20 - I-Q - CLI-Reference-ASRs-5000&5500 PDF
20 - I-Q - CLI-Reference-ASRs-5000&5500 PDF
20 - I-Q - CLI-Reference-ASRs-5000&5500 PDF
20
First Published: March 31, 2016
Last Modified: August 08, 2016
Americas Headquarters
Cisco Systems, Inc.
170 West Tasman Drive
San Jose, CA 95134-1706
USA
http://www.cisco.com
Tel: 408 526-4000
800 553-NETS (6387)
Fax: 408 527-0883
THE SPECIFICATIONS AND INFORMATION REGARDING THE PRODUCTS IN THIS MANUAL ARE SUBJECT TO CHANGE WITHOUT NOTICE. ALL STATEMENTS,
INFORMATION, AND RECOMMENDATIONS IN THIS MANUAL ARE BELIEVED TO BE ACCURATE BUT ARE PRESENTED WITHOUT WARRANTY OF ANY KIND,
EXPRESS OR IMPLIED. USERS MUST TAKE FULL RESPONSIBILITY FOR THEIR APPLICATION OF ANY PRODUCTS.
THE SOFTWARE LICENSE AND LIMITED WARRANTY FOR THE ACCOMPANYING PRODUCT ARE SET FORTH IN THE INFORMATION PACKET THAT SHIPPED WITH
THE PRODUCT AND ARE INCORPORATED HEREIN BY THIS REFERENCE. IF YOU ARE UNABLE TO LOCATE THE SOFTWARE LICENSE OR LIMITED WARRANTY,
CONTACT YOUR CISCO REPRESENTATIVE FOR A COPY.
The Cisco implementation of TCP header compression is an adaptation of a program developed by the University of California, Berkeley (UCB) as part of UCB's public domain version
of the UNIX operating system. All rights reserved. Copyright © 1981, Regents of the University of California.
NOTWITHSTANDING ANY OTHER WARRANTY HEREIN, ALL DOCUMENT FILES AND SOFTWARE OF THESE SUPPLIERS ARE PROVIDED “AS IS" WITH ALL FAULTS.
CISCO AND THE ABOVE-NAMED SUPPLIERS DISCLAIM ALL WARRANTIES, EXPRESSED OR IMPLIED, INCLUDING, WITHOUT LIMITATION, THOSE OF
MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT OR ARISING FROM A COURSE OF DEALING, USAGE, OR TRADE PRACTICE.
IN NO EVENT SHALL CISCO OR ITS SUPPLIERS BE LIABLE FOR ANY INDIRECT, SPECIAL, CONSEQUENTIAL, OR INCIDENTAL DAMAGES, INCLUDING, WITHOUT
LIMITATION, LOST PROFITS OR LOSS OR DAMAGE TO DATA ARISING OUT OF THE USE OR INABILITY TO USE THIS MANUAL, EVEN IF CISCO OR ITS SUPPLIERS
HAVE BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES.
Any Internet Protocol (IP) addresses and phone numbers used in this document are not intended to be actual addresses and phone numbers. Any examples, command display output, network
topology diagrams, and other figures included in the document are shown for illustrative purposes only. Any use of actual IP addresses or phone numbers in illustrative content is unintentional
and coincidental.
Cisco and the Cisco logo are trademarks or registered trademarks of Cisco and/or its affiliates in the U.S. and other countries. To view a list of Cisco trademarks, go to this URL: http://
www.cisco.com/go/trademarks. Third-party trademarks mentioned are the property of their respective owners. The use of the word partner does not imply a partnership
relationship between Cisco and any other company. (1110R)
ignore-pdp-data-inactivity 33
pdp-activate 35
source 296
tos 298
ttl 299
iu-recovery 342
iu-release-complete-timeout 343
loss-of-radio-coverage ranap-cause 344
mbms 346
network-sharing cs-ps-coordination 347
network-sharing failure-code 349
network-sharing non-shared 351
network-sharing stop-redirect-reject-cause 352
plmn 353
rab-assignment-response-timeout 355
radio-network-controller 356
rai-skip-validation 358
relocation-alloc-timeout 359
relocation-complete-timeout 360
reset 361
rnc 363
security-mode-complete-timeout 364
service-request-follow-on 365
srns-context-response-timeout 366
tigoc-timeout 367
tintc-timeout 368
max-tunnel-challenge-length 388
max-tunnels 389
peer-lns 390
proxy-lcp-authentication 392
retransmission-timeout-first 393
retransmission-timeout-max 394
single-port-mode 395
snoop framed-ip-address 397
trap 399
tunnel selection-key 400
tunnel-authentication 402
mtp3-p1-qlen 427
mtp3-p2-qlen 428
mtp3-p3-qlen 429
mtp3-test-pattern 430
priority 431
signaling-link-code 432
sscf-nni-n1 433
sscop-max-cc 434
sscop-max-pd 435
sscop-max-stat 436
timeout 437
standalone 473
timestamp-option-validation 474
timestamp-replay-protection 475
exit 519
CHAPTER 36 LTE Foreign PLMN GUTI Management Database Configuration Mode Commands 607
end 608
exit 609
plmn 610
CHAPTER 39 LTE MME HeNB-GW Management Database Configuration Mode Commands 625
end 626
exit 627
henbgw-global-enbid 628
CHAPTER 40 LTE Network Global MME ID Management Database Configuration Mode Commands 631
end 632
exit 633
plmn 634
max-retransmissions 743
mobility-header-checksum 744
mobility-option-type-value 745
policy 746
reg-lifetime 748
renew-percent-time 749
retransmission-policy 750
retransmission-timeout 751
signalling-packets 753
session 786
exit 825
hash-value 826
plmn-id 828
use-msc 830
isda-guard-timeout 895
isr-capability 897
legacy-tai-list-encoding 898
local-cause-code-mapping apn-mismatch 899
local-cause-code-mapping apn-not-subscribed 901
local-cause-code-mapping apn-not-supported-in-plmn-rat 902
local-cause-code-mapping auth-failure 904
local-cause-code-mapping congestion 906
local-cause-code-mapping ctxt-xfer-fail-mme 908
local-cause-code-mapping ctxt-xfer-fail-sgsn 910
local-cause-code-mapping gw-unreachable 912
local-cause-code-mapping hss-unavailable 914
local-cause-code-mapping newcall-policy-restrict 916
local-cause-code-mapping no-active-bearers 918
local-cause-code-mapping peer-node-unknown 920
local-cause-code-mapping pgw-selection-failure 922
local-cause-code-mapping restricted-zone-code 924
local-cause-code-mapping sgw-selection-failure 926
local-cause-code-mapping vlr-down 928
local-cause-code-mapping vlr-unreachable 930
location-reporting 932
mapping 933
max-bearers per-subscriber 935
max-paging-attempts 936
max-pdns per-subscriber 938
mme-id 939
mmemgr-recovery 941
msc 942
msc-mapping 944
nas gmm-qos-ie-mapping 945
nas-max-retransmission 946
network-sharing 947
nri 949
peer-mme 951
peer-sgsn rai 953
aaa 1008
associate 1009
attribute 1010
bind 1012
disconnect 1015
dns-P-GW 1017
end 1019
exit 1020
fqdn 1021
pgw-selection 1022
radius 1024
setup-timeout 1026
CHAPTER 61 Network Service Entity - Peer NSEI Configuration Mode Commands 1043
bssgp-timer 1044
end 1045
exit 1046
ns-reset-mode 1047
ns-vc 1049
end 1052
exit 1053
exit 1105
iiop-port 1106
iiop-transport 1107
iop-address 1108
max-attempt 1109
session-timeout 1110
siop-port 1111
ssl-auth-policy 1112
ssl-certificate 1113
ssl-private-key 1115
redistribute 1155
refresh timer 1157
router-id 1158
timers spf 1159
gt-address 1204
gt-format 1205
ni-indicator 1206
point-code 1207
routing-indicator 1208
ssf 1209
ssn 1210
subscription-attribute 1320
spr-profile-not-found 1322
threshold-condition usage-monitor 1323
user-access-network 1326
user-equipment-info esn 1328
user-equipment-info eui64 1330
user-equipment-info imeisv 1332
user-equipment-info mac 1334
user-equipment-info meid 1336
user-equipment-info modified-eui64 1338
user-location-info 1340
server 1477
setup-timeout 1529
nai-construction 1575
new-call conflict 1577
pcf-monitor 1578
pcf-session-id-change restart-ppp 1580
pdsn type0-tft attempt-inner-match 1582
peer-pcf 1583
pma-capability-indicator 1584
policy 1585
ppp 1589
qos-profile-id-mapping 1591
qos update 1594
radius accounting dropped-pkts 1596
registration-accept 1597
registration-ack-deny terminate-session-on-error 1598
registration-deny 1599
registration-discard 1602
registration-update 1604
retransmission-timeout 1606
service-option 1608
setup-timeout 1610
simple-ip allow 1611
spi 1612
tft-validation wait-timeout 1615
threshold a11-ppp-send-discard 1616
threshold a11-rac-msg-discard 1618
threshold a11-rrp-failure 1620
threshold a11-rrq-msg-discard 1622
threshold init-rrq-rcvd-rate 1624
map-initial-setup-auth-fail-to-gtp-cause-user-auth-fail 1679
message-timestamp-drift 1680
newcall 1682
pcscf-restoration 1684
plmn id 1685
session-delete-delay 1687
setup-timeout 1688
The Internet Group Management Protocol (IGMP) Profile Configuration Mode is used to create and manage
the IGMP parameters for an Ethernet interface.
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
Important Available commands or keywords/variables vary based on platform type, product version, and installed
license(s).
default ip igmp
Configures default IGMP parameters.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
Usage Guidelines Specify the IGMP parameters for the default profile. Refer to the remaining command description in this
chapter for additional information.
Examples To apply enable echo mode on this interface, use the following command:
bfd echo
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip igmp query
Configures the maximum response time for IGMP queries.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
max-response-time seconds
Specify the maximum number of seconds that the system will wait for an IGMP response as an integer from
1 through 25.
Usage Guidelines Specify the maximum response time for IGMP queries.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
no
Sets the router alert flag to OFF in IP IGMP packets.
ip igmp robustness
Sets the Robustness value in IP IGMP packets. The Robustness variable allows tuning for the expected packet
loss on a subnet. If a subnet is expected to be lossy, the Robustness variable may be increased. IGMP is robust
to packet losses. The Robustness variable should not be set to 1 (one).
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
value
Sets the robustness value as an integer from 1 through 10. Default: 2
ip igmp unsolicited-report-interval
Sets the Unsolicited Report Interval which is the time between repetitions of a host's initial report of membership
in a group.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
seconds
Specifies the number of seconds between repetitions of a host's initial report of membership in a group as an
integer from 1 through 25. Default: 10
Usage Guidelines Set the Unsolicited Report Interval which is the time between repetitions of a host's initial report of membership
in a group.
ip igmp version
Sets the IGMP version to be supported by this interface.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IGMP Profile Configuration
configure > context context_name > ip igmp profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-igmp-profile-<profile_name>)#
v1 | v2 | v3
Specifies the IGMP Version number:
• v1: Version 1, RFC 1112
• v2: Version 2, RFC 2236
• v3: Version 3, RFC 4604
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• default, page 12
• encryption, page 13
• end, page 15
• exit, page 16
• group, page 17
• hmac, page 19
• lifetime , page 21
• prf, page 22
default
Sets the default properties for the selected parameter.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
Usage Guidelines Configure default parameters for the IKEv2 IKE SA transform-set.
Examples Use the following configuration to set the default encryption algorithm:
default encryption
encryption
Configures the appropriate encryption algorithm and encryption key length for the IKEv2 IKE security
association. AES-CBC-128 is the default.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
3des-cbc
Data Encryption Standard Cipher Block Chaining encryption applied to the message three times using three
different cypher keys (triple DES).
aes-cbc-128
Advanced Encryption Standard Cipher Block Chaining with a key length of 128 bits.
aes-cbc-256
Advanced Encryption Standard Cipher Block Chaining with a key length of 256 bits.
des-cbc
Data Encryption Standard Cipher Block Chaining. Encryption using a 56-bit key size. Relatively insecure.
null
Configures no IKEv2 IKE Security Association Encryption Algorithm. All IKEv2 IPsec Child Security
Association protected traffic will be sent in the clear.
Note USE OF THIS ALGORITHM FOR IKE_SA ENCRYPTION IS A VIOLATION OF RFC 4306. THIS
ALGORITHM SHOULD ONLY BE USED FOR TESTING PURPOSES.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
group
Configures the appropriate key exchange cryptographic strength by applying a Diffie-Hellman group. Default
is Group 2.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
1
Configures crypto strength at the Group 1 level. Lowest security.
2
Configures crypto strength at the Group 2 (default) level. Medium security.
This is the default setting for this command.
5
Configures crypto strength at the Group 5 level. Higher security.
14
Configures crypto strength at the Group 14 level. Highest security
Usage Guidelines Diffie-Hellman groups are used to determine the length of the base prime numbers used during the key
exchange process in IKEv2. The cryptographic strength of any key derived depends, in part, on the strength
of the Diffie-Hellman group upon which the prime numbers are based.
Group 1 provides 768 bits of keying strength, Group 2 provides 1024 bits, Group 5 provides 1536 bits and
Group 14 provides 2048 bits of encryption strength.
Configuring a DH group also enables Perfect Forward Secrecy, which is disabled by default.
Examples This command configures crypto strength at the Group 14 level. Highest security group 14:
default group
hmac
Configures the IKEv2 IKE SA integrity algorithm. Default is SHA1-96.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
aes-xcbc-96
HMAC-AES-XCBC uses a 128-bit secret key and produces a 128-bit authenticator value.
md5-96
HMAC-MD5 uses a 128-bit secret key and produces a 128-bit authenticator value.
sha1-96
HMAC-SHA-1 uses a 160-bit secret key and produces a 160-bit authenticator value. This is the default setting
for this command.
sha2-256-128
HMAC-SHA-256 uses a 256-bit secret key and produces a 128-bit authenticator value.
sha2-384-192
HMAC-SHA-384 uses a 384-bit secret key and produces a 192-bit authenticator value.
sha2-512-256
HMAC-SHA-512 uses a 512-bit secret key and produces a 256-bit authenticator value.
lifetime
Configures the lifetime of a security association (SA) in seconds.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
lifetime sec
Sets the value of the timeout parameter in seconds as an integer from 60 through 86400. Default: 86400
Usage Guidelines The secret keys that are used for various aspects of a configuration should only be used for a limited amount
of time before timing out. This exposes a limited amount of data to the possibility of hacking. If the SA expires,
the options are then to either close the SA and open an new one, or renew the existing SA.
Examples The following command sets the lifetime timeout to 120 seconds:
lifetime 120
prf
Selects one of the HMAC integrity algorithms to act as the IKE Pseudo-Random Function. A PRF produces
a string of bits that an attacker cannot distinguish from random bit string without knowledge of the secret key.
The default is SHA1.
Product ePDG
PDIF
Command Modes Exec > Global Configuration > Context Configuration > IKEv2 Security Association Configuration
configure > context context_name > ikev2-ikesa transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(cfg-ctx-ikev2ikesa-tran-set)#
aes-xcbc-128
Configure IKEv2 IKE Security Association Pseudo Algorithm to be AES-XCBC-128.
md5
MD5 uses a 128-bit secret key and produces a 128-bit authenticator value.
sha1
SHA-1 uses a 160-bit secret key and produces a 160-bit authenticator value.
SHA-1 is considered cryptographically stronger than MD5, but it takes more CPU cycles to compute.
This is the default setting for this command.
sha2-256
PRF-HMAC-SHA-256 uses a 256-bit secret key.
sha2-384
PRF-HMAC-SHA-384 uses a 384-bit secret key.
sha2-512
PRF-HMAC-SHA-512 uses a 512-bit secret key.
Usage Guidelines This command generates keying material for all the cryptographic algorithms used in both the IKE_SA and
the CHILD_SAs.
The IMEI profile configuration mode defines a set of parameters controlling the SGSN behavior when a
Request is received from a device in the specified IMEI (International Mobile Equipment Identity) range.
An IMEI profile is a key element in the Operator Policy feature and an IMEI profile is not used or valid
unless it is associated with an IMEI range and this association is specified in an operator policy (see the
Operator Policy Configuration Mode Commands chapter elsewhere in the Command Line Interface Reference).
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• associate, page 26
• blacklist, page 27
• description, page 28
• direct-tunnel, page 29
• end, page 30
• exit, page 31
• ggsn-address, page 32
• ignore-pdp-data-inactivity, page 33
• pdp-activate, page 35
associate
Associate an APN remap table with this IMEI profile.
Note that an APN remap table can be associated with an IMEI profile before the table has actually been
created/configured.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
no
Disables the configured remap table association.
table_name
Define the name of an APN remap table that is to be associated with this IMEI profile for call routing based
in IMEI.
Usage Guidelines Use this command to associate an APN remap table with this IMEI profile. With such an association, it is
possible to override an APN call-routing based on an IMEI.
For example, with the APN exceptions defined in an APN remap table (refer to the APN Remap Table
Configuration Mode chapter), a blank APN or an incorrect APN could be overriden. So during PDP Activation
for in incoming call, the call could be rerouted based on an IMEI in the range defined for the IMEI profile.
Examples Associate the APN remap table 'remapHO' (remaps all calls with blank APNS to the head-office) to this IMEI
profile:
associate apn-remap-table remapHO
blacklist
Blacklist all mobile devices that fit the IMEI definitions associated with this IMEI profile.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
remove
Including this keyword with the command, removes the blacklist command from the IMEI profile configuration.
Usage Guidelines Blacklists subscribers whose devices bear IMEI that match the defined IMEI range for this profile.
Examples Use this command to black list all subscribers with IMEI that fall within the range set for this IMEI profile:
blacklist
description
Define a descriptive string relevant to the specific APN profile.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
remove
Removes the configured description from this APN profile.
description
Enter an alphanumeric string of 1 to 100 alphanumeric characters. The string may include spaces, punctuation,
and case-sensitive letters if the string is enclosed in double quotes ( " ).
Usage Guidelines Define information that identifies this particularly APN profile.
Examples Indicate that this IMEI profile IMEIprof1 is to be used for customers in the United Kingdom and that the
profile:
description "IMEIprof1 defines routing actions based on IMEI for customers in the UK."
direct-tunnel
Instruct the SGSN to enable/disable a direct tunnel between the RNC and the GGSN based on the IuPS service
configuration.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
no direct-tunnel
This command instructs the SGSN to disable the direct tunnel function between the GGSN and the RNC.
Usage Guidelines Direct tunnel is enabled by default on the GGSN and often on the RNC. This leaves it to the SGSN's
configuration to actually enable or disable a direct tunnel.
With the SGSN, the options for configuring a direct tunnel are complex -- enable/disable on the basis of APNs,
or RNCs, or GGSNs, or on the basis of the IMEI range. Refer to the SGSN Administration Guide for
configuration details.
Examples Assuming the IuPS service configuration has enabled DT for associated RNCs, then use this command to
enable DT from the RNC to the GGSN associated with this IMEI profile:
direct tunnel check-iups-service
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the context configuration mode.
Product SGSN
ggsn-address
Identify the target GGSN for traffic being managed by this IMEI profile.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
IPv4/IPv6_address
Enter the IP address of the target GGSN. Enter the address in either standard IPv4 dotted decimal format or
in standard IPv6 colon notation format.
Usage Guidelines Use this command to define the IP address of the target GGSN to be associated with this IMEI profile.
Examples The following command identifes the address of the GGSN associated with this IMEI profile as 123.123.11.1
ggsn-address 123.123.11.1
ignore-pdp-data-inactivity
On executing this command the SGSN ignores PDP Data Inactivity configuration under the APN profile for
one or more matching IMEIs.
Product SGSN
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
remove
This command is used to disable or remove the option to ignore PDP data inactivity from the IMEI profile.
Usage Guidelines The SGSN supports options to configure PDP Data Inactivity detection duration and actions to be performed
on timeout under the APN-Profile. The following configurable actions are supported under APN-Profile in
case of PDP Data Inactivity detection in the PDP context:
1 De-activate all PDPs of the subscriber
2 De-activate all PDPs of the bundle (all linked PDPs)
3 Detach the subscriber. This action is triggered when:
• Data in-activity is detected for all PDPs
• Data in-activity is detected for any of the PDPs
On the Detection of the PDP Data Inactivity, depending on the configuration option the SGSN either de-activates
the PDP or detaches the subscriber.
The ignore-pdp-data-inactivity CLI is added to provide an option under the IMEI-Profile to ignore PDP
Data Inactivity configuration for one or more IMEIs. On configuring this CLI, the SGSN ignores the application
of in-activity configuration (configured in the APN-Profile) for a specified set of IMEI's.
Note The IMEI range or set of IMEI's are mapped to specific IMEI-Profile using the CLI configuration option
under Operator-policy.
Examples Use this command to ignore PDP Data Inactivity configuration under the APN profile for one or more matching
IMEIs.
ignore-pdp-data-inactivity
pdp-activate
This command enables the operator to configure the SGSN to reject Secondary PDPActivation Requests from
the UE based on IMEI range.
Product SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > IMEI Profile Configuration
configure > imei-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-imei-profile-profile_name)#
Syntax Description [ remove ] pdp-activate { drop | restrict } secondary-activation access-type { gprs | umts }
remove
Removes the configured value and returns the pdp-activate configuration to the default 'not configured' state.
drop
This keyword causes the GSN to ignore the Request.
restrict
This keyword instructs the SGSN to reject Secondary PDP Activation Requests.
Usage Guidelines The SGSN administrator can use this command to configure the IMEI profile in the operator policy to either
ignore or reject Secondary PDP Activation Requests from UEs based on an IMEI range and UE access-type.
Restricting secondary PDP activation based on the IMEI (in the IMEI profile) takes precedence over secondary
PDP activation that might be configured in the call control profile.
Command Modes Exec > Global Configuration > LTE Policy Configuration > IMEI-TAC-Group
configure > lte-policy > imei-tac-groupgroup_name
Entering the above command sequence results in the following prompt:
[local]host_name(imei-tac-group)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• end, page 38
• exit, page 39
• tac, page 40
• tac-range, page 41
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
tac
Confitures individual TAC (type allocation code) values to be included in a IMEI-TAC group which will be
used as criteria for operator policy selection.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > IMEI-TAC-Group
configure > lte-policy > imei-tac-groupgroup_name
Entering the above command sequence results in the following prompt:
[local]host_name(imei-tac-group)#
no
Removes the identified TAC from the IMEI-TAC group configuration.
tac_value
Specifies the 8-digit number that identifies a specific "type allocation code". When entering more than one
TAC, simply use a space between each TAC. Additional TAC values can be added at any time after the
IMEI-TAC group is configured.
Usage Guidelines Use this command to enter one or more individual TAC (type allocation code) values to the IMEI-TAC group.
Up to 500 unique IMEI-TAC values can be included in an IMEI-TAC group.
The TAC, the first eight digits of the 15-digit IMEI or 16-digit IMEI-SV, identifies the equipment manufacturer,
the wireless device type and the model number (if there is one); for example, TAC of 35201906 identifies an
Apple iPhone 5S.
tac-range
Defines a range of IMEI-TAC values to be included in a IMEI-TAC group which will be used as criteria for
operator policy selection.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > IMEI-TAC-Group
configure > lte-policy > imei-tac-groupgroup_name
Entering the above command sequence results in the following prompt:
[local]host_name(imei-tac-group)#
no
Removes the identified TAC range from the IMEI-TAC group configuration.
start_tac_value to end_tac_value
tac_value - Specifies the 8-digit number that identifies a specific "type allocation code". The start TAC is
the first TAC in the range. The end TAC is the last TAC in the range.
Usage Guidelines Use this command to enter up to 20 IMEI-TAC value ranges. Ranges can be overlapping.
The TAC, the first eight digits of the 15-digit IMEI or 16-digit IMEI-SV, identifies the equipment manufacturer,
the wireless device type and the model number (if there is one); for example, TAC of 35201906 identifies an
Apple mobile phone. Defining ranges would enable carriers to select operator policies for call handling based
on multiple device types.
Examples The following command defines a TAC range to be added to the IMEI-TAC group:
tac-range from 23456789 to 23456889
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• end, page 44
• exit, page 45
• p-cscf discovery, page 46
• p-cscf table, page 48
• policy-control, page 51
• qos-update-timeout, page 52
• reauth-trigger, page 53
• signaling-flag, page 55
• signaling-flow, page 57
• traffic-policy, page 59
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
p-cscf discovery
This command defines the method of Proxy-Call Session Control Function (P-CSCF) discovery to be used.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
default
Sets the P-CSCF discovery to default parameter.
no
Removes/deletes configured parameters for P-CSCF discovery.
table { 1 | 2 }
Specifies that which P-CSCF table is to be used to obtain the primary and secondary P-CSCF addresses. Total
2 tables can be configured for P-CSCF discovery.
Default: round-robin
diameter-configured
This option enables the table number and algorithm specified by the diameter host-select table configuration
in Policy Control Configuration mode.
Usage Guidelines Use this command to configure the table and row selection methods to select IP address/host address for
P-CSCF discovery.
Examples The following command specifies table 1 with round-robin algorithm to select the rows with IP address for
P-CSCF discovery.
p-cscf discovery table 1 algorithm round-robin
p-cscf table
This command adds/appends rows with primary and/or secondary IPv4/IPv6 addresses to a P-CSCF discovery
table with precedence for P-CSCF discovery.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
no
Removes/deletes configured row with precedence in specified table for P-CSCF discovery address.
{1|2}
Specifies which P-CSCF table is to be used to add/append the primary and secondary P-CSCF addresses. Two
tables can be configured for P-CSCF discovery address.
row-precedence precedence_value
This keyword adds/appends the row with the specified row-precedence to the P-CSCF address table.
In 8.1 and later releases, precedence_value must be an integer from 1 through 128, and a maximum of 128
rows can be added to a table.
In release 8.0, precedence_value must be an integer from 1 through 100, and a maximum of 16 rows can be
added to a table.
secondary
Specifies the secondary IPv4/IPv6 address to be entered in P-CSCF table rows.
address ip_address
Specifies the primary and/or secondary IPv4 address for P-CSCF discovery table. This keyword, if used with
secondary keyword, specifies the secondary IPv4 address.
Important This keyword is available only in releases prior to 18. In 18 and later releases, this keyword is concealed
and is replaced with ipv4-address to support the PDN type v4v6 request for VoLTE setup.
ipv4-address ipv4_address
Specifies the primary and/or secondary IPv4 address for P-CSCF discovery table. This keyword, if used with
secondary keyword, specifies the secondary IPv4 address.
ipv4_address must be entered in IPv4 dotted-decimal notation.
Important This keyword is available in 18 and later releases to support the PDN type v4v6 request for VoLTE setup.
In releases prior to 18, the P-CSCF configuration accepts only one primary and one secondary P-CSCF IP
addresses – both IPv4 and IPv6 addresses per row in the P-CSCF address table. Two IP addresses are not
sufficient enough to address the requirement with PDN type v4v6 request for VoLTE setup. Hence, in release
18, the P-CSCF configuration has been enhanced to allow users to configure a maximum of two IPv4 addresses
(primary/secondary) and two IPv6 addresses (primary/secondary) per P-CSCF table row.
ipv6-address ipv6_address
Specifies the primary and/or secondary IPv6 address for P-CSCF discovery table. This keyword, if used with
secondary keyword, specifies the secondary IPv6 address.
ipv6_address must be entered in IPv6 colon-separated-hexadecimal notation.
In releases prior to 18, the P-CSCF configuration accepts only one primary and one secondary P-CSCF IP
addresses – both IPv4 and IPv6 addresses per row in the P-CSCF address table. Two IP addresses are not
sufficient enough to address the requirement with PDN type v4v6 request for VoLTE setup. Hence, in release
18, the P-CSCF configuration has been enhanced to allow users to configure a maximum of two IPv4 addresses
(primary/secondary) and two IPv6 addresses (primary/secondary) per P-CSCF table row.
weight value
This keyword designates weight to a row-precedence relative to other row-precedences configured under this
table, Default value is 1. value must be an integer from 1 through 10.
Within the IMS Authorization configuration, the P-CSCF address is selected based on round robin fashion.
This feature allows the customer to perform P-CSCF selection based on weight factor.
With this CLI option, the user can configure and add weight (in the scale of 1 to 10) to each row, and the rows
are selected based on weighted round-robin. That is, the row with higher weight parameter is selected more
number of times than the row with less number of weights.
Usage Guidelines Use this command to add rows with primary and/or secondary IP addresses for P-CSCF discovery. The row
is added with the specified row-precedence.
In releases prior to 17.0, IMSA will select the servers if requested server address type and selected row
server-address type are the same. Otherwise, it will return NULL. In 17.0 and later releases, P-CSCF server
selection algorithm is modified such that the P-CSCF server selection happens based on UE-requested
server-type.
The operator can add/remove rows to the table that is not currently selected by the diameter host-select table
command in Policy Control Configuration Mode.
In releases prior to 18, the look-up and forwarding of P-CSCF server information from P-CSCF table to the
session manager were performed by IMS Authorization (IMSA) server only during the setup. In 18 and later
releases, whenever IMSA receives a Modify Bearer request with P-CSCF Address request indication, then
the list of P-CSCF IP addresses are sent to the session manager through Modify Bearer Response message.
This look-up and forwarding functionality works even when the call is with the Local Policy (LP) engine
during the time the Modify Bearer Request is triggered.
Examples The following command adds a row in table 2 with primary IP address 10.2.3.4, secondary IP address as
50.6.7.8, and row-precedence value as 20 for P-CSCF discovery.
p-cscf table 2 row-precedence 20 address 10.2.3.4 secondary 50.6.7.8
policy-control
This command enters the Policy Control Configuration mode for Diameter Policy Control Application (DPCA)
to configure Diameter authorization and policy control parameter for IMS authorization.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
no
Disables the pre-configured policy control parameters for IMS authorization in this IMS authorization service.
Usage Guidelines Use this command to enter the Policy Control Configuration Mode to configure the policy control parameters
for Diameter authorization and charging policy in IMS Authorization Service.
Entering this command results in the following prompt:
[context_name]hostname(config-imsa-dpca)#
Policy Control configuration commands are described in the Policy Control Configuration Mode Commands
chapter.
qos-update-timeout
This command is obsolete in release 11.0 and later releases. This command sets the Quality of Service update
timeout for a subscriber in IMS authorization service.
Product GGSN
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
no
Disables the pre-configured QoS update timeout parameter in this IMS authorization service.
timeout_duration
Specifies the duration of timeout in seconds as an integer from 0 through 3600.
Default: 60
Usage Guidelines Use this command to set the maximum time to wait for a subscriber to initiate the update QoS procedure in
IMS authorization service.
Examples The following command sets the QoS update timeout to 90 seconds.
qos-update-timeout 90
reauth-trigger
This command specifies the trigger events to initiate re-authorization for a subscriber in IMS authorization
service.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
Default
Sets the pre-configured Re-authorization trigger to default value.
all
Sets the IMS authorization service to initiate re-authorization process for a subscriber on all events listed in
this command.
an-gw-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber whose access network
gateway changed.
bearer-loss
Sets the IMS authorization service to initiate re-authorization process for a subscriber on loss of bearer or
service.
bearer-recovery
Sets the IMS authorization service to initiate re-authorization process for a subscriber when a bearer or service
recovered after loss of bearer or service.
default-bearer-qos-change
Sets the IMS authorization service to initiate re-authorization process when QoS is changed and
DEFAULT_EPS_BEARER_QOS_CHANGE event triggered for the default EPS bearer context of a subscriber
in LTE network.
plmn-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber on change in Public
Land Mobile Network (PLMN) of subscriber.
policy-failure
Sets the IMS authorization service to initiate re-authorization process for a subscriber on failure of credit and
charging policy for subscriber.
qos-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber on change in Quality
of Service level/rating of subscriber.
rat-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber on change in Radio
Access Type (RAT) of subscriber node.
sgsn-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber on change in SGSN for
subscriber node.
tft-change
Sets the IMS authorization service to initiate re-authorization process for a subscriber on change in Traffic
Flow Template (TFT) of subscriber session.
tft-delete
Sets the IMS authorization service to initiate re-authorization process for a subscriber when Traffic Flow
Template (TFT) of subscriber session is deleted by a system administrative user.
Usage Guidelines Use this command to set the triggers to initiate QoS re-authorization process for a subscriber in IMS
authorization service.
Examples The following command sets the re-authorization trigger to bearer-loss, so that re-authorization of subscriber
session is initiated on loss of bearer.
reauth-trigger bearer-loss
signaling-flag
This command specifies whether a request for a PDP context dedicated to signaling (for IMS sessions) should
be granted or denied.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
default
Sets the signaling flag to default mode of deny.
deny
Denies the request for a signaling PDP context for IMS session and keeps signaling co-existed with other
traffic on PDP contexts. Default: Enabled
permit
Permits the request for a signaling PDP context for IMS session and a separate signaling context activated.
Default: Disabled
Usage Guidelines Use this command to allow or deny the activation of a dedicated PDP context for signaling. The user equipment
(UE) may indicate that the PDP context should be dedicated for IP multimedia (IM) signaling by setting the
IP Multimedia Core Network (IM-CN) signaling flag in the Protocol Configuration Options (PCO).
The deny option causes the system to inform the UE that the PDP context will not be dedicated for IM signaling
and signaling will co-exist with other traffic on PDP context.
The permit option is used to activate the signaling context for signal traffic and the other traffic uses other
PDP context for traffic with the following destinations:
• Towards the DHCP and DNS servers for the IMS domain
• Towards the P-CSCF(s)
The UE is not trusted to follow these restrictions, and the system monitors and restricts the traffic from the
dedicated PDP context. The signaling-flow class-map command is used to configure the restrictions.
Examples The following command denies the request for a signaling PDP context for IMS session.
default signaling-flag
signaling-flow
This command specifies the packet filters and policy servers for bandwidth control and singling context
enforcement that define the traffic that is allowed through the dedicated signaling context.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
Syntax Description signaling-flow permit server-address ipv4/ipv6_address [ server-port { port_num | range start_port to
end_port } ] [ description STRING ]
no signaling-flow permit server-address ipv4/ipv6_address [ server-port { port_num | range start_port to
end_port } ]
no
Disables the signaling flow option configured with this command.
server-address ipv4/ipv6_address
The server address refers to the destination IP address in uplink packets, and the source IP address in downlink
packets.
ipv4/ipv6_address is an IP address in IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal notation and
can be used with a subnet mask.
A maximum of 16 signaling server addresses can be configured per IMS Authorization service.
description STRING
Specifies the customized description for configured signaling server as an alphanumeric string of 1 through
63 characters.
Usage Guidelines Traffic that matches any instance of the signaling-flow command will be forwarded via the signaling PDP
context. In addition, the policy server gives policy gates to use for the signaling PDP context.
Examples The following command sets the packet filter server address to 10.2.3.4 with port number 1234 for packet
filtering.
signaling-flow server-address 10.2.3.4 server-port 1234
traffic-policy
This command specifies the action on packets which do not match any policy gates in the general purpose
PDP context.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Service Configuration
configure > context context_name > ims-auth-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-service)#
Syntax Description traffic-policy general-pdp-context no-matching-gates direction { downlink | uplink } { forward | discard
}
default traffic-policy general-pdp-context no-matching-gates direction { downlink | uplink }
default
Sets the default traffic policy for packets without any policy gate match in general purpose PDP context.
By default packets which do not have any matching policy gate are forwarded.
no-matching gates
Applies traffic policy for packets which do not match any policy gate.
forward
Forwards the packets which do not match any policy gates. Default: Enabled
discard
Discards the packets which do not match any policy gates. Default: Disabled
Usage Guidelines This command provides configuration on traffic policy applied on packets which are not matching any policy
gate in general PDP context. Packets can either be forwarded or discarded on the basis of operator's
configuration.
This command needs to be configured once for downlink and once for uplink separately.
Examples The following command discards uplink packets which do not match any policy gate in general purpose PDP
context.
traffic-policy general-pdp-context no-matching-gates direction uplink discard
The IMS Sh Interface Configuration Mode is used to configure various Diameter parameters in order for:
Command Modes Exec > Global Configuration > Context Configuration > IMS Sh Interface Configuration
configure > context context_name > ims-sh-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ims-sh-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• diameter, page 62
• end, page 64
• exit, page 65
• failure-handling, page 66
• request, page 68
diameter
This command configures Diameter parameters.
Product PDIF
SCM
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > IMS Sh Interface Configuration
configure > context context_name > ims-sh-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ims-sh-service)#
no
Removes previously configured endpoint.
default
Configures parameters to the default value.
dictionary
Specifies the dictionary to use.
• custom1: A custom dictionary
• standard: The standard dictionary
endpoint string
Selects an endpoint to use in the configuration.
string must be the endpoint name, and must be an alpha and/or numeric string of 1 through 63 characters in
length.
Usage Guidelines The Diameter endpoint contains information on the peer names and IP addresses and port, and the local IP
address to use for Diameter.
You can have more than one Diameter endpoint configured on the chassis and the ims-sh-service needs to
know which Diameter endpoint to use. This command is to select the appropriate Diameter endpoint, even if
only one has been configured.
end
Exits the current mode and returns to the Exec Mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
failure-handling
This command configures the action to take in the event of an HSS server request failure.
Product PDIF
SCM
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > IMS Sh Interface Configuration
configure > context context_name > ims-sh-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ims-sh-service)#
default
Resets configuration for the specified keyword to the default setting.
profile-update-request
Configures failure-handling as a result of a profile update request error.
user-data-request
Configures failure-handling as a result of a user data request.
result_code specifies either a result code value (diameter-result-code 3001) or a range of result code values
(diameter-result-code 3000 to 9999) to which the failure-handling applies.
action
Configures the action to take depending on the diameter-result-code:
• Continue the session
• Retry and then terminate
• Terminate the session
request-timeout action
Configures the action to take as a result of a request timeout error:
• Continue the session
• Retry and then terminate
• Terminate the session
Examples The following command configures profile-update-request failure-handling using a result-code configuration
with the terminate session option:
failure-handling profile-update-request diameter-result-code 3005 to 3600action terminate
request
Configures application request timeout.
Product PDIF
SCM
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > IMS Sh Interface Configuration
configure > context context_name > ims-sh-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ims-sh-service)#
no
Disables a configured timeout request.
default
Default: 300 seconds
Resets configuration to the default setting.
Usage Guidelines Specifies the session request timeout period in seconds after which the request is deemed to have failed.
Examples The following example configures the default timeout request of 300 seconds:
default request timeout
Important This is a license enabled external application support. For more information on this product, refer to the
IPMS Installation and Administration Guide.
The IPMS Client Configuration Mode is used to enable the Intelligent Packet Monitoring System (IPMS)
client service on an Access Gateway and to set basic service-wide options in a context.
Command Modes Exec > Global Configuration > Context Configuration > IPMS Configuration
configure > context context_name > ipms
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-ipms)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• end, page 70
• exit, page 71
• export keys, page 72
• heartbeat, page 73
• server, page 75
• source, page 77
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
export keys
Enables the encryption key export in specific key exchange events to IPMS server from IPMS-enabled AGW.
Product IPMS
Command Modes Exec > Global Configuration > Context Configuration > IPMS Configuration
configure > context context_name > ipms
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-ipms)#
Syntax Description
no
Removes the configured source IP address from this context for IPMS client communication with IPMS
server.
ikev2
Enables the security association (SA) key export for Internet Key Exchange (IKEv2) protocol to IPMS server.
Usage Guidelines Monitor subscribers which have complaints of service availability or to monitor a test user for system
verification.
Examples The following command assigns the IP address 10.2.3.4 to the IPMS client service in context to communicate
with IPMS server. This is the IP address allocated for IPMS client service on chassis.
source address 10.2.3.4
heartbeat
Configures the IPMS heartbeating between the IPMS-enabled AGW and the IPMS server.
Product IPMS
Command Modes Exec > Global Configuration > Context Configuration > IPMS Configuration
configure > context context_name > ipms
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-ipms)#
default
Configures the heartbeat period and permitted number of failures to the default values of 10 seconds and 1
failure respectively.
no
Disables/removes the configured heartbeat period and permitted number of failures.
period dur
Specifies the periodicity (in seconds) between heartbeat messages as an integer from 1 through 3600. Default:
10
permitted-failure no_of_failures
Specifies the number of errors/failures allowed before declaring an IPMS server as dead/unreachable as an
integer from 1 through 10. Default: 1
Usage Guidelines Use this command to configure the heartbeat message periodicity and permissible failure of heartbeat message
response before declaring an IPMS server as dead or unreachable. When an IPMS server is declared down
an SNMP trap is sent.
Examples Following command configures the heartbeat message periodicity to 5 second and number of failures allowed
as 3 to determine an IPMS server as dead.
heartbeat period 5 permitted-failure 3
server
Configures the IPMS server address and ports on which the IPMS client on an IPMS-enabled AGW
communicates. This is the IP address and port range of the IPMS server.
Product IPMS
Command Modes Exec > Global Configuration > Context Configuration > IPMS Configuration
configure > context context_name > ipms
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-ipms)#
Syntax Description server address ip_address [ seconary ] [ start-port start_port [ end-port end_port ]][ secondary
]
no server address ip_address
no
Removes the configured IPMS server IP address and port range from this context.
address ip_address
Specifies the IP address of the IPMS server to which the IPMS client service communicates in IPv4
dotted-decimal notation.
A maximum of 4 IPMS severs can be configured with this command in one context.
secondary
The secondary keyword is used to configure the specified server address as secondary IP address on the IPMS
client interface.
Usage Guidelines Use this command to configure/remove the IPMS servers. Up to 4 different IPMS servers can be configured
with this command. UDP port number can also be configured with this command. IPMS client will search
for this IP address to push the event and traffic logs.
Examples The following command configures IPMS server having IP address 10.2.3.4 in the IPMS client service export
the event and traffic logs for intelligent packet monitoring functionality. It also specifies the UDP port range
from 48000 to 48005 for communication.
server address 10.2.3.4 start-port 48000 end-port 48005
source
Configures the source address of the IPMS client in this context to communicate with the IPMS server. This
is the IP address for IPMS client on the chassis.
Product IPMS
Command Modes Exec > Global Configuration > Context Configuration > IPMS Configuration
configure > context context_name > ipms
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-ipms)#
address ip_address
Specifies the IP address of the IPMS client on the AGW in this context. This is the address which is bound
to the IPMS client service in this context.
ip_address is expressed in IPv4 dotted-decimal notation.
Usage Guidelines Monitor subscribers which have complaints of service availability or to monitor a test user for system
verification.
Examples The following command assigns the IP address 10.2.3.4 to the IPMS client service in context to communicate
with IPMS server. This is the IP address allocated for IPMS client service on chassis.
source address 10.2.3.4
The IPNE Endpoint Configuration Mode provides the commands to configure the parameters for an IPNE
Endpoint in an IPNE Service.
Command Modes Exec > Global Configuration > Context Configuration > IPNE Service Configuration > IPNE Endpoint
Configuration
configure > context context_name > ipne-service ipne_service_name > ipne-endpoint
Entering the above command sequence results in the following prompt:
[ context_name]host_name (config-ipne-endpoint)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• bind, page 80
• end, page 81
• exit, page 82
• peer, page 83
bind
This command binds the IPNE client socket to the IPNE endpoint.
Product MME
Command Modes Exec > Global Configuration > Context Configuration > IPNE Service Configuration > IPNE Endpoint
Configuration
configure > context context_name > ipne-service ipne_service_name > ipne-endpoint
Entering the above command sequence results in the following prompt:
[ context_name]host_name (config-ipne-endpoint)#
no
When included as a command prefix, the system removes the bind address from the IPNE endpoint
configuration.
ipv4-address | ipv6-address
Identifies whether the bind address uses IPv4 or IPv6 format.
ip_address
Enter either an IPv4 dotted-decimal address or an IPv6 colon-separated hexadecimal notation
Usage Guidelines The bind command defines the IP address of the IPNE client socket as the local address.
Examples Use a command similar to the following to bind the IPNE client socket to the IPNE endpoint.
bind ipv4-address 123.123.123.1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
peer
Identifies the MINE server as a peer for the IPNE endpoint.
Product MME
Command Modes Exec > Global Configuration > Context Configuration > IPNE Service Configuration > IPNE Endpoint
Configuration
configure > context context_name > ipne-service ipne_service_name > ipne-endpoint
Entering the above command sequence results in the following prompt:
[ context_name]host_name (config-ipne-endpoint)#
no
Removes the peer address from the IPNE endpoint configuration.
ipv4-address | ipv6-address
Informs the system of the format of the peer address.
ip_address
Enter either an IPv4 dotted-decimal address or an IPv6 colon-separated hexadecimal notation.
Usage Guidelines Use the peer command to configure a MINE server IP address as the peer for the IPNE endpoint.
The IPNE Service Configuration Mode is used to configure and manage the IPNE Service.
Command Modes Exec > Global Configuration > Context Configuration > IPNE Service Configuration
configure > context context_name > ipne-service ipne_service_name
Entering the above command sequence results in the following prompt:
[ context_name]host_name (config-ipne-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• end, page 86
• exit, page 87
• ipne-endpoint, page 88
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ipne-endpoint
Creates and configures an IPNE endpoint and enters the IPNE endpoint configuration mode. An IPNE endpoint
is a combination of a local IPP address, a peer address and, optionally, a port.
Product MME
Command Modes Exec > Global Configuration > Context Configuration > IPNE Service Configuration
configure > context context_name > ipne-service ipne_service_name
Entering the above command sequence results in the following prompt:
[ context_name]host_name (config-ipne-service)#
no
Causes the system to delete the IPNE endpoint configuration from the IPNE service configuration.
Usage Guidelines Use this command to create an IPNE endpoint and to enter the IPNE endpoint sub-configuration mode.
Examples Use the following command to access commands to configure the IPNE endpoint:
ipne-endpoint
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
• encryption, page 90
• end, page 93
• esn, page 94
• exit, page 96
• group, page 97
• hmac, page 99
• mode, page 101
encryption
Configures the appropriate IPSec ESP encryption algorithm and encryption key length. AES-CBC-128 is the
default.
Product ePDG
PDIF
SCM
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
3des-cbc
Data Encryption Standard Cipher Block Chaining encryption applied to the message three times using three
different cypher keys (triple DES).
aes-128-gcm-128
IKEv2 Child Security Association IPsec ESP Algorithm is AES-GCM-128 with 128-bit ICV (Integrity Check
Value). HMAC algorithm with this encryption algorithm should be None.
aes-128-gcm-64
IKEv2 Child SA (Security Association) IPsec ESP Algorithm is AES-GCM-128 with 64-bit ICV. HMAC
algorithm with this encryption algorithm should be None.
aes-128-gcm-96
IKEv2 Child SA IPsec ESP Algorithm to be AES-GCM-128 with 96-bit ICV. HMAC algorithm with this
encryption algorithm should be None.
aes-256-gcm-128
IKEv2 Child SA IPsec ESP Algorithm is AES-GCM-256 with 128-bit ICV. HMAC algorithm with this
encryption algorithm should be None.
aes-256-gcm-64
IKEv2 Child SA IPsec ESP Algorithm is AES-GCM-256 with 64-bit ICV. HMAC algorithm with this
encryption algorithm should be None.
aes-256-gcm-96
IKEv2 Child SA IPsec ESP Algorithm is AES-GCM-256 with 96-bit ICV. HMAC algorithm with this
encryption algorithm should be None.
aes-cbc-128
Advanced Encryption Standard Cipher Block Chaining with a key length of 128 bits. This is the default setting
for this command.
aes-cbc-256
Advanced Encryption Standard Cipher Block Chaining with a key length of 256 bits.
des-cbc
Data Encryption Standard Cipher Block Chaining. Encryption using a 56-bit key size. Relatively insecure.
null
The NULL encryption algorithm represents the optional use of applying encryption within ESP. ESP can then
be used to provide authentication and integrity without confidentiality.
default
Sets the default IPSec ESP algorithm to AES-CBC-128.
Usage Guidelines AES-GCM (Advanced Encryption Standard-Galois Counter Mode) is a block cipher mode of operation that
uses universal hashing over a binary Galois field to provide authenticated encryption (RFC 5288). It uses
mechanisms that are supported by a well-understood theoretical foundation, and its security follows from a
single reasonable assumption about the security of the block cipher. StarOS supports these AEAD
(Authenticated Encryption with Associated Data) algorithms for improved IPsec performance when using
OpenSSL to process ESP packets.
Important The AEAD algorithms are only supported on virtualized platforms. They are not supported on ASR 5x00
hardware.
In cipher block cryptography, the plaintext is broken into blocks usually of 64 or 128 bits in length. In cipher
block chaining (CBC) each encrypted block is chained into the next block of plaintext to be encrypted. A
randomly generated vector is applied to the first block of plaintext in lieu of an encrypted block. CBC provides
confidentiality, but not message integrity.
Because RFC 4307 calls for interoperability between IPSec and IKEv2, the IKEv2 confidentiality algorithms
must be the same as those configured for IPsec in order for there to be an acceptable match during the IKE
message exchange. In IKEv2, there is no NULL option.
Examples The following command configures the encryption to be the default aes-cbc-128:
default encryption
end
Exits the current configuration mode and returns to the Exec mode.
Product All
esn
Enables support for the use of 64-bit Extended Sequence Numbers (ESNs) in ikev2 Encapsulating Security
Payload (ESP) and Authentication Header (AH) packets. The ESN transform is included in an ikev2 proposal
used in the negotiation of IKE SAs as part of the IKE_SA_INIT exchange.
Product SecGW
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
Usage Guidelines Use this command to enable support for the use of 64-bit ESNs for ikev2. The ESN transform is included in
an ikev2 proposal used in the negotiation of IKE SAs as part of the IKE_SA_INIT exchange.
The ESN transform has the following meaning:
• A proposal containing one ESN transform with value 0 means "do not use extended sequence numbers".
• A proposal containing one ESN transform with value 1 means "use extended sequence numbers".
• A proposal containing two ESN transforms with values 0 and 1 means "I support both normal and
extended sequence numbers, you choose". This case is only allowed in requests; the response will contain
only one ESN transform.
In most cases, the exchange initiator will include either the first or third alternative in its SA payload. The
second alternative is rarely useful for the initiator: it means that using normal sequence numbers is not
acceptable (so if the responder does not support ESNs, the exchange will fail with NO_PROPOSAL_CHOSEN.
Enabling the esn command is the equivalent of sending ESN Transform = 0 and 1; support both 32-bit and
64-bit sequence numbers. If the esn command is not enabled, support only 32-bit sequence numbers (default
behavior).
Including the ESN transform is mandatory when creating ESP or AH SAs.
For additional information, see the IPSec Reference.
Important ESN is only supported on ASR 5500 and ASR 9000 Virtualized Services Modules (VSMs). It is not
supported on the ASR 5000 or VPC-SI.
Examples The following command enables support for 64-bit ESNs in ikev2 ESP and AH packets:
esn
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
group
Configures the appropriate key exchange cryptographic strength and activate Perfect Forward Secrecy by
applying a Diffie-Hellman group.
Product ePDG
PDIF
SCM
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
default group
Configures the default crypto strength to be none and disables Perfect Forward Secrecy.
1
Configures crypto strength at the Group 1 level. Lowest security.
2
Configures crypto strength at the Group 2 level. Medium security.
5
Configures crypto strength at the Group 5 level. Higher security.
14
Configures crypto strength at the Group 14 level. Highest security.
none
Applies no group and disables Perfect Forward Secrecy. This is the default.
default
Sets the default Diffie-Hellman group algorithm to none. This also deactivates PFS.
Usage Guidelines Diffie-Hellman groups are used to determine the length of the base prime numbers used during the key
exchange process. The cryptographic strength of any key derived depends, in part, on the strength of the
Diffie-Hellman group upon which the prime numbers are based.
Group 1 provides 768 bits of keying strength, Group 2 provides 1024 bits, Group 5 provides 1536 bits and
Group14 2048 bits. Selecting a group automatically activates Perfect Forward Secrecy. The default value is
none, which disables PFS
hmac
Configures the IPsec ESP integrity algorithm using a Hash-based Message Authentication Code (HMAC).
Product ePDG
PDIF
SCM
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
Syntax Description hmac { aes-xcbc-96 | md5-96 | none| null | sha1-96 | sha2-256-128 | sha2-384-192 | sha2-512-256 }
default hmac
default hmac
Sets the default IPSec hashing algorithm to SHA1-96.
aes-xcbc-96
AES-XCBC-96 uses a 128-bit secret key and produces a 128-bit authenticator value.
md5-96
MD5-96 uses a 128-bit secret key and produces a 128-bit authenticator value.
none
Sets the IPsec hashing algorithm to none. Used with OpenSSL AEAD algorithms.
null
Configures the HMAC value to be null. The NULL encryption algorithm represents the optional use of applying
encryption within ESP. ESP can then be used to provide authentication and integrity without confidentiality.
sha1-96
SHA-1 uses a 160-bit secret key and produces a 160-bit authenticator value. This is the default setting for this
command.
sha2-256-128
HMAC-SHA-256 uses a 256-bit secret key and produces a 128-bit authenticator value.
sha2-384-192
HMAC-SHA-384 uses a 384-bit secret key and produces a 192-bit authenticator value.
sha2-512-256
HMAC-SHA-512 uses a 512-bit secret key and produces a 256-bit authenticator value.
Usage Guidelines HMAC is an encryption technique used by IPsec to make sure that a message has not been altered.
A keyed-Hash-based Message Authentication Code (HMAC), is a type of message authentication code that
is calculated using a cryptographic hash function in combination with a secret key to verify both data integrity
and message authenticity. A hash takes a message of any size and transforms it into a message of a fixed size:
the authenticator value. This is truncated to 96 bits and transmitted. The authenticator value is reconstituted
by the receiver and the first 96 bits are compared for a 100 percent match.
Examples The following command configures the default HMAC value (SHA1-96):
default hmac
mode
Configures the security of IP datagrams based on header placement. Tunnel mode applies security to a
completely encapsulated IP datagram, while Transport does not. Default is Tunnel mode.
Product ePDG
PDIF
SCM
Command Modes Exec > Global Configuration > Context Configuration > IPSec Transform Set Configuration
configure > context context_name > ipsec transform-set set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
transport
In Transport mode, the IPSec header is applied only over the IP payload, not over the IP header in front of it.
The AH and/or ESP headers appear between the original IP header and the IP payload, as follows:
Original IP header, IPSec headers (AH and/or ESP), IP payload (including transport header).
Transport mode is used for host-to-host communications and is generally unsuited to PDIF traffic.
tunnel
In Tunnel mode, the original IP header is left intact, so a complete IP datagram is encapsulated, forming a
virtual tunnel between IPSec-capable devices. The IP datagram is passed to IPSec, where a new IP header is
created ahead of the AH and/or ESP IPSec headers, as follows:
New IP header, IPSec headers (AH and/or ESP), old IP header, IP payload.
Tunnel mode is used for network-to-network communications (secure tunnels between routers) or
host-to-network and host-to-host communications over the Internet.
This is the default setting for this command.
default mode
Sets the default IPSec Mode to Tunnel.
Usage Guidelines IPSec modes are closely related to the function of the two core protocols, the Authentication Header (AH)
and Encapsulating Security Payload (ESP). Both of these protocols provide protection by adding to a datagram
a header (and possibly other fields) containing security information. The choice of mode does not affect the
method by which each generates its header, but rather, changes what specific parts of the IP datagram are
protected and how the headers are arranged to accomplish this.
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
bind
This command allows you to configure the service to accept data on any interface configured in the context.
Optionally, you can also configure the system to limit the number of sessions processed by this service.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
no
If previously configured, deletes the binding configuration for the service.
max-subscribers max_sessions
Specifies the maximum number of subscriber sessions allowed for the service. If this option is not configured,
the system defaults to the license limit.
In StarOS 9.0 and later releases, max_sessions must be an integer from 0 through 4000000.
In StarOS 8.3 and earlier releases, max_sessions must be an integer from 0 through 3000000.
Usage Guidelines Use this command to initiate the service and begin accepting data on any interface configured in the context.
Examples The following command prepares the system to receive subscriber sessions on any interface in the context
and limits the sessions to 10000:
bind max-subscribers 10000
connection authorization
This command allows you to configure the RADIUS authorization password that must be matched by the
RADIUS accounting requests "snooped" by this service.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
no
Deletes the RADIUS connection authorization configuration from the current IPSG RADIUS snoop service.
Usage Guidelines RADIUS accounting requests being examined by the IPSG RADIUS snoop service are destined for a RADIUS
Accounting Server. Since the "snoop" service does not terminate user authentication, the user password is
unknown.
Use this command to configure the authorization password that the RADIUS accounting requests must match
in order for the service to examine and extract user information.
Examples The following command sets the RADIUS authorization password that must be matched by the RADIUS
accounting requests "snooped" by this service. The password is encrypted, and the password used in this
example is "secret".
connection authorization encrypted password secret
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
profile
This command allows you to configure the service to use APN or subscriber profile.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
default
Configures this command with its default setting.
APN
Specifies the service to support APN configuration required to enable Gx support.
subscriber
Specifies the service to support subscriber profile lookup.
Usage Guidelines Use this command to set the service to support APN profiles (supporting Gx through the enabling of
ims-auth-service) or for basic subscriber profile lookup.
radius
This command allows you to specify the RADIUS accounting servers where accounting requests are sent
after being "inspected" by this service.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
Syntax Description radius { accounting server ipv4_address [ port port_number | source-context context_name ] | dictionary
{ 3gpp2 | 3gpp2-835 | customXX | standard | starent | starent-835 | starent-vsa1 | starent-vsa1-835 } }
[ no ] radius accounting server ipv4_address [ port port_number | source-context context_name ]
no
Removes the RADIUS accounting server identifier from this service.
port port_number
Specifies the port number of the RADIUS Accounting Server where accounting requests are sent after being
"snooped" by this service.
port_number must be an integer from 1 through 65535.
Default: 1813
source-context context_name
Specifies the source context where RADIUS accounting requests are received.
context_name must be an alphanumeric string of 1 through 79 characters.
If this keyword is not configured, the system will default to the context in which the IPSG service is configured.
Dictionary Description
3gpp This dictionary consists not only of all of the attributes in the standard dictionary,
but also all of the attributes specified in 3GPP 32.015.
3gpp2 This dictionary consists not only of all of the attributes in the standard dictionary,
but also all of the attributes specified in IS-835-A.
3gpp2-835 This dictionary consists not only of all of the attributes in the standard dictionary,
but also all of the attributes specified in IS-835.
customXX These are customized dictionaries. For information on custom dictionaries, please
contact your Cisco account representative.
XX is the integer value of the custom dictionary.
standard This dictionary consists only of the attributes specified in RFC 2865, RFC 2866,
and RFC 2869.
starent This dictionary consists of all of the attributes in the starent-vsa1 dictionary and
incorporates additional Starent Networks VSAs by using a two-byte VSA Type
field. This dictionary is the master-set of all of the attributes in all of the
dictionaries supported by the system.
starent-835 This dictionary consists of all of the attributes in the starent-vsa1-835 dictionary
and incorporates additional Starent Networks VSAs by using a two-byte VSA
Type field. This dictionary is the master-set of all of the attributes in all of the
-835 dictionaries supported by the system.
starent-vsa1 This dictionary consists not only of the 3gpp2 dictionary, but also includes Starent
Networks vendor-specific attributes (VSAs) as well. The VSAs in this dictionary
support a one-byte wide VSA Type field in order to support certain RADIUS
applications. The one-byte limit allows support for only 256 VSAs (0–255). This
is the default dictionary.
starent-vsa1-835 This dictionary consists not only of the 3gpp2-835 dictionary, but also includes
Starent Networks vendor-specific attributes (VSAs) as well. The VSAs in this
dictionary support a one-byte wide VSA Type field in order to support certain
RADIUS applications. The one-byte limit allows support for only 256 VSAs
(0–255). This is the default dictionary.
Usage Guidelines Use this command to specify the RADIUS Accounting Servers where accounting requests are sent after being
snooped by this service.
Examples The following command specifies the IP address (10.2.3.4) of a RADIUS Accounting Server whose accounting
requests are to be "snooped", and the source context (aaa_ingress) where the requests are received on the
system:
radius accounting server 10.2.3.4 source-context aaa_ingress
sess-replacement
This command allows you to enable/disable session replacement.
Important This command is not supported in this release. The Session Replacement feature is under development
for future use.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
default
Configures this command with its default setting.
Default: Disabled.
no
If previously configured, deletes the configuration.
with-diff-acct-sess-id
Specifies to replace current session when a new session request comes with same IP address and same user
name/IMSI but different accounting session ID.
with-diff-ip
Specifies to replace current session when a new session request comes with same user name/IMSI but different
IP address.
with-diff-key
Specifies to replace current session when a new session request comes with same IP address but different user
name/IMSI.
Usage Guidelines Use this command to enable/disable session replacement. By default, session replacement is disabled.
Examples The following command enables session replacement specifying to replace the current session when a new
session request comes with same user name/IMSI but different IP address:
sess-replacement with-diff-ip
setup-timeout
This command allows you to configure the timeout value for IPSG session setup attempts.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Snoop Configuration
configure > context context_name > ipsg-service service_name mode radius-snoop
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-snoop)#
setup_timeout
Specifies the period of time (in seconds) the IPSG session setup is allowed to continue before the setup attempt
is terminated.
setup_timeout must be an integer from 1 through 1000000.
Default: 60
Usage Guidelines Use this command to prevent IPSG session setup attempts from continuing without termination.
Examples The following command configures the session setup timeout setting to 20 seconds:
setup-timeout 20
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
accounting-context
This command allows you to specify the GTPP accounting context.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
If previously configured, removes the accounting context configuration.
context_name
Specifies name of the GTPP accounting context.
context_name must be an alphanumeric string of 1 through 79 characters in length.
Usage Guidelines Use this command to specify the GTPP accounting context.
Examples The following command specifies to use the GTPP accounting context context12 for the eWAG service:
accounting-context context12
associate sgtp-service
This command allows you to associate an SGTP service with the current eWAG service.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
If previously configured, removes the service association from the configuration.
sgtp-service sgtp_service_name
Specifies name of the SGTP service to associate with this service.
sgtp_service_name must be the name of an SGTP service, and must be an alphanumeric string of 1 through
63 characters in length.
context sgtp_context_name
Specifies name of the context in which the SGTP service is configured.
sgtp_context_name must be the name of the context, and must be an alphanumeric string of 1 through 63
characters in length.
If a context is not specified, the current context is used.
Usage Guidelines Use this command to associate an SGTP service with the IPSG service. This enables the GTP functionality
for eWAG supporting GTP-C (GTP Control Plane) messaging and GTP-U (GTP User Data Plane) messaging
between eWAG and GGSN over the Gn' interface.
Important Any change to this configuration will result in restart of the eWAG service.
Examples The following command associates an SGTP service named service1, configured in the context named context2,
with the IPSG service:
associate sgtp-service service1 context context2
bind
This command allows you to bind the current IPSG/eWAG service to a logical AAA interface, and specify
the number of subscriber sessions allowed.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
Syntax Description bind accounting-proxy address ipv4_address [ max-subscribers max_sessions | port port_number |
source-context source_context ]
bind address ipv4_address [ disconnect-message [ src-port source_port_number ] | max-subscribers
max_sessions | port port_number | source-context source_context ]+
bind authentication-proxy address ipv4_address [ acct-port port_number | auth-port port_number |
max-subscribers max_sessions | source-context source_context ]
no bind
no
If previously configured, removes the binding for the service.
• acct-port port_number: Specifies the port number of the interface where accounting proxy requests are
received by this service.
port_number must be an integer from 0 through 65535.
Default: 1813
• auth-port port_number: Specifies the port number of the interface where authentication proxy requests
are received by this service.
port_number must be an integer from 0 through 65535.
Default: 1812
• max-subscribers max_sessions: Specifies the maximum number of subscriber sessions allowed for the
service. If this option is not configured, the system defaults to the license limit.
In StarOS 9.0 and later releases, max_sessions must be an integer from 0 through 4000000.
In StarOS 8.3 and earlier releases, max_sessions must be an integer from 0 through 3000000.
• source-context source_context: Specifies the source context where RADIUS accounting requests are
received.
source_context must be an alphanumeric string of 1 through 79 characters.
This keyword should be configured if the source of the RADIUS requests is in a different context then
the IPSG service. If this keyword is not configured, the system will default to the context in which the
IPSG service is configured.
• +: Indicates that more than one of the preceding options may be specified in a single command.
Usage Guidelines Use this command to bind the IPSG RADIUS Server/eWAG service to a logical AAA interface and specify
the number of allowed subscriber sessions. If the AAA interface is not located in this context, configure the
source-context parameter.
Use the accounting and authentication proxy settings to enable RADIUS proxy server functionality on the
IPSG. These commands are used when the NAS providing the RADIUS request messages is incapable of
sending them to two separate devices. The IPSG in RADIUS Server mode proxies the RADIUS request and
response messages while performing the user identification task in order to provide services to the session.
Examples The following command binds the service to a AAA interface with and IP address of 10.2.3.4 located in the
source context named aaa_ingress:
bind address 10.2.3.4 source-context aaa_ingress
connection authorization
This command allows you to configure the RADIUS authorization password that must be matched by the
RADIUS accounting requests received by the current IPSG service.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
Deletes the RADIUS authorization from the current IPSG RADIUS Server service.
Usage Guidelines The IPSG RADIUS server service does not terminate RADIUS user authentication so the user password is
unknown.
Use this command to configure the authorization password that the RADIUS accounting requests must match
in order for the service to examine and extract user information.
Examples The following command sets the RADIUS authorization password that must be matched by the RADIUS
accounting requests sent to this service. The password is encrypted, and the password used in this example
is "secret".
connection authorization encrypted password secret
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
gtp max-contexts-per-imsi
This command allows you to configure multiple primary contexts having the same IMSI number.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command to disable use of multiple primary contexts. Only one PDP context per user is
allowed.
max-contexts-per-imsi: 1
min-nsapi: 15
max-contexts-per-imsi max_value
Specifies the limit for the maximum number of contexts per IMSI.
max_value must be an integer from 1 through 11.
min-nsapi min_nsapi_value
Specifies the range of NSAPI values to be assigned to different PDP context of the same subscriber.
min_nsapi_valuemust be an integer from 5 through 15.
Usage Guidelines Use this command to configure the maximum number of contexts per IMSI, and the range of NSAPI values
to be assigned to different PDP context.
Examples The following command configures the maximum contexts per IMSI to 5 and specify the range of values
NSAPI valie to 7.
gtp max-contexts-per-imsi 5 min-nsapi 7
gtp peer-ip-address
This command allows you to configure GGSN IP address under the eWAG service.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
Deletes the configuration, if previously configured.
Usage Guidelines Use this command to configure the GGSN IP address under the eWAG service.
This command replaces the hidden mode command [ no ] ggsn-ip-address ipv4_address
Examples The following command configures the GGSN IP address 1.2.3.4 under the current eWAG service.
gtp peer-ip-address 1.2.3.4
ip
This command enables you to configure IP parameters for the current eWAG service.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command, for specified option, with default setting for all QoS Class Identifier (QCI) values.
• QCI-based DSCP map:
◦qci 1: ef
◦qci 2: ef
◦qci 3: af11
◦qci 4: af11
◦qci 5: ef
◦qci 6: ef
◦qci 7: af21
◦qci 8: af21
◦qci 9: be
◦qci 5 allocation-retention-priority 1: ef
◦qci 5 allocation-retention-priority 2: ef
◦qci 5 allocation-retention-priority 3: ef
◦qci 6 allocation-retention-priority 1: ef
◦qci 6 allocation-retention-priority 2: ef
◦qci 6 allocation-retention-priority 3: ef
◦qci 7 allocation-retention-priority 1: af21
◦qci 7 allocation-retention-priority 2: af21
◦qci 7 allocation-retention-priority 3: af21
◦qci 8 allocation-retention-priority 1: af21
◦qci 8 allocation-retention-priority 2: af21
◦qci 8 allocation-retention-priority 3: af21
no
Resets configured value for specified QCI with its default setting.
gnp-qos-dscp
Specifies, for uplink direction, the DiffServ Code Point marking to be used for sending packets of a particular
3GPP QoS class.
qos-dscp
Specifies, for downlink direction, the DiffServ Code Point marking to be used for sending packets of a particular
3GPP QoS class.
qci { 1 | 2 | 3 | 4 | 9 }
Specifies the QCI attribute of QoS.
• 1: QCI 1 attribute of QoS
• 2: QCI 2 attribute of QoS
• 3: QCI 3 attribute of QoS
• 4: QCI 4 attribute of QoS
• 9: QCI 9 attribute of QoS
qci { 5 | 6 | 7 | 8 } allocation-retention-priority { 1 | 2 | 3 }
Specifies the QCI attribute of QoS with ARP.
• 5: QCI 5 attribute of QoS
• 6: QCI 6 attribute of QoS
af11 | af12 | af13 | af21 | af22 | af23 | af31 | af32 | af33 | af41 | af42 | af43 | be | ef | pt
Specifies the Per-Hop Forwarding Behavior (PHB) to use.
• af11: Assured Forwarding 11 PHB
• af12: Assured Forwarding 12 PHB
• af13: Assured Forwarding 13 PHB
• af21: Assured Forwarding 21 PHB
• af22: Assured Forwarding 22 PHB
• af23: Assured Forwarding 23 PHB
• af31: Assured Forwarding 31 PHB
• af32: Assured Forwarding 32 PHB
• af33: Assured Forwarding 33 PHB
• af41: Assured Forwarding 41 PHB
• af42: Assured Forwarding 42 PHB
• af43: Assured Forwarding 43 PHB
• be: Best Effort Forwarding PHB
• ef: Expedited Forwarding PHB
• pt: Pass Through (do not modify the ToS)
Usage Guidelines Use this command to configure IP parameters for the eWAG service.
Examples The following command specifies to configure the DiffServ Code Point marking to be used for sending packets
specifying QCI as 1 and Assured Forwarding 11 PHB:
ip gnp-qos-dscp qci 1 af11
map ue-mac-to-imei
This command allows you to map the UE MAC received in the Calling-Station-Id RADIUS attribute to
IMEIsV in order to forward it in the GTP CPC message to the GGSN.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
If previously configured, disables mapping of UE MAC address to IMEIsV IE of GTP message in order to
forward it to GGSN.
Default: Mapping is disabled.
no
If previously configured, disables mapping of UE MAC address to IMEIsV IE of GTP message in order to
forward it to GGSN.
Usage Guidelines Use this command to enable or disable mapping of UE MAC address to IMEIsV IE of GTP message in order
to forward it to GGSN.
overlapping-ip-address
This command allows you to enable or disable overlapping of IP addresses which enables multiple users to
use the same IP address.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
If previously configured, disables IPSG support of overlapping IP addresses.
Using overlapping IP addresses is disabled by default.
no
If previously configured, disables IPSG support of overlapping IP addresses.
Usage Guidelines Use this command to enable or disable overlapping IP addresses for subscribers on different networks that
are independent of each other.
plmn id
This command allows you to configure Public Land Mobile Network (PLMN) identifier for the current eWAG
service.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
If previously configured, deletes the PLMN ID configuration.
mcc mcc_number
Specifies the mobile country code (MCC) part of the PLMN identifier for the eWAG service.
mcc_number must be a three-digit number ranging from 200 to 999.
mnc mnc_number
Specifies the mobile network code (MNC) part of the PLMN identifier for the eWAG service.
mnc_number must be a two- or three-digit number ranging from 00 to 999.
Usage Guidelines Use this command to configure the location-specific mobile network identifiers included in the Routing Area
Identity (RAI) field of the PDP Create Request messages sent to the GGSN.
Important Any change to this configuration will result in restart of the eWAG service.
Examples The following command configures the PLMN identifier for the eWAG service as MCC 333 and MNC 99:
plmn id mcc 333 mnc 99
profile
This command allows you to configure the IPSG/eWAG service to use APN or subscriber profile.
Important In release 14.0, eWAG service uses only the APN profile. In release 15.0, ReWAG uses the APN profile
and DeWAG uses the subscriber profile. Whereas, the IPSG service uses both APN and subscriber profiles.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command with its default setting.
Default: APN
APN
Specifies to use APN profile for the service.
default-apn apn_name
subscriber
Important This option is supported only for the IPSG RADIUS Server service, and in release 15.0 for DeWAG
service. For the DeWAG service, this command must be configured with the subscriber option. This is
because DeWAG will operate based on subscriber template profile selection only for connecting users. If
the APN profile selection is configured, the DeWAG service will not be started.
Usage Guidelines Use this command to set the service to support APN profiles (supporting Gx through the enabling of
ims-auth-service) or for basic subscriber profile lookup.
For the DeWAG service, this command must be configured with the subscriber option. This is because
DeWAG will operate based on subscriber template profile selection only for connecting users. If the APN
profile selection is configured, the DeWAG service will not be started.
radius accounting
This command allows you to specify the IP address and shared secret of the RADIUS accounting client from
which RADIUS accounting requests are received. The RADIUS client can be either the access gateway or
the RADIUS accounting server depending on which device is sending accounting requests.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
Syntax Description radius accounting { client { ipv4_address | ipv4_address/mask } [ encrypted ] key key [ acct-onoff [
aaa-context aaa_context_name ] [ aaa-group aaa_server_group_name ] [ clear-sessions ] + ] [ dictionary
dictionary ] [ disconnect-message [ release-on-acct-stop acct_stop_wait_timeout ] [ dest-port
destination_port_number ] + | interim create-new-call | validate-client-ip }
no radius accounting { client { ipv4_address | ipv4_address/mask } | interim create-new-call |
validate-client-ip }
default radius accounting { interim create-new-call | validate-client-ip }
no
If previously configured, removes the specified configuration.
ipv4_address | ipv4_address/mask
Specifies the IP address, and optionally subnet mask of the RADIUS client from which RADIUS accounting
requests are received.
ipv4_address/ipv4_address/mask must be in IPv4 dotted-decimal notation.
A maximum of 16 IP addresses can be configured.
In StarOS 12.2 and later releases, key with encryption must be an alphanumeric string of 1 through 236
characters, and without encryption an alphanumeric string of 1 through 127 characters. Note that key is
case sensitive.
In StarOS 12.1 and earlier releases, key must be an alphanumeric string of 1 through 127 characters and
is case sensitive.
Important In release 12.3 and earlier releases, this option is applicable only to the IPSG Proxy Mode.
Important In release 14.0 and later releases, this option is applicable to the IPSG Proxy and Server Modes.
dictionary dictionary
Specifies the dictionary to use.
Dictionary Description
3gpp2 This dictionary consists not only of all of the attributes in the standard
dictionary, but also all of the attributes specified in IS-835-A.
3gpp2-835 This dictionary consists not only of all of the attributes in the standard
dictionary, but also all of the attributes specified in IS-835.
Dictionary Description
standard This dictionary consists only of the attributes specified in RFC 2865, RFC
2866, and RFC 2869.
starent This dictionary consists of all of the attributes in the starent-vsa1 dictionary
and incorporates additional Starent Networks VSAs by using a two-byte
VSA Type field. This dictionary is the master-set of all of the attributes in
all of the dictionaries supported by the system.
starent-vsa1 This dictionary consists not only of the 3GPP2 dictionary, but also includes
Starent Networks vendor-specific attributes (VSAs) as well. The VSAs in
this dictionary support a one-byte wide VSA Type field in order to support
certain RADIUS applications. The one-byte limit allows support for only
256 VSAs (0–255). This is the default dictionary.
Important In StarOS 12.0 and later releases, no new attributes can be
added to the starent-vsa1 dictionary. If there are new attributes
to be added, you can only add them to the starent dictionary.
For more information, please contact your Cisco account
representative.
starent-vsa1-835 This dictionary consists not only of the 3GPP2-835 dictionary, but also
includes Starent Networks vendor-specific attributes (VSAs) as well. The
VSAs in this dictionary support a one-byte wide VSA Type field in order
to support certain RADIUS applications. The one-byte limit allows support
for only 256 VSAs (0–255). This is the default dictionary.
Important For information on the specific dictionary to use for your deployment contact your Cisco account
representative.
This keyword is configured on a per RADIUS accounting client basis and not for the entire service.
• dest-port destination_port_number: Specifies the port number to which the disconnect message must
be sent.
destination_port_number must be an integer from 1 through 65535.
interim create-new-call
Important This option does not apply to the IPSG Proxy Mode.
validate-client-ip
Specifies to enable the ipsgmgr to validate RADIUS accounting messages from different configured RADIUS
client IP address, and forward requests to the session manager.
Default: The RADIUS client IPs are validated.
Usage Guidelines Use this command to configure the communication parameters for the RADIUS client from which RADIUS
accounting requests are received.
Examples The following command configures the service to communicate with a RADIUS client with an IP address of
10.2.3.4 and an encrypted shared secret of key1234:
radius accounting client 10.2.3.4 encrypted key key1234
radius dictionary
This command allows you to specify the RADIUS dictionary for the current IPSG/eWAG service.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Specifies to use the default dictionary.
Default: starent-vsa1
dictionary dictionary_name
Specifies the dictionary to use.
Important In 15.0 and later releases, for DeWAG use the starent dictionary.
Dictionary Description
3gpp2 This dictionary consists not only of all of the attributes in the standard
dictionary, but also all of the attributes specified in IS-835-A.
3gpp2-835 This dictionary consists not only of all of the attributes in the standard
dictionary, but also all of the attributes specified in IS-835.
Dictionary Description
customXX These are customized dictionaries. For information on custom dictionaries,
please contact your Cisco account representative.
XX is the integer value of the custom dictionary.
standard This dictionary consists only of the attributes specified in RFC 2865, RFC
2866, and RFC 2869.
starent This dictionary consists of all of the attributes in the starent-vsa1 dictionary
and incorporates additional Starent Networks VSAs by using a two-byte
VSA Type field. This dictionary is the master-set of all of the attributes in
all of the dictionaries supported by the system.
starent-vsa1 This dictionary consists not only of the 3GPP2 dictionary, but also includes
Starent Networks vendor-specific attributes (VSAs) as well. The VSAs in
this dictionary support a one-byte wide VSA Type field in order to support
certain RADIUS applications. The one-byte limit allows support for only
256 VSAs (0–255). This is the default dictionary.
starent-vsa1-835 This dictionary consists not only of the 3GPP2-835 dictionary, but also
includes Starent Networks vendor-specific attributes (VSAs) as well. The
VSAs in this dictionary support a one-byte wide VSA Type field in order
to support certain RADIUS applications. The one-byte limit allows support
for only 256 VSAs (0–255). This is the default dictionary.
Important For information on the specific dictionary to use for your deployment contact your Cisco account
representative.
Usage Guidelines Use this command to specify the RADIUS dictionary to use for the IPSG RADIUS Server/eWAG service.
Examples The following command specifies to use the custom10 RADIUS dictionary:
radius dictionary custom10
respond-to-non-existing-session
Configures the IPSG service to respond to Radius Accounting-Stop messages even if a session does not exist.
Product IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command with its default setting.
Default: Disabled. IPSG service drops packets containing the Radius Accounting-Stop message if the session
does not exist.
no
If previously enabled, disables the configuration.
Usage Guidelines Use this command to enable/disable the IPSG service to respond to Radius Accounting-Stop messages with
a Radius Accounting-Response message for non-existing sessions.
sess-replacement
This command allows you to enable/disable the Session Replacement feature for eWAG and IPSG services.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command with its default setting.
Default: Disabled.
no
If previously configured, deletes the configuration.
with-diff-acct-sess-id
Specifies to replace current session when a new session request comes with same IP address and same user
name/IMSI but different accounting session ID.
with-diff-ip
Specifies to replace current session when a new session request comes with same user name/IMSI but different
IP address.
with-diff-key [ with-diff-acct-sess-id ]
Specifies to replace current session when a new session request comes with same IP address but different user
name/IMSI.
For IPSG, you can also use a combination of replacement options of different key and different account session
ID.
Usage Guidelines Use this command to enable/disable the Session Replacement feature. By default, the Session Replacement
feature is disabled.
Examples The following command enables session replacement specifying to replace the current session when a new
session request comes with same user name/IMSI but different IP address:
sess-replacement with-diff-ip
setup-timeout
This command allows you to configure a timeout for session setup attempts for the current IPSG/eWAG
service.
Product eWAG
IPSG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
default
Configures this command with its default setting.
Default: 60 seconds
setup_timeout_seconds
Specifies the time period, in seconds, for which a session setup attempt is allowed to continue before being
terminated.
setup_timeout_seconds must be an integer from 1 through 1000000.
Usage Guidelines Use this command to configure a timeout for IPSG/eWAG session setup attempts.
Examples The following command configures the timeout for session setup attempts to 30 seconds:
setup-timeout 30
w-apn
This command allows you to configure the W-APNs that can be connected through DeWAG, and the
default-gateway IP addresses to be used by the UEs for connecting to the W-APN network.
Product eWAG
Command Modes Exec > Global Configuration > Context Configuration > IPSG RADIUS Server Configuration
configure > context context_name > ipsg-service service_name mode radius-server
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipsg-service-radius-server)#
no
If previously configured, removes the specified configuration.
apn-name apn_name
Specifies the APN name.
apn_name must be the name of an APN and must be a string of 1 to 62 characters in length consisting of
alphabetic characters (A-Z and a-z), digits (0-9), dot(.) and the dash (-).
This value is compared against the subscribed APN returned by the AAA server or locally configured APN
in the subscriber-template configuration to find the default-gateway IP address to be used in DHCP signaling
packets.
default-gw ipv4/ipv6_address/maskbits
Specifies the IP address of the default gateway to be used by UE for W-APN access.
You can configure a maximum of four default gateways per W-APN. Multiple default-gateways are possible
as the APN can have different pools of different subnet with different default-gateway IP addresses.
ipv4/ipv6_address/maskbits must be an IPv4/IPv6 address and subnet-mask, for example 192.168.1.1/24.
This value should be in the same subnet as that of UE allocated IP address from GGSN for the W-APN. GGSN
does not supply subnet-mask along with IP address. Therefore, the identification of whether GGSN-allocated
IP address is in same subnet or not is done with the help of configured "/maskbits". This default-gateway
value is sent to the UE as default-gateway IP address using "Router" option in DHCP-OFFER message. The
maskbits is sent to the UE as subnet-mask using the "Subnet Mask" option in DHCP-OFFER message.
Usage Guidelines Use this command to configure the list of W-APN names that can be connected through DeWAG and the
default-gateway IP addresses to be used by UE for connecting to the W-APN network. During DHCP signaling
the configured default-gateway value will be notified to UE as the router. This command also configures the
subnet-mask to be used for the respective default-gateway IP address in order to find the network prefix of
the default-gateway.
Note that DeWAG will be acting as 'default-gateway' for the UE in its connected network.
Important This command can be configured a maximum of four times to configure four different APNs and the
corresponding default-gateways.
Examples The following command configures an APN named apn123 with the default gateway IP address and mask
192.168.1.1/24:
w-apn apn123 default-gw 192.168.1.1/24
Important For information on configuring and using IPSP refer to the System Administration Guide.
The IPSP Configuration Mode is used to configure properties for the IP Pool Sharing Protocol (IPSP).
System-based HA services use IPSP during an offline-software upgrade to avoid the assignment of duplicate
IP addresses to sessions while allowing them to maintain the same address, and to preserve network capacity
Command Modes Exec > Global Configuration > Context Configuration > Ethernet Interface Configuration > IP Pool Sharing
Protocol Configuration
configure > context context_name > interface interface_name broadcast > pool-share-protocol { primary
ip_address | secondary ip_address }
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
dead-interval
Configures the retry time to connect to the remote system for the IP Pool Sharing Protocol.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > Ethernet Interface Configuration > IP Pool Sharing
Protocol Configuration
configure > context context_name > interface interface_name broadcast > pool-share-protocol { primary
ip_address | secondary ip_address }
no
Disables the dead interval. On loss of connectivity to the remote system, no retries are attempted and the
remote system is marked dead immediately on failure.
default
Resets the dead interval to the default of 3600 seconds.
seconds
Default: 3600 seconds
The amount of time in seconds to wait before retrying the remote system. seconds must be an integer from
25 through 259200.
Usage Guidelines Use this command to set the amount of time to wait before retrying to connect with the remote system for the
IP pool sharing protocol.
Examples Use the following command to set the interval to 180 seconds (3 minutes):
dead-interval 180
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
reserved-free-percentage
This command is used to set the amount of free addresses reserved for use on the primary HA.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > Ethernet Interface Configuration > IP Pool Sharing
Protocol Configuration
configure > context context_name > interface interface_name broadcast > pool-share-protocol { primary
ip_address | secondary ip_address }
value
Default: 100
value specifies the percentage of free addresses reserved for the use on the primary HA for IP pool sharing
during upgrade. It must be an integer from 0 through 100.
Usage Guidelines This command is used with pool-sharing-protocol active mode on the primary HA. Before using this
command, pool-sharing-protocol in the Ethernet Interface Configuration Mode must be configured.
For more information, refer to the Ethernet Interface Configuration Mode Commands chapter in this guide.
Examples To reserve 40 percent of free addresses in primary HA for IP pool sharing, enter the following command:
reserved-free-percentage 40
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: Indicates the rule, when matched, drops the corresponding packets.
• permit: Indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
Usage Guidelines Define a rule when any packet from the IP addresses which fall into the group of addresses matching the IP
address masking. This allows the reduction of filtering rules as it does not require a rule for each source and
destination pair.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide.
Examples The following command defines two rules with the second logging filtered packets:
permit 2001:4A2B::1f3F
deny log 2001:4A2B::1f3F
The following sets the insertion point to before the first rule defined above:
before permit 2001:4A2B::1f3F
The following command sets the insertion point after the second rule defined above:
after deny log 2001:4A2B::1f3F
The following deletes the first rule defined above:
no permit 2001:4A2B::1f3F
deny/permit (any)
Used to filter subscriber sessions based on any packet received. This command is also used to set the access
control list insertion point.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: Indicates the rule, when matched, drops the corresponding packets.
• permit: Indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
any
Indicates all packets will match the filter regardless of source and/or destination.
Usage Guidelines Define a catch all rule to place at the end of the list of rules.
Important It is suggested that any rule which is added to be a catch all should also have the log option specified. The
logged packets may be used to determine if the current list of rules is adequate or needs modification to
ensure proper security. The maximum number of rules that can be configured per ACL varies depending
on how the ACL is to be used. For more information, refer to the Engineering Rules appendix in the System
Administration Guide.
Examples The following command defines two rules with the second logging filtered packets:
permit any
deny log any
The following sets the insertion point to before the first rule defined above:
before permit any
The following command sets the insertion point after the second rule defined above:
after deny log any
The following deletes the first rule defined above:
no permit any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: Indicates the rule, when matched, drops the corresponding packets.
• permit: Indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
Usage Guidelines Define a rule when a very specific remote host is to be blocked. In simplified networks where the access
controls need only block a few hosts, this command allows the rules to be very clear and concise.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide.
Examples The following command defines two rules with the second logging filtered packets:
permit host 2001:4A2B::1f3F
deny log host 2001:4A2B::1f3F
The following sets the insertion point to before the first rule defined above:
before permit host 2001:4A2B::1f3F
The following command sets the insertion point after the second rule defined above:
after deny log host 2001:4A2B::1f3F
The following deletes the first rule defined above:
no permit host 2001:4A2B::1f3F
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description { deny | permit } [ log ] icmp { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
after { deny | permit } [ log ] icmp { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
before { deny | permit } [ log ] icmp { source_address source_wildcard | any | host source_host_address }
{ dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
no { deny | permit } [ log ] icmp { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: Indicates the rule, when matched, drops the corresponding packets.
• permit: Indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 hexadecimal-colon-separated notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 hexadecimal-colon-separated notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. The type is an integer from 0 through
255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered. The type is an integer from 0 through
255.
Usage Guidelines Define a rule to block ICMP packets which can be used for address resolution and possible be a security risk.
The IP filtering allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the filtering of entire subnets if necessary.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide.
Examples The following command defines two rules with the second logging filtered packets:
permit icmp host 2001:4A2B::1f3F4 any 168
deny log icmp 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410 168 11
The following sets the insertion point to before the first rule defined above:
before permit icmp host 2001:4A2B::1f3F any 168
The following command sets the insertion point after the second rule defined above:
after deny log icmp 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410 168 11
The following deletes the first rule defined above:
no permit icmp host 2001:4A2B::1f3F any 168
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description { deny | permit } [ log ] ip { source_address source_wildcard | any | host source_host_address } { dest_address
dest_wildcard | any | host dest_host_address } [ fragment ] [ protocolnum ]
after { deny | permit } [ log ] ip { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocolnum ]
before { deny | permit } [ log ] ip { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocolnum ]
no { deny | permit } [ log ] ip { source_address source_wildcard | any | host source_host_address } {
dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocolnum ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: indicates the rule, when matched, drops the corresponding packets.
• permit: indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
fragment
Indicates packet filtering is to be applied to IP packet fragments only.
protocol num
Indicates that the packet filtering is to be applied to a specific protocol number.
num can be any integer ranging from 0 to 255.
Usage Guidelines Block IP packets when the source and destination are of interest.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide.
Examples The following command defines two rules with the second logging filtered packets:
permit ip host 2001:4A2B::1f3F any fragment
deny log ip 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410
The following sets the insertion point to before the first rule defined above:
before permit ip host 2001:4A2B::1f3F any fragment
The following command sets the insertion point after the second rule defined above:
after deny log ip 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410
The following deletes the first rule defined above:
no permit ip host 2001:4A2B::1f3F any fragment
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description { deny | permit } [ log ] { tcp | udp } { { source_address source_wildcard | any | host source_host_address
} [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address dest_wildcard |
any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
after { deny | permit } [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
before { deny | permit } [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
no { deny | permit } [ log ] { tcp | udp } { { source_address source_wildcard | any | host source_host_address
} [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address dest_wildcard |
any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
deny | permit
Specifies the rule is either block (deny) or an allow (permit) filter.
• deny: Indicates the rule, when matched, drops the corresponding packets.
• permit: Indicates the rule, when matched, allows the corresponding packets.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
tcp | udp
Specifies the filter is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Filter applies to TPC packets.
• udp: Filter applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer from 0 through 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer from 0 through 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide.
Examples The following command defines four rules with the second and fourth rules logging filtered packets:
permit tcp host 2001:4A2B::1f3F any
deny log udp 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410
permit tcp host 2001:4A2B::1f3F gt 1023 any
The following sets the insertion point to before the first rule defined above:
before permit tcp host 2001:4A2B::1f3F any
The following command sets the insertion point after the second rule defined above:
after deny log udp 2001:4A2B::1f3F 2001:4a2b::1f00 host fe80::a02:410
The following deletes the third rule defined above:
no permit tcp host 2001:4A2B::1f3F gt 1023 any
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
readdress server
Alter the destination address and port number in TCP or UDP packet headers to redirect packets to a different
server.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description readdress server redirect_address [ port port_number ] { tcp | udp } { { source_address source_wildcard
| any | host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } {
{ dest_address dest_wildcard any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq
dst_port ] }
after readdress server redirect_address [ port port_no ] { tcp | udp } { { source_address source_wildcard
| any | host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } {
{ dest_address dest_wildcard any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq
dst_port ] }
before readdress server redirect_address [ port port_no ] { tcp | udp } { { source_address source_wildcard
| any | host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } {
{ dest_address dest_wildcard any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq
dst_port ] }
no readdress server redirect_address [ port port_number ] { tcp | udp } { { source_address source_wildcard
| any | host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } {
{ dest_address dest_wildcard any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq
dst_port ] }
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
redirect_address
The IP address to which the IP packets are redirected. TCP or UDP packet headers are rewritten to contain
the new destination address. This must expressed in IPv6 colon-separated-hexadecimal notation.
port port_number
The number of the port at the redirect address where the packets are sent. TCP or UDP packet headers are
rewritten to contain the new destination port number.
tcp | udp
Specifies the redirect is to be applied to the IP-based transmission control protocol or the user datagram
protocol.
• tcp: Redirect applies to TCP packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer from 0 through 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer from 0 through 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
Usage Guidelines Use this command to define a rule that redirects packets to a different destination address. The TCP and UDP
packet headers are modified with the new destination address and destination port.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the server at fe80::c0a8:a04, UDP packets
coming from any host with a destination of any host are matched:
readdress server fe80::c0a8:a04 udp any any
The following sets the insertion point to before the rule defined above:
before readdress server fe80::c0a8:a04 udp any any
The following deletes the rule defined above:
no readdress server fe80::c0a8:a04 udp any any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
Specifies the context identification number of the context to which packets are redirected. At the executive
mode prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
Filters by the IP address(es) from which the packet originated. This option filters all packets from a specific
IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
Filters packets for a group of addresses specified in conjunction with the source_address option.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
Usage Guidelines Define a rule when any packet from the IP addresses which fall into the group of addresses matching the IP
address masking. This allows the reduction of redirect rules as it does not require a rule for each source and
destination pair.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23 and the
source IP and wildcard of 2002::c6a2:1600 and 2002::c6a2:1600:
redirect context 23 2002::c6a2:1600 2002::c6a2:1600
The following sets the insertion point to before the first rule defined above:
before redirect context 23 2002::c6a2:1600 2002::c6a2:1600
The following command sets the insertion point after the second rule defined above:
after redirect context 23 2002::c6a2:1600 2002::c6a2:1600
The following deletes the first rule defined above:
no redirect context 23 2002::c6a2:1600 2002::c6a2:1600
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
any
Indicates all packets will match the redirect regardless of source and/or destination.
Usage Guidelines Define a catch all rule to place at the end of the list of rules to provide explicit handling of rules which do not
fit any other criteria.
Important It is suggested that any rule which is added to be a catch all should also have the log option specified. The
logged packets may be used to determine if the current list of rules is adequate or needs modification to
ensure proper security. The maximum number of rules that can be configured per ACL varies depending
on how the ACL is to be used. For more information, refer to the Engineering Rules appendix in the System
Administration Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers
or all subscribers facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23 and any
source IP:
redirect context 23 any
The following sets the insertion point to before the first rule defined above:
before redirect context 23 any
The following command sets the insertion point after the second rule defined above:
after redirect context 23 any
The following deletes the first rule defined above:
no redirect context 23 any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
Usage Guidelines Define a rule when a very specific remote host is to be blocked. In simplified networks where the access
controls need only block a few hosts, this command allows the rules to be very clear and concise.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23 and a
host IP address of fe80::c0a8:c80b:
redirect context 23 host fe80::c0a8:c80b
The following sets the insertion point to before the first rule defined above:
before redirect context 23 host fe80::c0a8:c80b
The following command sets the insertion point after the second rule defined above:
after redirect context 23 host fe80::c0a8:c80b
The following deletes the first rule defined above:
no redirect context 23 host fe80::c0a8:c80b
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect context context_id [ log ] icmp { source_address source_wildcard | any | host source_host_address
} { dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
after redirect context context_id [ log ] icmp { source_address source_wildcard | any | host
source_host_address } { dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code
]]
before redirect context context_id [ log ] icmp { source_address source_wildcard | any | host
source_host_address } { dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code
]]
no redirect context context_id [ log ] icmp { source_address source_wildcard | any | host source_host_address
} { dest_address dest_wildcard | any | host dest_host_address } [ icmp_type [ icmp_code ] ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. Type is an integer from 0 through 255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered type is an integer from 0 through 255.
Usage Guidelines Define a rule to block ICMP packets which can be used for address resolution and possibly be a security risk.
The IP redirecting allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the redirecting of entire subnets if necessary.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23, and
ICMP packets coming from the host with the IP address 2002::c6a2:6419:
redirect context 23 icmp host 2002::c6a2:6419
The following sets the insertion point to before the first rule defined above:
before redirect context 23 icmp host 2002::c6a2:6419
The following command sets the insertion point after the second rule defined above:
after redirect context 23 icmp host 2002::c6a2:6419
The following deletes the first rule defined above:
no redirect context 23 icmp host 2002::c6a2:6419
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect context context_id [ log ] ip { source_address source_wildcard | any | host source_host_address }
{ dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocol num ]
after redirect context context_id [ log ] ip { source_address source_wildcard | any | host source_host_address
} { dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocol num ]
before redirect context context_id [ log ] ip { source_address source_wildcard | any | host
source_host_address } { dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocol
num ]
no redirect context context_id [ log ] ip { source_address source_wildcard | any | host source_host_address
} { dest_address dest_wildcard | any | host dest_host_address } [ fragment ] [ protocol num ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
fragment
Indicates packet redirection is to be applied to IP packet fragments only.
protocol num
Indicates that the packet filtering is to be applied to a specific protocol number.
num is an integer from 0 through 255.
Usage Guidelines Block IP packets when the source and destination are of interest.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23, and IP
packets coming from the host with the IP address 2002::c6a2:6419, and fragmented packets for any destination
are matched:
redirect context 23 ip host 2002::c6a2:6419 any fragment
The following sets the insertion point to before the first rule defined above:
before redirect context 23 ip host 198.162.100.25 any fragment
The following command sets the insertion point after the second rule defined above:
after redirect context 23 ip host 2002::c6a2:6419 any fragment
The following deletes the first rule defined above:
no redirect context 23 ip host 2002::c6a2:6419 any fragment
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect context context_id [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
after redirect context context_id [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
before redirect context context_id [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
no redirect context context_id [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port ] } { { dest_address
dest_wildcard | any | host dest_host_address } [ eq dest_port | gt dest_port | lt dest_port | neq dst_port ] }
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
tcp | udp
Specifies the redirect is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Redirect applies to TPC packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer from 0 through 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer from 0 through 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer from 0 through 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer from 0 through 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the context with the context ID of 23, and
UDP packets coming from any host are matched:
redirect context 23 udp any
The following sets the insertion point to before the rule defined above:
before redirect context 23 udp any
The following command sets the insertion point after the rule defined above:
after redirect context 23 udp any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rule definitions defined subsequent to this command are to be inserted after the command identified
by the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definitions which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
any
Indicates all packets will match the redirect regardless of source and/or destination.
Usage Guidelines Define a catch all rule definitions to place at the end of the list of rule definitions to provide explicit handling
of rule definitions which do not fit any other criteria.
Important It is suggested that any rule definition which is added to be a catch all should also have the log option
specified. The logged packets may be used to determine if the current list of rule definitions is adequate
or needs modification to ensure proper security.
Important Also note that "redirect" rule definitions are ignored for ACLs applied to specific subscribers or all
subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service with the name
css-svc1 and any source IP:
redirect css service css-svc1 any
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 any
The following command sets the insertion point after the second rule definitions above:
after redirect css service css-svc1 any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
Usage Guidelines Define a rule definition when a very specific remote host is to be blocked. In simplified networks where the
access controls need only block a few hosts, this command allows the rule definitions to be very clear and
concise.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service with the name
css-svc1 and a host IP address of fe80::c0a8:c80b:
redirect css service css-svc1 host fe80::c0a8:c80b
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 host fe80::c0a8:c80b
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 host fe80::c0a8:c80b
The following deletes the first rule definition above:
no redirect css service css-svc1 host fe80::c0a8:c80b
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] icmp { any | host source_host_address | source_address source_wildcard
} { any | host dest_host_address | dest_address dest_wildcard } [icmp_type [ icmp_code ] ]
after redirect css service svc_name [ log ] icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [icmp_type [ icmp_code ]
]
before redirect css service svc_name [ log ] icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [icmp_type [ icmp_code ]
]
no redirect css service svc_name [ log ] icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [icmp_type [ icmp_code ]
]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. The type can be an integer value from
0 through 255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered. The type is an integer from 0 through
255.
Usage Guidelines Define a rule definition to block ICMP packets which can be used for address resolution and possibly be a
security risk.
The IP redirecting allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the redirecting of entire subnets if necessary.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service named css-svc1,
and ICMP packets coming from the host with the IP address 2002::c6a2:6419:
redirect css service css-svc1 icmp host 2002::c6a2:6419
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 icmp host 2002::c6a2:6419
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 icmp host 2002::c6a2:64195
The following deletes the first rule definition above:
no redirect css service css-svc1 icmp host 2002::c6a2:6419
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] ip { any | host source_host_address | source_address source_wildcard
} { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after redirect css service svc_name [ log ] ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
before redirect css service svc_name [ log ] ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
no redirect css service svc_name [ log ] ip { any | host source_host_address | source_address source_wildcard
} { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions defined subsequent to this command are to be inserted before the command
identified by the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
fragment
Indicates packet redirection is to be applied to IP packet fragments only.
Usage Guidelines Block IP packets when the source and destination are of interest.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service named css-svc1,
and IP packets coming from the host with the IP address 2002::c6a2:6419, and fragmented packets for any
destination are matched:
redirect css service css-svc1 ip host 2002::c6a2:6419 any fragment
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 ip host 2002::c6a2:6419 any fragment
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 ip host 2002::c6a2:6419 any fragment
The following deletes the first rule definition above:
no redirect css service css-svc1 ip host 2002::c6a2:6419 any fragment
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
Usage Guidelines Define a rule definition when any packet from the IP addresses which fall into the group of addresses matching
the IP address masking. This allows the reduction of filtering rule definitions as it does not require a rule
definition for each source and destination pair.
Examples The following command defines a rule definition to redirect packets to a CSS service named css-svc1:
redirect css service css=svc1 2002::c6a2:6419
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after redirect css service svc_name [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after redirect css service svc_name [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
no redirect css service svc_name [ log ] { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
tcp | udp
Specifies the redirect is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Redirect applies to TPC packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer from 0 to 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer from 0 to 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer from 0 to 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer from 0 to 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer from 0 to 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer from 0 to 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer from 0 to 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer from 0 to 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service named css-svc1,
and UDP packets coming from any host are matched:
redirect css service css-svc1 udp any
The following sets the insertion point to before the rule definition above:
before redirect css service css-svc1 udp any
The following command sets the insertion point after the rule definition above:
after redirect css service css-svc1 udp any
The following deletes the rule definition above:
no redirect css service css-svc1 udp any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rule definitions defined subsequent to this command are to be inserted after the command identified
by the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
any
Indicates all packets will match the redirect regardless of source and/or destination.
Usage Guidelines Define a catch all rule definition to place at the end of the list of rule definitions to provide explicit handling
of rule definitions which do not fit any other criteria.
Important It is suggested that any rule definition which is added to be a catch all should also have the log option
specified. The logged packets may be used to determine if the current list of rule definitions is adequate
or needs modification to ensure proper security.
Important Also note that "redirect" rule definitions are ignored for ACLs applied to specific subscribers or all
subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the CSS service with the name
css-svc1 and any source IP:
redirect css service css-svc1 downlink any
The following sets the insertion point to before the first rule definition above:
before redirect service css-svc1 downlink any
The following command sets the insertion point after the second rule definition above:
after redirect service css-svc1 downlink any chgsvc1 downlink any
The following deletes the first rule definition above:
no redirect service css-svc1 downlink any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] downlink host source_host_address
after redirect css service svc_name [ log ] downlink host source_host_address
before redirect css service svc_name [ log ] downlink host source_host_address
no redirect css service svc_name [ log ] downlink host source_host_address
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions defined subsequent to this command are to be inserted before the command
identified by the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
Usage Guidelines Define a rule definition when a very specific remote host is to be blocked. In simplified networks where the
access controls need only block a few hosts, this command allows the rule definitions to be very clear and
concise.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service with the name
css-svc1 and a host IP address of fe80::c0a8:c80b:
redirect service css-svc1 downlink host fe80::c0a8:c80b
The following sets the insertion point to before the first rule definition above:
before redirect service css-svc1 downlink host fe80::c0a8:c80b
The following command sets the insertion point after the second rule definition above:
after redirect service css-svc1 downlink host fe80::c0a8:c80b
The following deletes the first rule definition above:
no redirect service css-svc1 downlink host fe80::c0a8:c80b
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] downlink icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ icmp_type [ icmp_code ]
]
after redirect css service svc_name [ log ] downlink icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ icmp_type [ icmp_code
]]
before redirect css service svc_name [ log ] downlink icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ icmp_type [ icmp_code
]]
no redirect css service svc_name [ log ] downlink icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ icmp_type [ icmp_code
]]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. The type can be an integer value from
0 through 255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered. The type can be an integer value from
0 through 255.
Usage Guidelines Define a rule definition to block ICMP packets which can be used for address resolution and possibly be a
security risk.
The IP redirecting allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the redirecting of entire subnets if necessary.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service named css-svc1,
and ICMP packets coming in the downlink (from the Mobile Node) direction from the host with the IP address
2002::c6a2:6419:
redirect css service css-svc1 downlink icmp host 2002::c6a2:6419
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 downlink icmp host 2002::c6a2:6419
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 downlink icmp host 2002::c6a2:6419
The following deletes the first rule definition above:
no redirect css service css-svc1 downlink icmp host 2002::c6a2:6419
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] downlink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after redirect css service svc_name [ log ] downlink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
before redirect css service svc_name [ log ] downlink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
no redirect css service svc_name [ log ] downlink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions defined subsequent to this command are to be inserted before the command
identified by the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
fragment
Indicates packet redirection is to be applied to IP packet fragments only.
Usage Guidelines Block IP packets when the source and destination are of interest.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service named css-svc1,
and downlink IP packets coming from the host with the IP address 2002::c6a2:6419, and fragmented packets
for any destination are matched:
redirect css service css-svc1 downlink ip host 2002::c6a2:6419 any fragment
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 downlink ip host 2002::c6a2:6419 any fragment
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 downlink ip host 2002::c6a2:6419 any fragment
The following deletes the first rule definition above:
no redirect css service css-svc1 downlink ip host 2002::c6a2:6419 any fragment
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] downlink source_address source_wildcard
after redirect css service svc_name [ log ] downlink source_address source_wildcard
before redirect css service svc_name [ log ] downlink source_address source_wildcard
no redirect css service svc_name [ log ] downlink source_address source_wildcard
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions defined subsequent to this command are to be inserted before the command
identified by the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
Usage Guidelines Define a rule definition when any packet from the IP addresses which fall into the group of addresses matching
the IP address masking. This allows the reduction of filtering rule definitions as it does not require a rule
definition for each source and destination pair.
Examples The following command defines a rule definition to redirect packets to a charging service named css-svc1:
redirect css service css-svc1 donwlink fe80::c0a8:a04
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] downlink { tcp | udp } { { source_address source_wildcard | any | host
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after redirect css service svc_name [ log ] downlink { tcp | udp } { { source_address source_wildcard | any
| host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after redirect css service svc_name [ log ] downlink { tcp | udp } { { source_address source_wildcard | any
| host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
no redirect css service svc_name [ log ] downlink { tcp | udp } { { source_address source_wildcard | any |
host source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
downlink
Apply this rule definition only to packets in the downlink (from the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
tcp | udp
Specifies the redirect is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Redirect applies to TPC packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer value from 0 to 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service named css-svc1,
and UDP packets coming from any host are matched:
redirect css service css-svc1 downlink udp any
The following sets the insertion point to before the rule definition above:
before redirect css service css-svc1 downlink udp any
The following command sets the insertion point after the rule definition above:
after redirect css service css-svc1 downlink udp any
The following deletes the rule definition above:
no redirect css service css-svc1 downlink udp any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
any
Indicates all packets will match the redirect regardless of source and/or destination.
Usage Guidelines Define a catch all rule definition to place at the end of the list of rule definitions to provide explicit handling
of rule definitions which do not fit any other criteria.
Important It is suggested that any rule definition which is added to be a catch all should also have the log option
specified. The logged packets may be used to determine if the current list of rule definitions is adequate
or needs modification to ensure proper security.
Important Also note that "redirect" rule definitions are ignored for ACLs applied to specific subscribers or all
subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service with the name
css-svc1 and any source IP:
redirect css service css-svc1 uplink any
The following sets the insertion point to before the first rule definition above:
before redirect css service css-svc1 uplink any
The following command sets the insertion point after the second rule definition above:
after redirect css service css-svc1 uplink any
The following deletes the first rule definition above:
no redirect css service css-svc1 uplink any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] uplink host source_host_address
after redirect css service svc_name [ log ] uplink host source_host_address
before redirect css service svc_name [ log ] uplink host source_host_address
no redirect css service svc_name [ log ] uplink host source_host_address
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
after
Indicates all rule definitions defined subsequent to this command are to be inserted after the command identified
by the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
Usage Guidelines Define a rule definition when a very specific remote host is to be blocked. In simplified networks where the
access controls need only block a few hosts, this command allows the rule definitions to be very clear and
concise.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service with the name
css-svc1 and a host IP address of fe80::c0a8:c80b:
redirect service css-svc1 uplink host fe80::c0a8:c80b
The following sets the insertion point to before the first rule definition above:
before redirect service css-svc1 uplink host fe80::c0a8:c80b
The following command sets the insertion point after the second rule definition above:
after redirect service css-svc1 uplink host fe80::c0a8:c80b
The following deletes the first rule definition above:
no redirect service css-svc1 uplink host fe80::c0a8:c80b
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] uplink icmp { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ icmp_type [ icmp_code ]
]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. The type can be an integer value from
0 through 255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered. The type can be an integer value from
0 through 255.
Usage Guidelines Define a rule definition to block ICMP packets which can be used for address resolution and possibly be a
security risk.
The IP redirecting allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the redirecting of entire subnets if necessary.
Important A maximum of 16 rule definitions can be configured per ACL. Also note that "redirect" rule definitions
are ignored for ACLs applied to specific subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule definition that redirects packets to the charging service named chgsvc1,
and ICMP packets in the uplink (to the Mobile Node) direction from the host with the IP address
198.162.100.25:
redirect css service chgsvc1 uplink icmp host 198.162.100.25
The following sets the insertion point to before the first rule definition above:
before redirect css service chgsvc1 uplink icmp host 198.162.100.25
The following command sets the insertion point after the second rule definition above:
after redirect css service chgsvc1 uplink icmp host 198.162.100.25
The following deletes the first rule definition above:
no redirect css service chgsvc1 uplink icmp host 198.162.100.25
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] uplink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after redirect css service svc_name [ log ] uplink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
before redirect css service svc_name [ log ] uplink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
no redirect css service svc_name [ log ] uplink ip { any | host source_host_address | source_address
source_wildcard } { any | host dest_host_address | dest_address dest_wildcard } [ fragment ]
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
Important If the options specified do not exactly match an existing rule definition, the insertion point does not change.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
fragment
Indicates packet redirection is to be applied to IP packet fragments only.
Usage Guidelines Block IP packets when the source and destination are of interest.
Examples The following command defines a rule definition that redirects packets to the charging service named chgsvc1,
and uplink IP packets going to the host with the IP address 198.162.100.25, and fragmented packets for any
destination are matched:
redirect css service chgsvc1 uplink ip host 198.162.100.25 any fragment
The following sets the insertion point to before the first rule definition above:
redirect css service chgsvc1 uplink ip host 198.162.100.25 any fragment
The following command sets the insertion point after the second rule definition above:
after redirect css service chgsvc1 uplink ip host 198.162.100.25 any fragment
The following deletes the first rule definition above:
no redirect css service chgsvc1 uplink ip host 198.162.100.25 any fragment
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] uplink source_address source_wildcard
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the filter are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Usage Guidelines Define a rule definition when any packet from the IP addresses which fall into the group of addresses matching
the IP address masking. This allows the reduction of filtering rule definitions as it does not require a rule
definition for each source and destination pair.
Examples The following command defines a rule definition to redirect packets to a charging service named chgsvc1:
redirect css service chgsvc1 uplink 1:1:1:1:1:1:1:1
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect css service svc_name [ log ] uplink { tcp | udp } { { source_address source_wildcard | any |
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after redirect css service svc_name [ log ] uplink { tcp | udp } { { source_address source_wildcard | any |
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
before redirect css service svc_name [ log ] uplink { tcp | udp } { { source_address source_wildcard | any
| source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
no redirect css service svc_name [ log ] uplink { tcp | udp } { { source_address source_wildcard | any |
source_host_address } [ eq source_port | gt source_port | lt source_port | neq source_port | range
start_source_port end_source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port | range start_dest_port end_dest_port ] }
after
Indicates all rule definitions subsequent to this command are to be inserted after the command identified by
the exact options listed.
This moves the insertion point to be immediately after the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, after the matching rule definition.
before
Indicates all rule definitions subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule definition which matches the exact options
specified such that new rule definitions will be added, in order, before the matching rule definition.
no
Removes the rule definition which exactly matches the options specified.
uplink
Apply this rule definition only to packets in the uplink (to the Mobile Node) direction.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
tcp | udp
Specifies the redirect is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Redirect applies to TPC packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
any
Specifies that the rule definition applies to all packets.
host
Specifies that the rule definition applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer value from 0 to 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Examples The following command defines a rule definition that redirects packets to the charging service named chgsvc1,
and UDP packets coming from any host are matched:
redirect css service chgsvc1 uplink udp any
The following sets the insertion point to before the rule definition above:
before redirect css service chgsvc1 uplink udp any
The following command sets the insertion point after the rule definition above:
after redirect css service chgsvc1 uplink udp any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] source_address
source_wildcard
after redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] source_address
source_wildcard
before redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] source_address
source_wildcard
no redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] source_address
source_wildcard
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an alpha
and/or numeric string from 1 to 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
Usage Guidelines Define a rule when any packet from the IP addresses which fall into the group of addresses matching the IP
address masking. This allows the reduction of redirect rules as it does not require a rule for each source and
destination pair.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the next hop host at 192.168.10.4, the context
with the context ID of 23 and the source IP and wildcard of 198.162.22.0 and 0.0.0.31:
redirect nexthop 192.168.10.4 context 23 198.162.22.0 0.0.0.31
The following sets the insertion point to before the first rule defined above:
before redirect nexthop 192.168.10.4 context 23 198.162.22.0 0.0.0.31
The following command sets the insertion point after the second rule defined above:
after redirect nexthop 192.168.10.4 context 23 198.162.22.0 0.0.0.31
The following deletes the first rule defined above:
no redirect nexthop 192.168.10.4 context 23 198.162.22.0 0.0.0.31
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] any
after redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] any
before redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] any
no redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] any
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an alpha
and/or numeric string from 1 to 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
any
Indicates all packets will match the redirect regardless of source and/or destination.
Usage Guidelines Define a catch all rule to place at the end of the list of rules to provide explicit handling of rules which do not
fit any other criteria.
Important It is suggested that any rule which is added to be a catch all should also have the log option specified. The
logged packets may be used to determine if the current list of rules is adequate or needs modification to
ensure proper security. The maximum number of rules that can be configured per ACL varies depending
on how the ACL is to be used. For more information, refer to the Engineering Rules appendix in the
System Administration Guide. Also note that "redirect" rules are ignored for ACLs applied to specific
subscribers or all subscribers facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the next hop host at 192.168.10.4, the context
with the context ID of 23 and any source IP:
redirect nexthop 192.168.10.4 context 23 any
The following sets the insertion point to before the first rule defined above:
before redirect nexthop 192.168.10.4 context 23 any
The following command sets the insertion point after the second rule defined above:
after redirect nexthop 192.168.10.4 context 23 any
The following deletes the first rule defined above:
no redirect nexthop 192.168.10.4 context 23 any
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] host
source_ip_address
after redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] host
source_ip_address
before redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] host
source_ip_address
no redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] host
source_ip_address
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an alpha
and/or numeric string from 1 to 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
Usage Guidelines Define a rule when a very specific remote host is to be blocked. In simplified networks where the access
controls need only block a few hosts, this command allows the rules to be very clear and concise.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the next hop host at 192.168.10.4, the context
with the context ID of 23 and a host IP address of 192.168.200.11:
redirect nexthop 192.168.10.4 context 23 host 192.168.200.11
The following sets the insertion point to before the first rule defined above:
before redirect nexthop 192.168.10.4 context 23 host 192.168.200.11
The following command sets the insertion point after the second rule defined above:
after redirect nexthop 192.168.10.4 context 23 host 192.168.200.11
The following deletes the first rule defined above:
no redirect nexthop 192.168.10.4 context 23 host 192.168.200.11
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] icmp { source_address
source_wildcard | any | host source_host_address } { dest_address dest_wildcard | any | host
dest_host_address } [ icmp_type [ icmp_code ] ]
after redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] icmp {
source_address source_wildcard | any | host source_host_address } { dest_address dest_wildcard | any |
host dest_host_address } [ icmp_type [ icmp_code ] ]
before redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] icmp {
source_address source_wildcard | any | host source_host_address } { dest_address dest_wildcard | any |
host dest_host_address } [ icmp_type [ icmp_code ] ]
no redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] icmp {
source_address source_wildcard | any | host source_host_address } { dest_address dest_wildcard | any |
host dest_host_address } [ icmp_type [ icmp_code ] ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an alpha
and/or numeric string from 1 to 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
icmp_type
Specifies that all ICMP packets of a particular type are to be filtered. The type can be an integer value from
0 through 255.
icmp_code
Specifies that all ICMP packets of a particular code are to be filtered. The type can be an integer value from
0 through 255.
Usage Guidelines Define a rule to block ICMP packets which can be used for address resolution and possible be a security risk.
The IP redirecting allows flexible controls for pairs of individual hosts or groups by IP masking which allows
the redirecting of entire subnets if necessary.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the next hop host at fe80::c0a8:a04, the context
with the context ID of 23, and ICMP packets coming from the host with the IP address 2002::c6a2:64195:
redirect nexthop fe80::c0a8:a04 context 23 icmp host 2002::c6a2:6419
The following sets the insertion point to before the first rule defined above:
before redirect nexthop fe80::c0a8:a04 context 23 icmp host 2002::c6a2:6419
The following command sets the insertion point after the second rule defined above:
after redirect nexthop fe80::c0a8:a04 context 23 icmp host 2002::c6a2:6419
The following deletes the first rule defined above:
no redirect nexthop fe80::c0a8:a04 context 23 icmp host 2002::c6a2:6419
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] ip { source_address
source_wildcard | any | host source_host_address } { dest_address dest_wildcard | any | host
dest_host_address } [ fragment ] [ protocol num ]
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to be immediately after the rule which matches the exact options specified
such that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an
alphanumeric string from 1 through 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the source_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
fragment
Indicates packet redirection is to be applied to IP packet fragments only.
protocol num
Indicates that the packet filtering is to be applied to a specific protocol number.
num can be an integer from 0 through 255.
Usage Guidelines Block IP packets when the source and destination are of interest.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IPv6 ACL Configuration
configure > context context_name > ipv6 access-list ipv6_acl_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ipv6-acl)#
Syntax Description redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] { tcp | udp } { {
source_address source_wildcard | any | host source_host_address } [ eq source_port | gt source_port | lt
source_port | neq source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq dest_port
| gt dest_port | lt dest_port | neq dest_port ] }
after redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] { tcp | udp }
{ { source_address source_wildcard | any | host source_host_address } [ eq source_port | gt source_port |
lt source_port | neq source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port ] }
before redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] { tcp | udp
} { { source_address source_wildcard | any | host source_host_address } [ eq source_port | gt source_port
| lt source_port | neq source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq
dest_port | gt dest_port | lt dest_port | neq dest_port ] }
no redirect nexthop nexthop_addr { context context_id | interface interface_name } [ log ] { tcp | udp } {
{ source_address source_wildcard | any | host source_host_address } [ eq source_port | gt source_port | lt
source_port | neq source_port ] } { { dest_address dest_wildcard | any | host dest_host_address } [ eq dest_port
| gt dest_port | lt dest_port | neq dest_port ] }
after
Indicates all rules defined subsequent to this command are to be inserted after the command identified by the
exact options listed.
This moves the insertion point to immediately after the rule which matches the exact options specified such
that new rules will be added, in order, after the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
before
Indicates all rules defined subsequent to this command are to be inserted before the command identified by
the exact options listed.
This moves the insertion point to be immediately before the rule which matches the exact options specified
such that new rules will be added, in order, before the matching rule.
Important If the options specified do not exactly match an existing rule, the insertion point does not change.
no
Removes the rule which exactly matches the options specified.
nexthop nexthop_addr
The IP address to which the IP packets are redirected.
context context_id
The context identification number of the context to which packets are redirected. At the executive mode
prompt, use the show context all command to display context names and context IDs.
interface interface_name
The name of the logical interface to which the packets should be redirected. interface_name must be an
alphanumeric string from 1 through 79 characters.
log
Default: packets are not logged.
Indicates all packets which match the redirect are to be logged.
tcp | udp
Specifies the redirect is to be applied to IP-based transmission control protocol or the user datagram protocol.
• tcp: Redirect applies to TCP packets.
• udp: Redirect applies to UDP packets.
source_address
The IP address(es) form which the packet originated.
This option is used to filter all packets from a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this option. The range can then
be configured using the source_wildcard parameter.
source_wildcard
This option is used in conjunction with the source_address option to specify a group of addresses for which
packets are to be filtered.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB).
any
Specifies that the rule applies to all packets.
host
Specifies that the rule applies to a specific host as determined by its IP address.
source_host_address
The IP address of the source host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
dest_host_address
The IP address of the destination host to filter against expressed in IPv6 colon-separated-hexadecimal notation.
eq source_port
Specifies a single, specific source TCP port number to be filtered.
source_port must be configured to an integer value from 0 to 65535.
gt source_port
Specifies that all source TCP port numbers greater than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
lt source_port
Specifies that all source TCP port numbers less than the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
neq source_port
Specifies that all source TCP port numbers not equal to the one specified are to be filtered.
source_port must be configured to an integer value from 0 to 65535.
dest_address
The IP address(es) to which the packet is to be sent.
This option is used to filter all packets to a specific IP address or a group of IP addresses.
When specifying a group of addresses, the initial address is configured using this parameter. The range can
then be configured using the dest_wildcard parameter.
dest_wildcard
This option is used in conjunction with the dest_address option to specify a group of addresses for which
packets are to be filtered.
The mask must be entered as a complement:
• Zero-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be identical.
• One-bits in this parameter mean that the corresponding bits configured for the dest_address parameter
must be ignored.
Important The mask must contain a contiguous set of one-bits from the least significant bit (LSB). Therefore, allowed
masks are 0, 1, 3, 7, 15, 31, 63, 127, and 255. For example, acceptable wildcards are 0.0.0.3, 0.0.0.255,
and 0.0.15.255. A wildcard of 0.0.7.15 is not acceptable since the one-bits are not contiguous.
eq dest_port
Specifies a single, specific destination TCP port number to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
gt dest_port
Specifies that all destination TCP port numbers greater than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
lt dest_port
Specifies that all destination TCP port numbers less than the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
neq dest_port
Specifies that all destination TCP port numbers not equal to the one specified are to be filtered.
dest_port must be configured to an integer value from 0 to 65535.
Usage Guidelines Block IP packets when the source and destination are of interest but for only a limited set of ports.
Important The maximum number of rules that can be configured per ACL varies depending on how the ACL is to
be used. For more information, refer to the Engineering Rules appendix in the System Administration
Guide. Also note that "redirect" rules are ignored for ACLs applied to specific subscribers or all subscribers
facilitated by a specific context.
Examples The following command defines a rule that redirects packets to the next hop host at fe80::c0a8:a04, the context
with the context ID of 23, and UDP packets coming from any host are matched:
redirect nexthop fe80::c0a8:a04 context 23 udp any
The following sets the insertion point to before the rule defined above:
before redirect nexthop fe80::c0a8:a04 context 23 udp any
The following command sets the insertion point after the rule defined above:
after redirect nexthop fe80::c0a8:a04 context 23 udp any
The following command deletes the first rule defined above:
no redirect nexthop fe80::c0a8:a04 context 23 udp any
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4
Tunnel Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
destination address
Configures the destination of the tunnelled packets for a manual tunnel.
Product All
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4 Tunnel
Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
no
Removes configuration for the specified keyword.
address
Specifies the IP address of the destination device. address must be specified in IPv4 dotted decimal or IPv6
colon-separated-hexadecimal notation.
Usage Guidelines Use this command to configure the IP address of the destination end of the tunnel.
Examples The following command sets the destination address for packets on this tunnelled interface to 10.2.3.4:
destination address 10.2.3.4
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
mode
Configures the mode of IPv6 to IPv4 tunneling. The default is set to manual mode.
Product All
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4 Tunnel
Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
6to4
Configures automatic IPv6-to-IPv4 (6to4) tunnels as specified in RFC 3056.
manual
Configures point-to-point manual IPv6-to-IPv4 tunnels by specifying the IPv4 address of the tunnel remote
end.
default
Resets the mode of IPv6-to-IPv4 tunneling to manual mode.
Usage Guidelines There can be only one IPv6-to-IPv4 tunnel possible in a context. Once an IPv6-to-IPv4 tunnel is configured,
all subsequent tunnels will be configured as manual tunnels.
source
Configures the source of tunneled packets.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4 Tunnel
Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
address ip_address
Specifies the IPv4 address to use as the source address of the tunnel.
ip_address must be expressed in IPv4 dotted-decimal notation.
interface interface_name
Specifies the name of a non-tunnel IPv4 interface, whose address is used as the source address of the tunnel.
interface must be an alphanumeric string of 1 through 79 characters.
Usage Guidelines Configures the source IPv4 address of the tunnel by either specifying the IP address (host address) or by
specifying another configured non-tunnel IPv4 interface. The source address must be an existing interface
address before it is used. State of source address will affect the operational state of the tunnel.
Examples The following command configures the source address of the tunnel.
source address 10.2.3.4
tos
Configures the type of service (TOS) settings of the outer IPv4 header of the tunneled packets.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4 Tunnel
Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
copy
Copies the DC octet of the IPv6 packet to the TOS octet of IPv4 packet.
default
Configures default setting for the specified keyword.
value tos_value
Configures the raw TOS value ranging from 0 to 255. The default is 0.
Usage Guidelines Sets the TOS parameter to be used in the tunnel transport protocol or copies the TOS value from the original
IPv6 DC byte to the TOS value of the encapsulating IPv4 header.
ttl
Configures the TTL (Time to live) value of the outer IPv4 header of the tunneled packets.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > Tunnel Interface Configuration > IPv6 to IPv4 Tunnel
Interface Configuration
configure > context context_name > interface interface_name tunnel > tunnel-mode ipv6ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-tunnel-ipv6ip)#
default
Configures default setting for the specified keyword.
value ttl_value
ttl_value is an integer from 1 through 255. The default is 16.
Usage Guidelines Configures the TTL parameter to be used in the tunnel transport protocol.
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate l2-mapping-table
Associates a global QoS-to-Level 2 mapping table with this VRF.
Product ePDG
HSGW
P-GW
SAEGW
S-GW
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
no
Disassociates an existing L2 mapping table from this VRF.
name table_name
Specifies the name of an existing internal table from which to map QoS to L2 values.
table_name is an alphanumeric string of 0 through 80 characters.
system-default
Associates the system-default table with this VRF. This is useful if the base-context has a different explicit
mapping.
Usage Guidelines Use this command to associates a global QoS-to-Level 2 mapping table with this VRF.
Internal-QoS will be mapped to an actual L2 value (either or both of 802.1p/MPLS) using a per-VRF based
table.
Important If an l2-mapping-table association is made at both the VRF and VPN level, the VRF level takes precedence.
The mapping table is configured via the Global Configuration mode qos l2-mapping-table command.
Examples The following command associates this VRF with Qos-to-L2 mapping table vrf10.
associate l2-mapping-table name vrf10
description
Allows you to enter descriptive text for this configuration.
Product All
no
Clears the description for this configuration.
text
Enter descriptive text as an alphanumeric string of 1 to 100 characters.
If you include spaces between words in the description, you must enclose the text within double quotation
marks (" "), for example, "AAA BBBB".
Usage Guidelines The description should provide useful information about this configuration.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip aggregate-address
Specifies an IPv4 address/mask for aggregating frame routes in the VRF.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
no
Deletes the specified IPv4 aggregate address.
ipv4_address/mask
Specifies the IP address and mask in IPv4 CIDR dotted-decimal notation.
[summary-only]
When this option is configured, the constituent routes are removed from the VRF.
Usage Guidelines Use this command to configure aggregate framed-routes in a VRF. It enables inserting an aggregate-address
in a VRF and its advertisement in the routing domain if at least one constituent framed-route exists in that
VRF. By default, the constituent routes will also be present along with the aggregate address. However, if the
summary-only option is configured, the constituent routes will be removed from the VRF. Up to 32 aggregate
addresses can be configured in a VRF.
Examples The following example sets an IPv4 aggregate address for the VFR:ip aggregate-address 196.168.34.55/24
ip guarantee
Enables and disables local switching of framed route packets.
Product GGSN
P-GW
SAEGW
Privilege Administrator
no
Disables local switching of framed route packets.
framed-route local-switching
Enables local switching of framed route packets. By default, this functionality is disabled.
Usage Guidelines Use this command to enable and disable local switching of framed route packets. This functionality will be
applicable only when there are some NEMO/framed route sessions in a context.
Examples The following command enables local switching of framed route packets:
ip guarantee framed-route local-switching
ip maximum-routes
This command configures the maximum number of routes in an IP VRF routing table configured in this
context.
Important This command should only be used for framed or NEMO (Network Mobility) routes of the VRF.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
no
Disables the configured maximum routes in specific IP VRF context.
max_routes
Sets the maximum number of routes in a specific IP VRF context.
max_routes must be an integer from 1 through 32768 or 65536 (release 17.0+).
Usage Guidelines Use this command to configure the maximum number of routes in a particular VRF routing table. When the
number of routes in the VRF is more than the maximum limit configured, a critical log is generated indicating
that the number of routes is over the limit. Once the number of routes in the VRF goes under the limit, a clear
log is generated.
The maximum routes configured using this command will be sent to the threshold configuration logic for
appropriate action. For more information on threshold configuration, refer to descriptions of the threshold
route-service and threshold poll route-service interval commands in the Global Configuration Mode
Commands chapter.
Examples The following command sets 1000 routes as a maximum limit for specific VRF context:
ip maximum-routes 1000
mpls map-dscp-to-exp
This command maps the final differentiated services code point (DSCP) bit value in the IP packet header to
the final Experimental (EXP) bit value in the MPLS header for incoming traffic.
Important This command has been deprecated beginning with Release 15.0.
Product GGSN
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
no
Disables the configured DSCP bit value mapping to the EXP bit value from a specific IP VRF context.
dscp dscp_bit_value
Specifies the final DSCP bit value which is to map with the final EXP bit value in MPLS header for incoming
traffic.
dscp_bit_value specifies the value of DSCP bit values separated in eight groups and represented with integers
from 0 through 7.
The default representation of DSCP value in eight groups is given in the following table:
DSCP Marking Value DSCP Map Group
0-7 0
8-15 1
16-23 2
24-31 3
exp exp_bit_value
Specifies the final EXP bit value in MPLS header to which the final DSCP bit value 0 to 7 (represented in
eight values) coming from incoming traffic will be mapped.
exp_bit_value is the value of EXP bit in MPLS header and must be an integer between 0 through 7.
Usage Guidelines Use this command to map the final DSCP value coming from incoming IP traffic to a final EXP value in
MPLS header. This mapping determines the QoS and service parameters to which the packet is assigned.
Examples The following command maps the DSCP value 3 (24 to 31) to EXP bit 3 in MPLS header:
mpls map-dscp-to-exp dscp 3 exp 3
mpls map-exp-to-dscp
Maps incoming the Experimental (EXP) bit value in MPLS header to the internal differentiated services code
point (DSCP) bit value in IP packet headers for outgoing traffic.
Important This command has been deprecated beginning with Release 15.0.
Product GGSN
Command Modes Exec > Global Configuration > Context Configuration > IP VRF Context Configuration
configure > context context_name > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-context-vrf)#
no
Disables the configured EXP bit value mapping to DSCP bit value from specific IP VRF context.
exp exp_bit_value
Specifies the incoming EXP bit value in MPLS header to which the internal DSCP bit value 0 to 7 (represented
in 8 values) in IP traffic will be mapped.
exp_bit_value is the value of the EXP bit in an MPLS header and must be an integer from 0 through 7.
dscp dscp_bit_value
Maps the DSCP bit value with the incoming EXP bit value in an MPLS header.
dscp_bit_value specifies the value of the DSCP bit values separated in eight groups and represented with
integers between 0 through 7.
The default representation of DSCP value in eight groups is given in the following table:
DSCP Marking Value DSCP Map Group
0-7 0
Usage Guidelines Use this command to map the incoming EXP bit value in MPLS headers to the DSCP bit value in IP traffic.
This mapping determines the QoS and service parameters to which the packet is assigned.
Examples The following command maps the EXP bit value 4 to DSCP value 6 (48 to 55) in IP header:
mpls map-exp-to-dscp exp 4 dscp 6
The ISAKMP Configuration Mode is used to configure Internet Security Association Key Management
Protocol (ISAKMP) policies that are used to define Internet Key Exchange (IKE) security associations (SAs).
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
authentication
Configures the ISAKMP policy authentication mode.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
default authentication
Restores the default setting of this parameter. This command is enabled by default.
no authentication
Disables the preshared key authentication mode.
preshared-key
Specifies that the policy will be authenticated through the use of the pre-shared key.
Usage Guidelines When the system is configured to use ISAKMP-type crypto maps for establishing IPSec tunnels, this command
is used to indicate that the policy will be authenticated through the use of the pre-shared key configured in
the ISAKMP crypto map.
Examples The following command sets policy authentication mode to use a pre-shared key:
authentication preshared-key
encryption
Configures the encryption protocol to use to protect subsequent IKE SA negotiations.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
default encryption
Restores the default setting of this parameter.
no encryption
Removes a previously configured encryption type.
3des-cbc
Specifies that the encryption protocol is Triple Data Encryption Standard (3DES) in chain block (CBC) mode.
aes-cbc-128
Specifies that the encryption protocol is Advanced Encryption Standard (AES) in CBC mode with a 128-bit
key.
aes-cbc-256
Specifies that the encryption protocol is Advanced Encryption Standard (AES) in CBC mode with a 256-bit
key.
des-cbc
Specifies that the encryption protocol is DES in CBC mode. This is the default setting.
Usage Guidelines Once the D-H exchange between the system and the security gateway has been successfully completed,
subsequent IKE SA negotiations will be protected using the protocol specified by this command.
Examples The following command sets the IKE encryption method to 3des-cbc:
encryption 3des-cbc
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
group
Configures the Oakely group (also known as the Diffie-Hellman [D-H] group) in which the D-H exchange
occurs.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
default group
Restores the default setting of this parameter.
no group
Removes a previously configured group.
{1|2|5}
Default: 1
Specifies the number of the Oakley group. The following groups are allowed:
• 1: Enables Oakley Group 1 using a 768-bit modp as defined in RFC 2409.
• 2: Enables Oakley Group 2, using a 1024-bit modp as defined in RFC 2409.
• 5: Enables Oakley Group 5, using a 1536-bit modp as defined in RFC 3526.
Usage Guidelines Specifies the Oakley group that determine the length of the base prime numbers that are used during the key
exchange process.
Examples The following command sets the group to 5 which specifies 1536-bit base prime numbers:
group 5
hash
Configures the IKE hash protocol to use during IKE SA negotiations.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
default
Restores the default setting of this parameter.
no
Removes a previously configured hash algorithm.
md5
Specifies that the hash protocol is Message Digest 5 truncated to 96 bits.
sha1
Specifies that the hash protocol is Secure Hash Algorithm-1 truncated to 96 bits. This is the default setting
for this command.
Usage Guidelines Use this command to configure the hash algorithm used during key negotiation.
Examples Set the hash algorithm to Message-Digest 5 by entering the following command:
hash md5
lifetime
Configures the lifetime of the IKE Security Association (SA).
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > ISAKMP Configuration
configure > context context_name > isakmp policy policy_number
default lifetime
Restores the default setting of this parameter.
seconds
Default: 86400
The number of seconds for the SA to live. seconds must be an integer from 60 to 86400.
Usage Guidelines Use this command to set the time that an ISAKMP SA will be valid. The lifetime is negotiated with the peer
and the lowest configured lifetime duration is used.
The IuPS Service configuration mode is used to define properties for the IuPS service which controls the
Iu-PS interface connections to Radio Network Controllers (RNCs) of the UMTS Terrestrial Radio Access
Network (UTRAN).
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
Note From R15.0 onwards, License Control is implemented on all Network Sharing related commands.
access-protocol
This command configures the access protocol parameters for the IuPS service.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Removes a previously configured access protocol value.
sccp-network sccp_net_id
Specifies the Signaling Connection Control Part (SCCP) for this IuPS service to use.
sccp_net_id must be an integer from 1 to 16.
Usage Guidelines Use this command to configure access protocol parameters for the current IuPS service.
Examples The following command specifies that the current Iu-PS service should use SCCP 1:
access-protocol sccp-network 1
associate
This command associates a configured DSCP marking template with this IuPS service and associated Iu
interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Removes a previously configured association.
dscp_template_name
Specifies a DSCP marking template that was previously configured with the commands in the DSCP Template
configuration mode.
dscp_template_name- Enter an alphanumeric string of 1 to 64 characters, including dots (.), dashes (-), and
forward slashes (/), to identify a unique instance of a DSCP marking template.
Usage Guidelines Use this command to associate a specific DSCP marking template with this IuPS service and associated Iu
interface. The DSCP template provides a mechanism for differentiated services code point (DSCP) marking
of control packets and signaling messages at the SGSN's M3UA level on the Iu interface. This DSCP marking
enables the SGSN to perform classifying and managing of network traffic and to determine quality of service
(QoS) for the interface to the IP network.
Examples The following command associates a DSCP marking template named dscptemp1 with the Iu interface:
associate dscp-template downlink dscptemp1
The following command disassociates a previously associated DSCP marking template named template4 with
this IuPS service configuration:
no associate dscp-template downlink
blacklist-timeout-gtpu-bind-addresses
This command specifies the time period that a GTP-U bind address (loopback address) will not be used (is
blacklisted) in RAB-Assignment requests after a RAB assignment request, with that GTP-U bind address,
has been rejected by an RNC with the cause - Unspecified Error. This is a failure at the RNC's GTP-U IP
interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the blacklist time to 60 seconds.
seconds
Number of seconds that the GTP-U bind (loopback) address will not be used in a RAB-Assignment request.
seconds : Must be an integer from 1 to 1800.
Examples The following command specifies a 15 minutes (460 seconds) blacklist period.
blacklist-timeout-gtpu-bind-addresses 460
empty-cr
This command allows the operator to determine how empty Connection Request messages will be handled.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default | no
Using either default or no with the command disables the rejection function and returns the system to the
default behavior, which is to ignore receipt of the empty CRs.
Usage Guidelines Use this command to enable/disable the procedure for handling empty (not containing dataparameters)
Connection Request (CR) messages.
This feature can be used in the following scenario: During 4G to 3G handovers, some Connection Requests
from mobile subscribers might be ignored by the SGSN, even though their UE would display that the WCDMA
was available. The RNC would send an SCCP Connection Request (CR) over the Iu interface to the SGSN.
Normally, this message contains a RANAP message and GMM, but according to 3GPP and ITU Q.713
standards, it is permissible to send an SCCP CR without any data parameters. In such a situation, normally
the SGSN would ignore these SCCP CR messages, because without these data parameters the SGSN would
be unable to derive the DeMux key which is the basis for determining the Session Manager instance to be
used for a subscriber. Using this feature allows the SGSN to send a Reject to the mobile subscriber when an
"empty" SCCP CR is sent from their UE.
Fields have been added to the output of the following CLI show commands to track the receipt and rejection
of Connect Request (CR) messages:
• show gmm-sm statistics
• show gmm-sm statistics verbose
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the previous configuration mode, the context configuration
mode.
Product SGSN
force-authenticate consecutive-security-failure
Disable/enable authentication when the MS/UE security fails and configures the procedures and frequency
for authentication
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the values to defaults. Forced authentication is enabled for all the types of event procedures with the
default values for determining frequency for authentication.
no
Disables the specified authentication configuration.
inter-sgsn-rau
Default: enabled
Enables/disables authentication for inter-SGSN RAU.
The SGSN does not remember previous inter-SGSN-RAU failures for a P-TMSI/RAI because the SGSN
clears all contexts on the occurrence of an inter-SGSN-RAU security failure. So the next inter-SGSN-RAU
can only be authenticated forcefully if it comes before the previous context is cleared. This type of forced
authentication is enabled by default because this type of failure is fairly common.
Enables/ disables authentication for local messages (such as local RAUs, Service Requests, Detach Requests,
etc) . Consecutive security failures is fairly rare for local messages so the default count frequency is fairly
high, 5. Setting the count frequency enables the feature and sets the number of consecurity local message
security failures that must occur prior t o authentication being forced.
frequency: Enter an integer from 1 to 10.
Usage Guidelines GMM authentication is optional for UMTS. When GMM authentication is skipped, the SGSN and the MS
continue to re-use the latest keys exchanged during the most recent GMM authentication procedure. This can
result in the SGSN and the MS going out of sync with the CK and IK currently in use. If a mismatch occurs
when the MS continues to use the correct parameters (e.g., cksn or P-TMSI signature) in the next Iu and if
the SGSN skips authentication on the Iu, then, usually, the security mode will timeout or be rejected because
the MS will not be able to decipher or perform an integrity check on the network messages. This scenario
results in a lot of useless signaling in the network. This command allows the operator to enable a forced GMM
authentication that will either resolve this type of problem or avoid it. As well, the operator can configure a
frequency of authentication that best meets their needs.
Examples The following command enables forced authentication after every 3rd local message security failure:
force-authenticate consecutive-security-failure local-messages count 3
gtpu
This commands configures parameters for the GTP user (GTP-U) dataplane.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
Syntax Description gtpu { bind ip_addr | echo-interval seconds | max-retransmissions number | retransmission-timeout
seconds | sync-echo-with-peer }
no gtpu { bind address ip_addr | echo-interval | max-retransmissions | retransmission-timeout |
sync-echo-with-peer }
default gtpu { echo-interval | max-retransmissions | retransmission-timeout | sync-echo-with-peer }
no
Removes the configured parameter value.
default
Sets the specified parameter to its default setting.
echo-interval seconds
Default: 60
Configures the rate, in seconds, at which GTP-U echo packets are sent to the UTRAN over the Iu-PS interface.
seconds : Must be an integer from 60 through 3600.
max-retransmissions number
Default: 5
Configures the maximum number of transmission retries for GTP-U packets.
number : Must be an integer from 0 through 15.
retransmission-timeout seconds
Default: 5
Configures the retransmission timeout for GTPU packets in seconds.
seconds : Must be an integer from 1 through 20.
sync-echo-with-peer
This keyword is applicable to the SGSN only.
This keyword enables the SGSN to synchronize path management procedures with the peer after a GTP service
restart recovery.
After GTP service recovery, the SGSN restarts the timers for GTP echo transmission, hence a drift in echo
request transmission time (from the pre-recovery time) can occur causing the SGSN to be out of sync with
the peer. By using this keyword, when the SGSN receives the first Echo Request (GTPC or GTPU) from the
peer after the GTP service restart, in addition to replying with an ECHO Response, the SGSN transmits an
ECHO Request to the peer and the SGSN restarts the timers associated with the path management procedures.
This causes the path management procedure at SGSN to synchronize with the peer node.
Default: Enabled
Usage Guidelines Use this command to configure GTP-U parameters for the Iu-PS interface.
Examples The following command binds the IP address 192.168.0.10 to the Iu-PS interface for communication with the
UTRAN:
gtpu bind address 192.168.0.10
inter-rnc-procedures
This command enables the processing of SRNS relocation when the source RNC is behaving as the target
RNC
Product SGSN
Insert product and tag this paragraph appropriately.
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Disables SRNS relocation when the source RNC is behaving as the target RNC. This is the default behavior.
source-rnc-as-target-rnc
Configures the SGSN to complete SRNS relocation when the source RNC is behaving as the target RNC. For
example, in the case of a Femtocell-to-Femtocell handoff - the femtocell gateway may act both as the source
and target RNC to the femtocells, although from the SGSN's perspective it is the same RNC.
use-old-location-info
Selects and uses the old values of LAC, RAC and SAC for S-CDRs and ULI information sent to the GGSN
during an intra-SRNS procedure.
Usage Guidelines Use this command to enable/disable SRNS relocation when the source RNC is behaving as the target RNC.
Examples Enter this command to enable SRNS relocation for those scenarios where the source RNC is behaving as the
target RNC.
inter-rnc-procedures source-rnc-as-target-rnc
iu-hold-connection
Defines the type and duration of the Iu hold connection.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the Iu hold connection parameters to requested-by-ms and 100 second duration.
no
Removes the configuration information for the specified Iu hold connection parameter.
always
Specifies that there is always to be an Iu hold connection procedure.
requested-by-ms
Specifies that there is only an Iu hold connection procedure if requested by the MS/UE.
This is the default setting for Iu-hold-connection.
hold-time time
This variable configures the interval (in seconds) that the SGSN holds the Iu connection.
time: must be an integer from 1 to 3600.
time: must be an integer from 10 to 3600.
Important It is recommended to use a minimum value of "10" seconds. If a value less than "10" seconds is used,
more collisions may be observed. If the minimum value of "1" is set, after a re-load, INTRA-RAU (with
unknown ptmsi, old-rai known) will be released in "1" second if the Identity Rsp does not come within
"1" second.
Default is 100.
Usage Guidelines Define the amount of time the Iu connection will be held open.
Examples Instruct the SGSN to hold the Iu connection open for 120 seconds
iu-hold-connection always hold-time 120
iu-recovery
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
iu-release-complete-timeout
Configures the SGSN's timer for waiting for an Iu Release Complete message from the RNC.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer to its default setting.
time
This variable defines the amount of time (in seconds) that the SGSN waits to receive an 'Iu Release Complete'
message from the RNC.
Default: 10.
time: Must be an integer from 1 to 60.
Usage Guidelines Configure the number of seconds that the SGSN waits to receive the Iu Release Complete message.
loss-of-radio-coverage ranap-cause
This command sets the detection cause included in the Iu Release message. This command is unique to releases
9.0 and higher.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
This keyword resets the configuration to the default cause ID number.
ranap-cause cause_number
This number identifies the reason the SGSN has detected, from Iu Release messages, for the loss of radio
coverage (LORC). This value is included in the IE messages the SGSN sends to either the GGSN or the GGSN
and the peer SGSN to indicate LORC state. The range of reasons is a part of the set defined by 3GPP 25413.
cause_number : Must be an integer from 1 to 512.
Default: 46 (MS/UE radio connection lost)
Usage Guidelines By defining a cause code, the SGSN knows to detect the LORC state of the mobile from Iu Release messages
it receives for the subscriber. This configuration also instructs the SGSN to include the defined cause code
for the LORC state in the IE portion of various messages sent to the GGSN and optionally the peer SGSN.
This command is one of the two commands required to enable the SGSN to work with the GGSN and,
optionally the peer SGSN, to implement the Overcharging Protection feature (see the SGSN Overview in the
SGSN Administration Guide for feature details. The other command needed to implement the Overcharging
Protection feature is the gtp private extension command explained in the SGSN APN Policy Configuration
Mode chapter of the Command Line Interface Reference.
Examples Use the following command to set the cause code to indicate that there are no radio resources available in the
target cell, cause 53.
loss-or-radio-coverage ranap-cause 53
mbms
This command is in development for future use so the command and keywords that you might see are not
currently supported.
network-sharing cs-ps-coordination
Enables/disables the SGSN service to perform a CS-PS coordination check.
Important With the release of 15.0, both 2G and 3G MOCN functionality is license controlled and the license is
required to use all previously available network sharing SGSN configuration commands. For additional
information, contact your Cisco Account Representative.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Including this keyword resets the SGSN service to allow the check to be performed.
no
Disables this CS-PS coordination checking for this IuPS service.
homer
Enables checking for CS-PS co-ordination for homers (UEs registered in the home network) only.
roamer
Enables checking for CS-PS co-ordination for roamers (UEs from outside the home network) only.
Usage Guidelines Use this command to facilitate the network sharing functionality. With this command, the SGSN can be
instructed to perform a check to determine if CS-PS coordination is needed.
3GPP TS 25.231 section 4.2.5 describes the functionality of the SGSN to handle CS-PS
(circuit-switching/packet-switching) coordination for attached networks not having a Gs-interface. In compliance
with the standard, the SGSN rejects an Attach in a MOCN configuration with cause 'CS-PS coordination
required', after learning the IMSI, to facilitate the RNC choosing the same operator for both CS and PS
domains.
Examples Use the following syntax to disable the CS-PS coordination check:
no network-sharing cs-ps-coordination
Use the following command to enable the CS-PS coordination check only for UEs from outside the home
network:
no network-sharing cs-ps-coordination roamer
network-sharing failure-code
Configure the reject cause code to included in network-sharing Reject messages.
Important With the release of 15.0, both 2G and 3G MOCN functionality is now license controlled and the license
is required to use all previously available network sharing SGSN configuration commands. For additional
information, contact your Cisco Account Representative.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the SGSN service to use the default cause code,14 (GPRS services not allowed in this PLMN).
failure_code
Enter one of the GMM failure cause codes listed below (from section 10.5.5.14 of the 3GPP TS 124.008
v7.2.0 R7):
• 2 - IMSI unknown in HLR
• 3 - Illegal MS
• 6 - Illegal ME
• 7 - GPRS services not allowed
• 8 - GPRS services and non-GPRS services not allowed
• 9 - MSID cannot be derived by the network
• 10 - Implicitly detached
Usage Guidelines Use this command to determine which failure code will be included in Reject messages sent by the SGSN
when there is a network-sharing failure.
Examples Use the following syntax to indicate that roaming is not allowed (#13) as the cause for the network-sharing
failure:
network-sharing failure 13
network-sharing non-shared
This command allows non-shared area access when network-sharing is enabled.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the default to disable non-shared access.
Usage Guidelines When non-shared area access is enabled, the SGSN sends the selected-plmn value in Attach/RAU accept if
LAI is having one of the selected-plmn and "selected-plmn" or "Redirect-attempt flag" IEs are not included
in the request message.
network-sharing stop-redirect-reject-cause
Enables the operator to disable the default behavior which sends Redirection Indication IE in RANAP Reject
messages when reject is due to GMM cause #17 (network failure) related to System Failure or Unexpected
Data value MAP errors from the HLR. This change of the default behavior would only be applicable to 3G
Roamers.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Instructs the SGSN to use the default behavior and send redirect indication in Attach Reject or RAU Reject
if reject is due to GMM cause 'network failure' which resulted from one of the MAP errors unexpected data
value or system failure.
no
Disables this function and returns to the default behavior.
Usage Guidelines With this command, the operator would change the SGSN's default behavior (complies with 3GPP Release
11) for roaming subscribers and send Redirection Complete IE in Attach and RAU Reject messages when the
reject is due to GMM cause #17 (network failure) in response to receiving System Failure or Unexpected
Data value MAP errors from the HLR
Examples Configure the SGSN to send Redirect Indication IE in RANAP reject messages:
default network-sharing stop-redirect-reject-cause
plmn
Configures the PLMN (public land mobile network) related parameters for the IuPS service. This command
is appicable to releases 8.1 and higher.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
Syntax Description plmn id mcc mcc_num mnc mnc_num [ network-sharing common-plmn mcc mcc_num mnc mnc_num [
plmn-list mcc mcc_num mnc mnc_num [ mcc mcc_num mnc mnc_num+ ] ] ]
no plmn id
no
Removes the PLMN ID from the configuration.
id
Creates a PLMN configuration instance based on the PLMN ID (comprised of the MCC and MNC). In
accordance with TS 25.413, the SGSN supports up to 32 PLMN configurations for shared networks.
mcc mcc_num
Specifies the mobile country code (MCC) portion of the PLMN's identifier.
mcc_num: The PLMN MCC identifier and can be configured to any integer value between 100 and 999.
mnc mnc_num
Specifies the mobile network code (MNC) portion of the PLMN's identifier.
mnc_num: The PLMN MNC identifier and can be configured to any 2-digit or 3-digit value between 00 and
999.
Important With the release of 15.0, both 2G and 3G MOCN functionality is now license controlled and the license
is required to use all previously available network sharing SGSN configuration commands. For additional
information, contact your Cisco Account Representative.
Usage Guidelines Use this command to configure the PLMN associated with the SGSN. There can only be one PLMN associated
with an SGSN unless one of the following features is enabled and configured: network sharing or multiple
PLMN.
For network sharing, use of the network-sharing keywords make it possible to identify more than one PLMN.
Including the PLMN identified initially. None have precedence. They are all treated equally but they must
each be unique. In a MOCN configuration, the PLMN list will not be used as there would only be one local
PLMN.
For multiple PLMN support, the SGSN can support up to 8 Iu-PS configurations for PLMNs. These Iu-PS
service configurations must be associated with the SGSN via the ran-protocol command in the SGSN Service
configuration mode.
Examples Use the following command to identify a PLMN by the MCC 313 and MNC 23 and instruct the SGSN to
perform network sharing with a single common PLMN identified by MCC 404 and MNC 123:
plmn id mcc 313 mnc 23 network-sharing common-plmn mcc 404 mnc 123
rab-assignment-response-timeout
Configures the RAB assignment timer.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer to its default setting.
time
This variable configures the amount of time (in seconds) that the SGSN waits to receive a RAB assignment
from the RNC.
time: must be an integer from 1 to 60.
Default: 8.
Usage Guidelines This command defines time the SGSN waits for the completion of the RAB assignment procedure.
radio-network-controller
This command creates an instance of an RNC configuration to associate with the IuPS service for the SGSN.
This command is only available in release 8.0; use the rnc command for releases 8.1 and higher.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Removes the configuration information for the specified RNC.
id rnc_id
Define the instance number of the RNC configuration.
rnc_id : Must be an integer from 0 to 4095.
mcc mcc_num
Specifies the mobile country code (MCC).
mcc_num : Must be an integer between 100 and 999.
mnc mnc_num
Specifies the mobile network code (MNC).
mnc_num : Must be an integer between 00 and 999.
Usage Guidelines Use this command to configure information for the IuPS service to use to contact specific RNCs.
This command also provides access to the RNC configuration mode.
Examples The following command creates or accesses RNC configuration instance #1 with MCC of 131 and MNC of
22:
radio-network-controller id 1 mcc 131 mnc 22
rai-skip-validation
Enable or disable if validation checks are done to verify the MCC and MNC fields received in the old RAI
IE in Attach/RAU Requests.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Disables skipping the validation of the old RAI MCC/MNC fields and enables the default behavior to validate.
Usage Guidelines This command configures the SGSN to enable or disable rejection of RAU requests with invalid MCC/MNC
values in the old RAI field. By default, this configuration is disabled allowing the default behavior to validate
the old RAI MCC/MNC fields.
This command also impacts the PTMSI attaches where the old RAI field is invalid. If the OLD RAI field is
invalid and if the validation is enabled, the identity of the MS is requested directly from the MS instead of
the peer SGSN.
Validation checks are done per 3GPP TS 24.008 for the MCC/MNC fields of the old RAI IE in Attach/RAU
Requests. RAU requests with invalid MCC/MNC values in the old RAI field are rejected. For Attach requests
with invalid MCC/MNC values in the old RAI field, the identity of the MS is retrieved directly from the MS
instead of sending an identity request to the peer Node where the MS identity is derived from the valid old-RAI.
Examples Use this command to configure rejection of RAU requests with invalid MCC/MNC values in the old RAI
field:
no rai-skip-validation
relocation-alloc-timeout
This command defines the amount of time the SGSN waits for a Relocation Request message.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the configuration to a 5 second wait time.
timeout_value
Time in seconds that the SGSN waits to receive a Relocation Request message.
timeout_value : Must be an integer from 1 to 60.
Default : 5 seconds.
Usage Guidelines Use this command to configure the number of seconds the SGSN will wait for a Relocation Request message
to be received. This timeout needs to be set with sufficient time so that SRNS procedure aborts can be avoided
if the peer fails to respond in a timely fashion in the case of a hard handoff.
relocation-complete-timeout
This command specifies the maximum time for the SGSN to wait for a Relocation Completion from the core
network.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the configuration to a 5 second wait time.
timeout_value
Time in seconds that the SGSN waits for relocation to be completed.
timeout_value : Must be an integer from 1 to 60.
Default : 5 seconds.
Usage Guidelines Use this command to configure the number of seconds the SGSN will wait for a relocation to be completed.
This timeout needs to be set with sufficient time so that SRNS procedure aborts can be avoided if the peer
fails to respond in a timely fashion in the case of a hard handoff.
Examples The following command sets the wait time for 10 seconds.
relocation-complete-timeout 10
reset
Defines the configuration specific to the RESET procedure.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
Syntax Description reset { ack-timeout time | guard-timeout time | max-retransmissions retries | sgsn-initiated }
default reset { ack-timeout | guard-timeout | max-retransmissions | sgsn-initiated }
no reset sgsn-initiated
default
Returns to the default settings for the Reset procedure.
no
Removes the SGSN-initiated reset procedure from the configuration.
ack-timeout time
Configures the interval (in seconds) for which the SGSN waits for RESET-ACK from the RNC.
time must be an integer from 5 to 60.
Default: 20.
guard-timeout
Configures the interval (in seconds) after which the SGSN sends RESET-ACK to the RNC.
time must be an integer from 5 to 60.
Default : 10
max-retransmissions
Configures maximum retries for RESET message.
sgsn-initiated
Enables SGSN initiated RESET procedure.
Default: disabled.
Examples Use the following to have the SGSN initiate the RESET procedure:
reset sgsn-initiated
rnc
This command creates or accesses an instance of an RNC (radio network controller) configuration.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
no
Removes the configuration information for the specified RNC.
id rnc_id
Set the identification number of the RNC configuration instance.
rnc_id : Must be an integer from 0 to 4095 for 8.1 releases. Must be an integer from 0 to 65535 for releases
9.0 and higher.
Usage Guidelines Use this command to configure information for the IuPS service to use to contact specific RNCs.
This command also provides access to the RNC configuration mode.
security-mode-complete-timeout
This command configures the security mode timer.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer configuration to the default settings.
time
Configures the interval (in seconds) the SGSN waits for the security mode from the MS to complete.
time must be an integer from 1 to 60.
Default is 5
Usage Guidelines Use this command to configure the timer that determines how long the SGSN waits for a Security Mode
Complete message from the MS (mobile station).
service-request-follow-on
Instructs the SGSN not to release an Iu immediately.
Product SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the configuration to the default, this function is enabled.
no
Disables this function so that Iu is released without waiting for the Iu-Hold-Timer to expire.
Usage Guidelines For an Iu established as the result of a Service Request (signaling), the SGSN, by default, waits for the
Iu-Hold-Timer to expire.
Use this command with the 'no' prefix to disable this function.
Use this command with the 'default' prefix or without any prefix if the configuration was modified previously
with by no service-request-follow-on.
srns-context-response-timeout
This command configures the SGSN context response timer.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer configuration to the default setting.
time
Configures the interval (in seconds) for which the SGSN waits for an SRNS Context Request message.
time must be an integer from 1 to 60.
Default: 5.
Usage Guidelines Configures the time to wait before the SGSN sends a response to the SRNS Context-Request message.
Examples Configure the SGSN to wait 7 seconds for an SRNS Context-Request response:
srns-context-response-timeout 7
tigoc-timeout
This command configures the TigOc interval.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer configuration to the default setting.
time
This command sets the time in seconds.
time : Must be an integer from 1 to 60.
Default: 5.
Usage Guidelines Define the amount of time that the SGSN ignores any overload messages for TigOc interval after receiving
one overload message from the RNC.
Examples Use the following command to change the default TigOc interval to 4 seconds:
tigoc-timeout 4
tintc-timeout
This command configures the TinTc interval..
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > IuPS Service Configuration
configure > context context_name > iups-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ctx-iups-service)#
default
Resets the timer configuration to the default setting.
time
Set the number of seconds to wait.
time : Must be an integer from 1 to 60.
Default: 30.
Usage Guidelines Define 4 as the number of seconds that the SGSN waits before decrementing (by one) the traffic level of the
RNC.
Examples tintc-timeout 4
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
allow
This command configure the system to allow different attributes in the LAC Hostname Attribute Value Pair
(AVP) and Called-Number AVP for L2TP messages exchanged between LAC and LNS.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
Syntax Description allow { aaa-assigned-hostname | called-number value apn | calling-number value imsi }
default allow { aaa-assigned-hostname | called-number value apn }
no allow { aaa-assigned-hostname | called-number value apn | calling-number }
no
Disable the configured attribute and returns to the behavior that uses the LAC-Service name as the HostName
AVP.
aaa-assigned-hostname
When enabled if AAA assigns a valid Tunnel-Client-Auth-ID attribute for the tunnel, it is used as the HostName
AVP in the L2TP tunnel setup message.
This keyword works in conjunction with the local-hostname hostname keyword applied via the tunnel l2tp
command in APN Configuration mode.
When Tunnel parameters are not received from the RADIUS Server, Tunnel parameters configured in an
APN are considered for the LNS peer selection. When APN configuration is selected, the local-hostname
configured with the tunnel l2tp command in the APN for the LNS peer will be used as an LAC Hostname.
Important This is a customer-specific keyword available for PDSN. Please contact your local Cisco sales representative
for more information.
Usage Guidelines Use this command to configure the attribute for the HostName AVP for L2TP messages exchanged between
LAC and LNS.
LAC Hostname will be different for the subscribers corresponding to the different corporate APNs. In the
absence of a AAA assigned HostName, the LAC-Service name is used as HostName. By default the
LAC-Service name is used as the HostName AVP.
Examples The following command enables the use of the value of Tunnel-Client-Auth-ID attribute for the HostName
AVP:
allow aaa-assigned-hostname
Use the following command to reset the behavior so that the LAC-Service uses the LAC-Service name as the
HostName AVP:
no allow aaa-assigned-hostname
bind
This command assigns a local end point address to the LAC service in the current context.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Unassign, or unbind, the local end point to the LAC service.
ip_address
This must be a valid IP address entered using IPv4 dotted-decimal notation.
max-subscribers
The maximum number of subscribers that can use the endpoint for this LAC service. Must be an integer from
1 to 2500000.
Usage Guidelines Use this command to bind a local end point IP address to the LAC service.
Examples The following command binds the local end point IP address 10.10.10.100 to the LAC service in the current
context:
bind 10.10.10.100
The following command removes the binding of the local end point to the LAC service:
no bind
data sequence-number
Enables data sequence numbering for sessions that use the current LAC service. Data sequence numbering is
enabled by default.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disables data sequence numbering for sessions.
Usage Guidelines An L2TP data packet header has an optional data sequence numbers field. The data sequence number may be
used to ensure ordered delivery of data packets. This command is used to re-enable or disable the use of the
data sequence numbers for data packets.
Examples Use the following command to disable the use of data sequence numbering:
no data sequence-number
Use the following command to re-enable data sequence numbering:
data sequence-number
default
This command sets the specified LAC service parameter to its default value or setting.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
data sequence-number
Enables data sequence numbering for sessions.
hide-attributes
Disables hiding attributes in control messages sent from the LAC to the LNS.
keepalive-interval
Sets the interval for send L2TP Hello keepalive if there is no control or data transactions to the default value
of 60 seconds.
load-balancing
Sets the load balancing algorithm to be used when many LNS peers have been configured to the default of
round robin.
local-receive-window
Sets the window size to be used for the local side for the reliable control transport to the default of 16.
max-retransmission
Sets the maximum number of retransmissions to the default of 5.
max-session-per-tunnel
Sets the maximum number of sessions per tunnel at any point in time to the default of 512.
max-tunnel-challenge-length
Sets the maximum length of the tunnel challenge to the default of 16 bytes.
max-tunnels
Sets the maximum number of tunnels for this service to the default of 32000.
proxy-lcp-authentication
Sets sending of proxy LCP authentication parameters to the LNS to the default state of enabled.
retransmission-timeout-first
Sets the first retransmit interval to the default of 1 second.
retransmission-timeout-max
Sets the maximum retransmit interval to the default of 8 seconds.
trap all
Generates all supported SNMP traps.
tunnel-authentication
Sets tunnel authentication to the default state of enabled.
Usage Guidelines Use the default command to set LAC service parameters to their default states.
Examples Use the following command to set the keep alive interval to the default value of 60 seconds:
default
keepalive-interval
Use the following command to set the maximum number of sessions per tunnel to the default value of 512:
default max-session-per-tunnel
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
hide-attributes
Enables hiding certain attributes (such as proxy-auth-name and proxy-auth-rsp) in control messages sent from
the LAC to the LNS. The LAC hides such attributes only if tunnel authentication is enabled between the LAC
and the LNS.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disable hiding attributes.
Usage Guidelines Use this command to hide certain attributes from control messages when tunnel authentication is enabled
between the LAC and the LNS.
keepalive-interval
This command specifies the amount of time to wait before sending a Hello keep alive message.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disables the generation of Hello keepalive messages on the tunnel.
seconds
Default: 60
The number of seconds to wait before sending a Hello keepalive message. The number can be configured to
an integer from 30 to 2147483648.
Usage Guidelines Use this command to set the amount of time to wait before sending a Hello keepalive message or disable the
generation of Hello keep alive messages completely. A keepalive mechanism is employed by L2TP in order
to differentiate tunnel outages from extended periods of no control or data activity on a tunnel. This is
accomplished by injecting Hello control messages after a specified period of time has elapsed since the last
data or control message was received on a tunnel. As for any other control message, if the Hello message is
not reliably delivered then the tunnel is declared down and is reset. The transport reset mechanism along with
the injection of Hello messages ensures that a connectivity failure between the LNS and the LAC is detected
at both ends of a tunnel.
Examples Use the following command to set the Hello keepalive message interval to 120 seconds:
keepalive-interval 120
Use the following command to disable the generation of Hello keepalive messages:
no keepalive-interval
load-balancing
Configures how LNSs are selected for this LAC service.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
balanced
LNS selection is made without regard to prioritization, but in a sequential order that balances the load across
the total number of LNS nodes available.
prioritized
LNS selection is made based on the priority assigned in the Tunnel-Preference attribute. An example of this
method is three LNS nodes, with preferences of 1, 2, and 3 respectively. In this example, the RADIUS server
always tries the tunnel with a preference of 1 before using any of the other LNS nodes.
random
Default: Enabled
LNS selection is random in order, wherein the RADIUS server does not use the Tunnel-Preference attribute
in determining which LNS to select.
Usage Guidelines Use this command to configure the load-balancing algorithm that defines how the LNS node is selected by
the LAC when there are multiple peer LNSs configured in the LAC service.
Examples The following command sets the LAC service to connect to LNSs in a sequential order;
load-balancing balanced
The following command sets the LAC service to connect to LNSs according to the priority assigned through
the Tunnel-Preference attribute:
load-balancing prioritized
local-receive-window
Specifies the number of control messages the remote peer LNS can send before waiting for an
acknowledgement.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 4
Specifies the number of control messages to send before waiting for an acknowledgement. The number can
be configured to an integer from 1 to 256.
Usage Guidelines Use this command to set the size of the control message receive window being offered to the remote peer
LNS. The remote peer LNS may send the specified number of control messages before it must wait for an
acknowledgment.
Examples The following command sets the local receive window to 10 control messages:
local-receive-window 10
max-retransmission
Sets the maximum number of retransmissions of a control message to a peer before the tunnel and all sessions
within it are cleared.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 5
Specifies the maximum number of retransmissions of a control message to a peer. This value must be an
integer from1 through 10.
Usage Guidelines Each tunnel maintains a queue of control messages to be transmitted to its peer. After a period of time passes
without acknowledgement, a message is retransmitted. Each subsequent retransmission of a message employs
an exponential backoff interval. For example; if the first retransmission occurs after 1 second, the next
retransmission occurs after 2 seconds has elapsed, then the next after 4 seconds. If no peer response is detected
after the number of retransmissions set by this command, the tunnel and all sessions within are cleared.
Use this command to set the maximum number of retransmissions that the LAC service sends before closing
the tunnel and all sessions within. it.
Examples The following command sets the maximum number of retransmissions of a control message to a peer to 7:
max-retransmissions 7
max-session-per-tunnel
Sets the maximum number of sessions that can be facilitated by a single a tunnel at any time.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 512
The maximum number of sessions expressed as an integer from 1 through 65535.
Usage Guidelines Use this command to set the maximum number of sessions you want to allow in a tunnel.
Examples The following command sets the maximum number of sessions in a tunnel to 5000:
max-sessions-per-tunnel 5000
max-tunnel-challenge-length
Sets the maximum length of the tunnel challenge in bytes.The challenge is used for tunnel authentication
purposes during tunnel creation.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
bytes
Default: 16
Specifies the maximum length (in bytes of the tunnel challenge. This must be an integer from 4 through 32.
Usage Guidelines Use this command to set the maximum length (in bytes) for the tunnel challenge that is used during tunnel
creation.
Examples The following command sets the maximum length of the tunnel challenge to 32 bytes:
max-tunnel-challenge-length 32
max-tunnels
The maximum number of tunnels that the current LAC service can support.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 32000
The maximum number of tunnels expressed as an integer from 1 through 32000.
Usage Guidelines Use this command to set the maximum number tunnels that this LAC service can support at any on time.
Examples Use the following command to set the maximum number of tunnels for the current LAC service to 20000:
max-tunnels 20000
peer-lns
Adds a peer LNS address for the current LAC service. Up to eight peer LNSs can be configured for each LAC
service.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
Syntax Description peer-lns ip_address [ encrypted ] secret secret [ crypto-map map_name { [ encrypted ] isakmp-secret
secret } ] [ description text ] [ preference integer ]
no peer-lns ip_address
no peer-lns ip_address
Deletes the peer LNS at the IP address specified by ip_address. ip_address must be entered in IPv4
dotted-decimal notation.
ip_address
The IP address of the peer LNS for the current LAC service. ip_address must be entered in IPv4 dotted-decimal
notation.
description text
Specifies the descriptive text to use to describe the specified peer LNS. text must be an alphanumeric string
of 0 through 79 characters.
preference integer
This sets the priority of the peer LNS if multiple peer LNSs are configured. integer must be an integer from
1 through 128.
Usage Guidelines Use this command to add a peer LNS address for the current LAC service.
Examples The following command adds a peer LNS to the current LAC service with the IP address of 10.10.10.100,
sets encryption on, specifies the shared secret to be 1b34nnf5d, and sets the preference to 3:
peer-lns 10.10.10.100 encrypted secret 1b34nnf5d preference 3
The following command removes the peer LNS with the IP address of 10.10.10.200 for the current LAC
service:
no peer-lns 10.10.10.200
proxy-lcp-authentication
Enables and disables the sending of proxy LCP authentication parameters to the LNS.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disables the sending of proxy LCP authentication parameters to the LNS.
proxy-lcp-authentication
Default: Enabled
Enables the sending of proxy LCP authentication parameters to the LNS.
Usage Guidelines Use this feature in situations where the peer LNS does not understand the proxy LCP Auth AVPs that the
system sends and does not do an LCP renegotiation and tears down the call.
Examples Use the following command to disable the sending of proxy LCP authentication parameters to the LNS;
no proxy-lcp-authentication
Use the following command to re-enable the sending of proxy LCP authentication parameters to the LNS:
proxy-lcp-authentication
retransmission-timeout-first
Each tunnel maintains a queue of control messages to transmit to its peer. After a period of time passes without
acknowledgement, a message is retransmitted. This command sets the initial timeout for retransmission of
control messages.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 1
The amount of time to wait (in seconds) before sending the first control message retransmission. This must
be an integer from 1 through 100.
Usage Guidelines Use this command to set the initial timeout before retransmitting control messages to the peer.
Examples The following command sets the initial retransmission timeout to 3 seconds:
retransmission-timeout-first 3
retransmission-timeout-max
Configures maximum amount of time between two retransmission of control messages.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
integer
Default: 8
integer is the maximum time (in seconds) to wait before retransmitting control messages expressed as e an
integer from 1 through 100.
Usage Guidelines Use this command to set the maximum amount of time that can elapse before retransmitting control messages.
Each tunnel maintains a queue of control messages to transmit to its peer. After a period of time passes without
acknowledgement, a message is retransmitted. Each subsequent retransmission of a message employs an
exponential backoff interval.
Examples The following command sets the maximum retransmission time-out to 10 seconds:
retransmission-timeout-max 10
single-port-mode
This command enables/disables the L2TP LAC service always to use standard L2TP port 1701 as source port
for all L2TP control and data packets originated from LAC node.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
default
Default: Enabled
Sets this command to its default state of disabled. By default single source port configuration for L2TP LAC
packets is disabled.
no
Disables the configured single source port configuration from this LAC service.
Usage Guidelines Use this command to enable or disable the single port mode for L2TP LAC service.
If this feature is enabled, then L2TP LAC service will always use standard L2TP port 1701 as source port for
all L2TP control/data packets originated from LAC (instead of the default scheme in which each L2TPMgr
uses a dynamic source port). L2TPMgr instance 1 will handle all L2TP calls for the service.
Caution Changing this configuration, while the service is already running, will cause restart of the service.
Examples The following command enables the LAC service to use port 1701 as source port for all L2TP control and
data packets:
single-port-mode
snoop framed-ip-address
When enabled, this feature allows the LAC to detect IP Control Protocol (IPCP) packets exchanged between
the mobile node and the LNS and extract the framed-ip-address assigned to the mobile node. The address will
be reported in accounting start/stop messages and displayed for subscriber sessions.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
default
Disabled.
no
Disables the feature. Accounting start/stop will occur before the PPP session is established and the framed IP
address field will be reported as 0.0.0.0.
Usage Guidelines This feature is available to address simple IP roaming scenarios. If this feature is enabled, the Accounting
Start will be sent only after the framed-ip-address is detected. If the framed-ip-address is not detected within
16 seconds, an Accounting Start will be sent for the session with the 0.0.0.0 address. If the session is
disconnected during the detection attempt, Accounting Start/Stop will be sent for the session. If the session
renegotiates IPCP, an Accounting Stop will be generated with a framed-ip-address from the old session, and
an Accounting Start will be generated with an IP address for the new session. IPv6 address detection is not
supported.
Important When this feature is enabled and the show subscribers all command is invoked, the framed-IP-address is
displayed for the PDSN Simple IP subscriber in the output display.
trap
This command generates SNMP traps.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disables SNMP traps.
Usage Guidelines Use this command to enable/disable all supported SNMP traps.
Examples To enable all supported SNMP traps, enter the following command:
trap all
tunnel selection-key
Enables the creation of tunnels between an L2TP service and an LNS server on the basis of a key received
from AAA server.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
default
Disables the creation of tunnel between LAC service and LNS based on a key value received from AAA
server.
none
Default: Enabled
This keyword disables the creation of multiple tunnels between a pair of LAC service and LNS server. LAC
will not make use of the key to choose a tunnel with LNS in this setup.
tunnel-client-auth-id
Default: Disabled
This keyword enables the creation of tunnels between LAC service and an LNS server on the basis of domain
attribute "Tunnel-Client-Auth-ID" value received from AAA server.
tunnel-server-auth-id
Default: Disabled
This keyword enables the creation of tunnels between LAC service and an LNS server on the basis of domain
attribute "Tunnel-Server-Auth-ID" value received from AAA server.
Usage Guidelines Use this command to enable or disable the creation of additional L2TP tunnels between LAC service and LNS
server on the basis of "Tunnel-Client-Auth-ID" or "Tunnel-Server-Auth-ID" attribute value received from
AAA Server in Access-Accept message. This value of attribute is treated as a key for tunnel selection and
creation.
When the LAC needs to establish a new L2TP session, it first checks for an existing L2TP tunnel with the
peer LNS based on the value of the key configured. If no such tunnel exists for the key, it will create a new
tunnel with the LNS.
The default configuration has the selection-key as none. Hence, LAC will not make use of key to choose a
tunnel with LNS in default setup.
The maximum number of sessions, as configured via the max-sessions-per-tunnel command, is applicable
for each tunnel created through this command. By default, each tunnel supports 512 sessions.
If the LAC service needs to establish a new tunnel for a new L2TP session with LNS and the tunnel create
request fails because maximum tunnel creation limit is reached, LAC will try other LNS addresses received
from AAA server in Access-Accept message for the APN/subscriber. If all available peer-LNS are exhausted,
LAC service will reject the call.
Examples The following command enables the use of "Tunnel-Server-Auth-ID" attribute value received from AAA
Server in Access-Accept message as a key for tunnel selection and creation:
tunnel selection-key tunnel-server-auth-id
tunnel-authentication
Enables tunnel authentication. When tunnel authentication is enabled, a configured shared secret is used to
ensure that the LAC service is communicating with an authorized peer LNS. The shared secret is configured
by the peer-lns command in the LAC Service Configuration mode, the tunnel l2tp command in the Subscriber
Configuration mode, or the Tunnel-Password attribute in the subscribers RADIUS profile.
Product GGSN
PDSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > LAC Service Configuration
configure > context context_name > lac-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lac-service)#
no
Disables tunnel authentication.
Tunnel authentication is enabled by default.
Usage Guidelines Disable or enable the usage of secrets to authenticate a peer LNS when setting up a tunnel.
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
length
Configures the output for the display's length (number of rows).
Product All
default
Restores the default value for the number of rows (length) that will be displayed in the output.
number
Specifies the number of rows (lines) of output that can be displayed on the terminal. number must be 0 or an
integer from 5 through 512, where the special value 0 implies an infinite number of rows.
Usage Guidelines Use this command to set the display terminal's output length other than the default. The special infinite value
(0) is typically used when logging the output of a session from a remote machine since this will result in no
pagination of output.
Examples The following command sets the length of the display to 33 rows.
length 33
width
Configures the output for the displays width (number of characters in a single row).
Product All
default
Restores the default value for the number of characters in a single row (width) that will be displayed in the
output on the terminal.
number
Specifies the number of characters in a single row that can be displayed on the terminal. number must be an
integer from 5 through 512.
Usage Guidelines Use this command to set the display terminal's output width other than the default.
Examples The following command sets the width of the display to 75 characters.
width 75
The Link configuration mode defines the MTP3 link parameters for a specific link in a linkset of an SS7
routing domain instance.
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
arbitration
This command configures link arbitration.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
no
Removes the arbitration configuration for the link.
active
The SSCOP initiates the transmission of PDUs.
passive
The SSCOP waits to receive PDUs.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
mtp2-aerm-emergency-threshold
Configure the alignment error rate monitor (AERM) emergency threshold value. This command is only
available for a lowspeed-narrowband link-type.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value of 1.
value
value : Enter an integer from 1 to 50. Default: 1.
Usage Guidelines This command sets the emergency threshold for the MTP2 alignment error rate monitor.
mtp2-aerm-normal-threshold
Configure the alignment error rate monitor (AERM) normal threshold value. This command is only available
for a lowspeed-narrowband link-type.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value of 4.
value
value : Enter an integer from 4 to 100. Default: 4.
Usage Guidelines This command sets the normal threshold for the MTP2 alignment error rate monitor.
mtp2-eim-decrement
Configure the errored interval monitor (EIM) emergency decrement value. This command is only available
for a highspeed-narrowband link-type.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value of 11.
value
value : Enter an integer from 1 to 63. Default: 11.
Usage Guidelines This command sets the emergency decrement value for the EIM.
mtp2-eim-increment
Configure the errored interval monitor (EIM) emergency increment value. This command is only available
for a highspeed-narrowband link-type.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value of 198.
value
value : Enter an integer from 1 to 1023. Default: 198.
Usage Guidelines This command sets the emergency increment value for the EIM.
mtp2-eim-threshold
Configure the errored interval monitor (EIM) emergency threshold value. This command is only available
for a highspeed-narrowband link-type.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value of 794.
value
value : Enter an integer from 1 to 65535. Default: 794.
Usage Guidelines This command sets the emergency threshold value for the EIM.
mtp2-error-correction
Configure the error correction method to be used. This command is only available for lowspeed or highspeed
narrowband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value.
basic
Basic error correction (BEC) is a positive / negative acknowledgement method that uses backwards
retransmission. This method is best for links with less than 30 ms one-way propagation delays.
preventtive-cyclic-retransmission
PCR is recommended for links with 125 ms, or higher, propagation delays.
Usage Guidelines Set the method of MTP2 layer error correct to be used on the link.
mtp2-lssu-len
This command sets the length of the link status signal unit (LSSU) which carries link status information used
to manage link alignment and indicate the status of the signaling points to each other. This command is only
available for lowspeed or highspeed narrowband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Using this keyword with the command resets the length to the default of 1 octet.
#_octets
Sets the number of octets for the length of the LSSU.
#_octets: Must be either 1 or 2.
Usage Guidelines Use this command to define the maximum amount of link status information that is to be shared between
signaling points.
Examples You can use the following command to set the LSSU length to 2 octets - the maximum length:
mtp2-lssu-len 2
mtp2-max-outstand-frames
This command sets the maximum number of outstanding packets to be sent by the link manager (linkmgr) -
applicable for both high speed (HSL) and low speed (LSL) narrowband links.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Using this keyword with the command resets number of packets to the default of 7 bytes.
#_bytes
Sets the maximum number of packets sent by the linkmgr that can be allowed to be outsanding.
#_bytes: Must be an integer from 5 to 10.
Usage Guidelines The linkmgr (MTP2) sends data at a higher rate, than the narrowband (NB) E1 link speed, when in congestion
and performing retransmission. This can lead to more congestion leading to more time taken for the link to
come out of congestion. If using a value of 10 during congestion, then linkmgr pumps data at a rate higher
than 2.5 mbps. To avoid this problem, a lower value is usually considered optimal. This configuration holds
good for both HSL and LSL.
Examples Use the following command to reset the default number of outstanding packets sent by the LinkMgr:
default mtp2-max-outstand-frames
Set the maximum number of oustanding packets the linkmgr can send to 6:
mtp2-max-outstand-frames 6
mtp2-suerm-threshold
Configure the signal unit error rate monitor (SUERM) threshold. This command is only available for
lowspeed-narrowband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the parameter to the default value.
value
Defines the threshold for number of bad frames
value: Enter an integer from 64 to 1023. Default is 64.
Usage Guidelines Sets the threshold for link monitoring of bad frames.
Examples Set a new link monitoring bad frames (SEURM) threshold of 256:
mtp2-suerm-threshold 256
mtp3-discard-priority
Configure MTP3 message discard priority.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the priority to the default value.
priority
priority: must be an integer between 0 and 3.
Default is 0.
Examples mtp3-discard-priority 2
mtp3-max-slt-try
Configure maximum number of times to retry SLT (signaling link test).
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the number of retries to the default value.
retries
retries: must be an integer between 1 to 65535.
Default is 10.
Examples mtp3-max-slt-try 35
mtp3-msg-priority
Configures the priority for sending MTP3 management messages.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the number of priority to the default value.
priority
priority: must be an integer from 0 to 3.
Default: 0
Usage Guidelines Use this command to set the priority for sending MTP3 management messages.
mtp3-msg-size
Configures the size of messages from layer 3 to layer 2.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the the size to the default value which is 4096 (for q.2140) or 272 (for MTP2)
size
size: must be an integer from 1 to 272 for high-speed or low-speed narrowband SS7 links.
size: must be an integer from 1 to 4096 for ATM broadband links.
Usage Guidelines Use this command to set the maximum message size, in bytes.
Examples Use this command to set the MTP3 message size to 4096 bytes:
mtp3-msg-size 4096
mtp3-p1-qlen
Configure the size for the MTP3 p1 queue length.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the number of size of the priority 1 queue to the default value.
size
size: integer from 1 to 65535. Size should be less than MTP3 p2 qlen and p3 qlen.
Default: 1024
Usage Guidelines Use this command to configure the queue length threshold for raising the congestion priority to level 1.
Examples Use this command to set the queue length priority to 128:
mtp3-p1-qlen 128
mtp3-p2-qlen
Configure the size of the priority 2 queue.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the number of size of the priority 2 queue to the default value.
size
size: integer from 1 to 65535. Size should be less than MTP3 p3 qlen and greater than p1 qlen.
Default: 1024
Usage Guidelines Use this command to configure the queue length threshold for raising the congestion priority to level 2.
Examples Use this command to set the queue length threshold to 256:
mtp3-p2-qlen 256
mtp3-p3-qlen
Configure the size of the priority 3 queue.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the number of size of the priority 3 queue to the default value.
size
size: integer from 1 to 65535. Size should be greater than MTP3 p1 qlen and p2 qlen .
Default: 1024
Usage Guidelines Use this command to configure the queue length threshold for raising the congestion priority to level 3.
mtp3-test-pattern
Configures the character string for the test message.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Resets the pattern to the default value.
pattern
pattern: 1 to 15 alphanumeric characters.
Default: SGSN-ORIGINATED
Usage Guidelines Use this command to define a test pattern string for the signalling link test match (SLTM).
priority
Configures the MTP3 Link Priority.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
no
Removes the priority configuration.
pri_value
pri_value: 0 represents highest priority and 15 represents the lowest priority.
Usage Guidelines Use this command to configure the link priority within the MTP3 link set.
Examples priority 2
signaling-link-code
Configures the signaling link code (SLC).
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
no
Removes the SLC configuration.
code
code: integer from 0 to 15.
Usage Guidelines Use this command to uniquely identify the signaling link to be used for MTP3 management messages.
Examples signaling-link-code 4
sscf-nni-n1
Configures the SSCF NNI N1. This command is only available for ATM-broadband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Removes the sscf-nni-n1 configuration.
value
value: integer from 1 to 65535.
Default: 1000
Usage Guidelines Use this command to identify the network-to-node interface (NNI) between the MTP3 and SSCOP layers.
sscop-max-cc
Configure the maximum value for the SSCOP connection control (CC) state variable. his command is only
available for ATM-broadband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Removes the sscop-max-cc configuration.
value
value: integer from 1 to 65535.
Default: 4
Usage Guidelines Use this command as part of the configuration responsible for managing the SSCOP connection. This command
sets the number of times retries.
sscop-max-pd
Configures the maximum acceptable value for the SSCOP state variable VT(PD). his command is only
available for ATM-broadband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Removes the sscop-max-pd configuration.
value
value: integer from 1 to 65535.
Default: 500
Usage Guidelines Use this command to define the maximum number of data PDUs transmitted between POLL PDUs.
sscop-max-stat
Configures the maximum number of elements included in a status PDU. his command is only available for
ATM-broadband link-types.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
default
Removes the sscop-max-stat configuration.
value
value: integer from 3 to 65535. This parameter should be an odd integer greater than or equal to 3.
Defaultz; 67
Usage Guidelines Received in response to a POLL PDU, the STAT PDU includes information about the number of SD PDUs
that have been received.
timeout
This command enables configuration of an array of signaling and flow control timers - for MTP, SSCF, and
SSCOP.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration > Link
Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id > link id link_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id-link-link_id)#
no
Adding no to the timeout command removes the timer configuration.
timer timer_value
Select the timer and enter a value from the range.
For timers having different ranges for highspeed and lowspeed links or for different variants, the appropriate
ranges will be displayed based on the link-type configured.
Usage Guidelines For a single link and specified link-type (highspeed or lowspeed), this command sets the timer values listed
above. The SS7 variant is determined when the SS7 routing domain is first defined from the Global
Configuration mode.
Repeat the timeout command (one timer and value per entry) as needed to configure all required timers.
Important Currently, the China variant uses the same timers, values, and defaults as the ITU variant.
The Linkset configuration mode defines the MTP3 linkset parameters for a specific SS7 routing domain
instance.
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
adjacent-point-code
This command defines the point-code for the adjacent (next) network element in the SS7 network.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id)#
point-code
Point-code is an SS7 address for an element in the SS7 network. Point-codes must be defined in dotted-decimal
format in a string of 1 to 11 digits. Format options include:
• 0.0.1 to 7.255.7 for point-code in the ITU range.
• 0.0.1 to 255.255.255 for point-code in the ANSI range.
• 0.0.1 to 15.31.255 for point-code in the TTC range.
• a string of 1 to 11 digits in dotted-decimal to represent a point-code in a different range.
no
Removes the adjacent-point-code configuration for this linkset in the SS7 routing domain
Important Removing the linkset configuration will result in the termination of all of the links within the linkset.
Usage Guidelines Use this command to define the point-code for the adjacent element in the SS7 network.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
link
This command creates an MTP3 link configuration for the SS7 linkset and enters the Link configuration mode.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id)#
no
Disables the specified link configuration.
Important Removing the link configuration will result in the termination of traffic on the specified link.
#_octets
Sets the number of octets for the length of the LSSU.
id
This number uniquely identifies the link in the linkset.
id: an integer between 1 and 16.
link-type
Identifies the signalling type for this link; options include:
• ATM broadband -- ATM AAL5 over an optical line card (OLC2)
• high speed-narrowband -- 64 kbps over a channelized optical line card (CLC2)
• low speed-narrowband -- 4.8 kbps over a channelized optical line card (CLC2)
Important Be default link-type is ATM-broadband. To support narrowband SS7, one of the other options must be
set.
Usage Guidelines Access the Link configuration mode to configure the parameters for the the link.
self-point-code
This command defines the SS7 network point-code to identify this SGSN.
Product SGSN
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Linkset Configuration
configure > ss7-routing-domain domain_id variant var_type > linkset id linkset_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-linkset-linkset_id)#
point-code
Point-code is an SS7 address for an element in the SS7 network. Point-codes must be defined in dotted-decimal
format in a string of 1 to 11 digits. Format options include:
• 0.0.1 to 7.255.7 for point-code in the ITU range.
• 0.0.1 to 255.255.255 for point-code in the ANSI range.
• 0.0.1 to 15.31.255 for point-code in the TTC range.
• a string of 1 to 11 digits in dotted-decimal to represent a point-code in a different range.
no
Removes the self-point-code configuration for this linkset in the SS7 routing domain.
Important Removing the self-point-code will result in the termination of all traffic on this link.
Usage Guidelines Use this command to define the SS7 point-code to identify this system.
Command Modes Exec > Global Configuration > Context Configuration > LMA Service Configuration
configure > context context_name > lma-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lma-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
aaa accounting
Enables the LMA to send AAA accounting information for subscriber sessions.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > LMA Service Configuration
configure > context context_name > lma-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lma-service)#
default
Sets the command to the default condition of enabled.
no
Disables the ability of the LMA to send AAA accounting information.
Usage Guidelines Use this command to enable the LMA service to send all accounting data (start, stop, and interim) to the
configured AAA servers.
Important In order for this command to function properly, AAA accounting must be enabled for the context in which
the LMA service is configured using the aaa accounting subscriber radius command.
Examples The following command disables aaa accounting for the LMA service:
no aaa accounting
alt-coa-allowed
Allows Alternate Care-of-address support to be added at LMA to separate signaling and control plane traffic.
Product P-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > LMA Service Configuration
configure > context context_name > lma-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lma-service)#
default
Including this keyword with the command disables the feature. The feature is disabled by default.
no
Disables the specified functionality.
Usage Guidelines This command allows Alternate Care-of-address support to be added at LMA to separate signaling and control
plane traffic.
Important The support of the extensions and functionality is defined in RFC 6275 and RFC 6463 for IPv6 and IPv4
transport respectively.
Proxy Mobile IPv6 is a network-based mobility management protocol. The mobility entities involved in the
Proxy Mobile IPv6 protocol, the Mobile Access Gateway (MAG) and the Local Mobility Anchor (LMA),
setup tunnels dynamically to manage mobility for a mobile node within the Proxy Mobile IPv6 domain. There
is an extension to the Proxy Mobile IPv6 protocol to register an IPv4 or IPv6 data plane address that is different
from the Proxy Care-of Address with the LMA. This allows separation of control and data plane. Some of
the deployments of Proxy Mobile IPv6 separated the control and data plane end points for Mobile Access
Gateway. There will be a separate IP address for the entity that sends and received the Proxy Mobile IPv6
signaling messages. Similarly, there will be a separate IP address for the entity that encapsulates and
decapsulates the data traffic to and from the mobile node.
In order to allow the separation of the control and data plane, the address of the data plane traffic endpoint
needs to be sent in a separate extension to register two IP addresses with the LMA. The IP address used for
the signaling messages will continue to be called the Proxy Care-of-Address. A separate IP address for the
data plane is carried in the Proxy Binding Update to indicate the tunnel end point for the data traffic.
The extension Alternate Care-of-Address Mobility Option defined in RFC 6275 should be used. When using
IPv6 transport and IPv4 transport, Alternate Ipv4 Care of Address Mobility Option defined in RFC 6463
should be used.
Normally, a binding update specifies the desired care-of-address in the source address field of the IPv6 header.
However, in some cases such as when the mobile node wishes to indicate a Care-of Address that it cannot
use as a topologically correct source address or when the used security mechanism does not protect the IPv6
header it is not possible.
The Alternate Care-of-Address option is for this type of situation. This option is valid only in binding update.
The Alternate Care-of Address field contains an address to use as the care-of-address for binding rather than
using the source address of the packet as the care-of-address.
bind address
Binds the LMA service to a logical IP interface serving as the S2a (HSGW) or S5/S8 (S-GW) interface and
specifies the maximum number of subscribers that can access this service over the configured interface.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > LMA Service Configuration
configure > context context_name > lma-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lma-service)#
no
Removes the interface binding from this service.
address ipv6_address
Specifies the IPv6 address of the interface configured as the S2a or S5/S8 interface. ipv6_address is specified
in colon separated notation.
ipv4-address ipv4_address
Specifies optional IPv4 HA/P-GW address to support DSMIP6 session using IPv4 transport.ipv4_address
must be entered as a standard IPv4 address in dotted decimal notation.
max-subscribers num
Default: 3000000
Specifies the maximum number of subscribers that can access this service on this interface. num must be
configured to an integer between 0 and 3,000,000.
Important The maximum number of subscribers supported is dependant on the license key installed and the number
of active PSCs in the system. A fully loaded system with 13 active PSCs can support 3,000,000 total
subscribers. Refer to the license key command and the Usage section (below) for additional information.
Usage Guidelines Associate the LMA service to a specific logical IP address. The logical IP address or interface takes on the
characteristics of an S2a or S5/S8 interface that provides the session connectivity to an HSGW (S2a) or S-GW
(S5/S8). Only one interface can be bound to a service. The interface should be configured prior to issuing this
command.
This command also sets a limit as to the number of simultaneous subscribers sessions that can be facilitated
by the service/interface at any given time.
When configuring the max-subscribers option, be sure to consider the following:
• The total number of S2a or S5/S8 interfaces you will configure
• The total number of subscriber sessions that all of the configured interfaces may handle during peak
busy hours
• An average bandwidth per session multiplied by the total number of sessions
• The type of physical port (10/100Base-T or 1000Base-Tx) that these interfaces will be bound to
Taking these factors into account and distributing your subscriber session across all available interfaces will
allow you to configure your interfaces to optimally handle sessions without degraded performance.
Examples The following command would bind the logical IP interface with the address of
4551:0db8:85a3:08d3:3319:8a2e:0370:1344 to the LMA service and specifies that a maximum of 300,000
simultaneous subscriber sessions can be facilitated by the interface/service at any given time:
bind address 4551:0db8:85a3:08d3:3319:8a2e:0370:1344 max-subscribers 300000
end
Exits the current mode and returns to the Exec Mode.
Product All
Privilege Administrator
exit
Exits the current mode and returns to the previous mode.
Product All
Privilege Administrator
heartbeat
Configures the PMIPv6 heartbeat message interval, retransmission timeout, and max retransmission for the
LMA Service.
Product P-GW
Privilege Administrator
Syntax Description heartbeat { interval seconds | retransmission { max number | timeout seconds } }
default heartbeat { interval | retransmission { max | timeout } }
no heartbeat
no
Disables the PMIPv6 heartbeat functionality. The P-GW starts sending heartbeat request to peers when the
heartbeat interval is configured.
default
Resets the specified parameter to the system default value.
interval seconds
The interval in seconds at which heartbeat messages are sent.
seconds is an integer from 30 through 3600.
Default: 60
Usage Guidelines Proxy Mobile IPv6 (PMIPv6) is a network-based mobility management protocol to provide mobility without
requiring the participation of the mobile node in any PMIPv6 mobility related signaling. The Local Mobility
Anchor (LMA) service sets up tunnels dynamically to manage mobility for a mobile node.
This command provides configuration of heartbeat messages between the LMA and MAG services to know
the reachability of the peers, to detect failures, quickly inform peers in the event of a recovery from node
failures, and allow a peer to take appropriate action.
Examples The following command enables PMIPv6 heartbeat messaging to known LMA service peers and sets the
heartbeat interval to 160 seconds.
heartbeat interval 160
mobility-option-type-value
Changes the mobility option type value used in mobility messages.
Product P-GW
SAEGW
Privilege Administrator
default
Sets the command to the default value of custom1.
custom1
(Default) Non-standard type values used before they were defined by IANA.
custom2
Standard type values, as defined by IANA, and some customer-specific message formats.
standard
Standard type values as defined by IANA. In addition, standard option uses type values defined in RFC 5844
for HoA options for PMIPv6 PBU/PBA/revocation message.
Usage Guidelines Use this command to change the mobility option type value used in mobility messages.
Examples The following command changes the mobility option type value to standard:
mobility-option-type-value standard
refresh-advice-option
Configures inclusion of a refresh advice option in the binding acknowledgement message sent by the LMA.
Product P-GW
SAEGW
Privilege Administrator
default
Returns the command setting to the default setting of disabled.
no
Disables the inclusion of the refresh advice option in the binding acknowledgement message sent by the LMA
Usage Guidelines Use this command to enable the LMA to include this option in a binding acknowledgment sent to the requesting
MAG. The option provides a "hint" to the MAG of when it should refresh the binding.
As defined in RFC 3775 "Mobility Support in IPv6", the binding refresh advice option can only be present
in the binding acknowledgement sent from the mobile node's home agent in reply to a registration request. A
refresh interval parameter determines the amount of time until the mobile node must send a new registration
to the home agent to avoid de-registration and loss of session.
Important Refer to the refresh-interval-percent and reg-lifetime commands for a complete understanding of registration
(binding) lifetimes and refresh intervals.
refresh-interval-percent
Configures percentage of the granted registration lifetime to be used in the refresh interval mobility option in
a binding acknowledgement message sent by the LMA service.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of 75.
number
Default: 75
Sets the percent value for session lifetimes for this service.
number must be an integer value from 1 to 99.
Usage Guidelines Use this command to configure the amount of the granted registration lifetime to be used in the refresh interval
mobility option in the binding acknowledgement message sent by the LMA service to the requesting MAG.
Refreshing a binding or registration is based on the granted registration lifetime. Since a refresh request must
be within the granted range of a registration lifetime, this command provides a method of setting the interval
of when a refresh request is sent.
As described in RFC 3775 "Mobility Support in IPv6", if a binding refresh advice option is present in the
binding acknowledgement, the refresh interval field in the option must be a value less than the binding lifetime
(also returned in the binding acknowledgement). The mobile node then should attempt to refresh its registration
at the shorter refresh interval. The home agent will still honor the registration for the lifetime period, even if
the mobile node does not refresh its registration within the refresh period.
Important Refer to the refresh-advice-option and reg-lifetime commands for a complete understanding of registration
(binding) lifetimes and refresh intervals.
Examples The following command sets the refresh interval percent to 90:
refresh-interval-percent 90
reg-lifetime
Configures the Mobile IPv6 session registration lifetime for this service.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of 600.
seconds
Default: 600
Sets the time value for session lifetimes for this service.
seconds must be an integer value from1 to 262140.
Usage Guidelines Use this command to limit PMIPv6 lifetime on this service. If the PBU contains a lifetime shorter than what
is specified, it is granted. If the lifetime is longer, then HA service will limit the granted lifetime to the
configured value.
Important Refer to the refresh-interval-percent and refresh-advice-option commands for a complete understanding
of registration (binding) lifetimes and refresh intervals.
Examples The following command sets the registration lifetime for Mobile IPv6 sessions using this service to 1200
seconds (20 minutes):
reg-lifetime 1200
revocation
Enables the MIP revocation feature and configures revocation parameters.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the keyword to its default value.
no
Disables revocation for this service.
enable
Default: disabled
Enables the MIP registration revocation feature for the LMA service. When enabled, if revocation is negotiated
with a MAG and a MIP binding is terminated, the LMA can send a Revocation message to the MAG. This
feature is disabled by default.
max-retransmission number
Default: 3
The maximum number of retransmissions of a Revocation message before the revocation fails. number must
be an integer value from 0 through 10.
retransmission-timeout msecs
Default: 3000
The number of milliseconds to wait for a Revocation Acknowledgement from the MAG before retransmitting
the Revocation message. msecs must be an integer value from 500 through 10000.
Usage Guidelines Use this command to enable or disable the MIP revocation feature on the LMA or to change settings for this
feature.
Examples The following command sets the maximum number of retries for a Revocation message to 6:
revocation max-retransmission 6
The following command sets the timeout between retransmissions to 10:
revocation retransmission-timeout 10
sequence-number-validate
Configures sequence number validation of the received MIPv6 control packets by the LMA service according
to RFC 3775.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of enabled.
no
Disables the feature.
Usage Guidelines Use this command to configure the sequence number validation of the received MIPv6 control packets (PBUs)
by the LMA service. This feature validates MIPv6 control packets and insures that any incoming packets with
a sequence number prior to the last number received is consider invalid.
If this service has no cache entry of the home address included in the PBU, it will accept any sequence value
in the initial PBU from the mobile node.
setup-timeout
The maximum amount of time allowed for session setup.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of 60.
seconds
Default: 60 seconds
The maximum amount of time, in seconds, to allow for setup of a session in this service. seconds must be an
integer value from 1 through 1000000.
Usage Guidelines Use this command to set the maximum amount of time allowed for setting up a session.
Examples The following command sets the maximum time allowed for setting up a session to 5 minutes (300 seconds):
setup-timeout 300
signalling-packets
Enables the DSCP marking feature for IP headers carrying outgoing signalling packets.
Product P-GW
SAEGW
Privilege Administrator
default
Restores the specified parameter to its default setting of 0x0.
no
Disables the specified functionality.
ip-header-dscp value
Used to configure the QoS Differentiated Services Code Point (DSCP) marking for IP header encapsulation.
value: Represents the DSCP setting. It represents the first six most-significant bits of the ToS field. It can be
configured to any hex value from 0x0 through 0x3F. Default is 0x0.
Usage Guidelines Use this command to enable or disable the DSCP marking feature for IP headers carrying outgoing signalling
packets. DSCP marking is disabled by default.
Examples The following command configures the HSGW service to support DSCP marking for IP headers carrying
outgoing signalling packets:
signalling-packets ip-header-dscp 0x21
simul-bindings
Specifies the maximum number of "care-of" addresses that can simultaneously be bound for the same user as
identified by NAI and Home address.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of 1.
number
Default: 1
Configures maximum number of "care of" addresses that can be simultaneously bound for the same user as
identified by their NAI and home address. number must be an integer value between 1 and 3.
Usage Guidelines Per RFC 5213 (and 3775), the LMA service creates a binding record known as a binding cache entry (BCE)
for each subscriber session it is facilitating. Each BCE is associated with a care-of address. As the mobile
node roams, it is possible that the session will be associated with a new care of address.
Typically, the LMA service will delete an old binding and create a new one when the information in the
registration request changes. However, the mobile node could request that the LMA maintains previously
stored BCEs. This command allows you to configure the maximum number of BCEs that can be stored per
subscriber if more than one is requested.
Examples The following command configures the service to support up to 2 addresses per subscriber:
simul-bindings 2
standalone
Configures the LMA service to start in standalone mode.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting.
no
Disables the feature.
Usage Guidelines Use this command to start the LMA service in standalone mode.
timestamp-option-validation
Configures validation of timestamp option in binding update messages. By default, timestamp option is
mandatory.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of enabled.
no
Disables the feature.
Usage Guidelines Use this command to configure timestamp validation in binding update messages.
timestamp-replay-protection
Designates timestamp replay protection scheme as per RFC 4285.
Product P-GW
SAEGW
Privilege Administrator
default
Resets the command value to the default setting of 7.
no
Disables the timestamp replay protection feature.
tolerance seconds
Default: 7
Defines the acceptable difference in timing (between timestamps) before rejecting packet, in seconds. seconds
must be an integer value between 0 and 65535.
Usage Guidelines Use this command to define the acceptable difference in timing (between timestamps) before rejecting packet.
Examples The following command sets the acceptable difference for timestamps to 10 seconds:
timestamp-replay-protection tolerance 10
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
aaa accounting
Enables the sending of authentication, authorization, and accounting (AAA) accounting information by the
LNS.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables this option.
roaming
Enables the sending of AAA accounting information by the LNS only for roaming subscribers.
Usage Guidelines Use this command to enable the sending of AAA accounting information by the LNS. By default this is
enabled.
Examples The following command enables the sending of AAA accounting information by the LNS:
aaa accounting
authentication
Configures the type of subscriber authentication for PPP sessions terminated at the current LNS.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
Syntax Description authentication { { [ allow-noauth ] [ chap chap_priority ] [ mschap mschap_priority ] [ pap pap_priority
] } | msid-auth }
allow-noauth
Default: Disabled
Configures the LNS to allow PPP sessions access even though they have not been authenticated. This command
issued by itself causes the LNS not to attempt authentication for any PPP sessions.
When the allow-noauth option is used in conjunction with commands specifying other authentication protocols
and priorities to use, then if attempts to use those protocols fail, the system treats the allow-noauth option as
the lowest priority.
If no authentication is allowed, the system constructs an Network Access Identifier (NAI) to provide accounting
records for the PPP session.
chap chap_priority
Default: 1
Configures the LNS to attempt to use Challenge Handshake Authentication Protocol (CHAP) to authenticate
the PPP session.
A chap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
chap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference. CHAP
is enabled by default as the highest preference.
mschap mschap_priority
Default: Disabled
Configures the LNS to attempt to use the Microsoft Challenge Handshake Authentication Protocol (MSCHAP)
to authenticate the PPP session.
A mschap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
mschap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference.
pap pap_priority
Default: 2
This option configures the LNS to attempt to use the Password Authentication Protocol (PAP) to authenticate
the PPP session.
A pap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
pap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference. PAP
is enabled by default as the second highest preference.
msid-auth
Default: Disabled
This option configures the LNS to attempt to authenticate the PPP session based on the Mobile Station Identity
(MSID).
Usage Guidelines Use to specify how the LNS service should handle authentication and what protocols to use. The flexibility
is given to configure this option to accommodate the fact that not every mobile will implement the same
authentication protocols.
By default LNS authentication options are set as follows:
• allow-noauth disabled
• chap enabled with a priority of 1
• mschap disabled
• msid-auth disabled
• pap enabled with a priority of 2
Important At least one of the keywords must be used to complete the command.
Examples The following command configures the LNS service to allow no authentication for PPP sessions and would
perform accounting using the default NAI-construct of username@domain:
authentication allow-noauth
The following command configures the system to attempt authentication first using CHAP, then MSCHAP,
and finally PAP. If the allow-noauth command was also issued, when all attempts to authenticate the subscriber
using these protocols failed, then the subscriber would be allowed access:
authentication chap 1 mschap 2 pap 3
Product GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
default
Disables mapping.
no
Disables mapping.
Usage Guidelines For LNS calls received through a LAC, the ICRQ message includes an APN name in the Called Number
AVP. This mapping function enables a GGSN system to provide RADIUS authentication/authorization via
a defined APN in place of an LNS configuration. If the mapped APN has not been defined within the GGSN
configuration then the call will be rejected.
bind
This command assigns the IP address of an interface in the current context to the LNS service.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Unassign, or unbind, the local end point to the LNS service.
ip_address
Specifies the IP address of an interface in the current context. This must be a valid IP address entered using
IPV4 dotted-decimal notation.
max-subscribers max_value
Default: 10000
Specifies the maximum number of subscribers that can be connected to this service at any time. max_value
must be an integer from 1 through 2500000.
Usage Guidelines Use this command to bind the IP address of an interface in the current context to the LNS service.
Examples The following command binds the current context interface IP address 192.168.100.10 to the current LNS
service:
bind 192.168.100.10
The following command removes the binding of the IP address from the LNS service:
no bind
data sequence-number
Enables data sequence numbering for sessions that use the current LNS service. Data sequence numbering is
enabled by default.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables data sequence numbering for sessions.
Usage Guidelines An L2TP data packet header has an optional data sequence numbers field. The data sequence number may be
used to ensure ordered delivery of data packets. This command is used to re-enable or disable the use of the
data sequence numbers for data packets.
Examples Use the following command to disable the use of data sequence numbering:
no data sequence-number
Use the following command to re-enable data sequence numbering:
data sequence-number
default
This command sets the specified LAC service parameter to its default value or setting.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
authentication
Sets the authentication parameters for PPP sessions to the following defaults:
• allow-noauth disabled
• chap enabled with a priority of 1
• mschap disabled
• msid-auth disabled
• pap enabled with a priority of 2
data sequence-number
Enables data sequence numbering for sessions.
ip source-violation
Sets the IP source violation parameters to the following defaults:
• drop-limit 10
keepalive-interval
Sets the interval for send L2TP Hello keepalive if there is no control or data transactions to the default value
of 60 seconds.
local-receive-window
Sets the window size to be used for the local side for the reliable control transport to the default of 4.
max-retransmission
Sets the maximum number of retransmissions to the default of 5.
max-session-per-tunnel
Sets the maximum number of sessions per tunnel at any point in time to the default of 65535.
max-tunnel-challenge-length
Sets the maximum length of the tunnel challenge to the default of 16 bytes.
max-tunnels
Sets the maximum number of tunnels for this service to the default of 32000.
proxy-lcp-authentication
Sets sending of proxy LCP authentication parameters to the LNS to the default state of enabled.
retransmission-timeout-first
Sets the first retransmit interval to the default of 1 second.
retransmission-timeout-max
Sets the maximum retransmit interval to the default of 8 seconds.
setup-timeout
Sets the maximum time allowed for session setup to the default of 60 seconds.
single-port-mode
Disables assignment of only port 1107 for incoming tunnels and allows dynamic assignment of ports.
subscriber
Sets the name of the default subscriber configuration to use.
tunnel-authentication
Sets tunnel authentication to the default state of enabled.
trap all
Generates all supported SNMP traps.
tunnel-switching
Sets the ability of the LNS to create subsequent tunnels to the default of enabled.
Usage Guidelines Use the default command to set LAC service parameters to their default states.
Examples Use the following command to set the keepalive interval to the default value of 60 seconds:
default keepalive-interval
Use the following command to set the maximum number of sessions per tunnel to the default value of 512:
default max-session-per-tunnel
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip source-violation
This command configures settings related to IP source-violation detection.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
Syntax Description ip source-violation { clear-on-valid-packet | drop-limit num | period secs | reneg-limit num }
no ip source-violation clear-on-valid-packet
clear-on-valid-packet
Default: disabled
Configures the service to reset the reneg-limit and drop-limit counters after receipt of a properly addressed
packet.
drop-limit num
Default: 10
Sets the number of allowed source violations within a detection period before forcing a call disconnect. If
num is not specified, the value is set to the default.
num can be an integer from 1 through 1000000.
period secs
Default: 120
The length of time (in seconds) for a source violation detection period to last. drop-limit and reneg-limit
counters are decremented each time this value is reached.
The counters are decremented in this manner: reneg-limit counter is reduced by one (1) each time the period
value is reached until the counter is zero (0); drop-limit counter is halved each time the period value is reached
until the counter is zero (0). If secs is not specified, the value is set to the default.
reneg-limit num
Default: 5
Sets the number of allowed source violations within a detection period before forcing a PPP renegotiation. If
num is not specified, the value is set to the default.
num can be an integer from 1 through 1000000.
Usage Guidelines This function allows the operator to configure a network to prevent problems such as when a user gets handed
back and forth between two PDSNs a number of times during a handoff scenario.
When a subscriber packet is received with a source address violation, the system increments both the IP
source-violation reneg-limit and drop-limit counters and starts the timer for the IP-source violation period.
Every subsequent packet received with a bad source address during the IP-source violation period causes the
reneg-limit and drop-limit counters to increment.
For example, if reneg-limit is set to 5, the system allows five packets with a bad source address (source
violations), but on the fifth packet, it re-negotiates PPP.
If the drop-limit is set to 10, the above process of receiving five source violations and renegotiating PPP occurs
only once. After the second 5-source violation, the call is dropped. The period timer continues to count
throughout this process.
If at any time before the call is dropped, the configured source-violation period is exceeded, the counters for
drop-limit is decremented by half and reneg-limit is decremented by 1. See period definition above.
Examples To set the maximum number of source violations before dropping a call to 100, enter the following command:
ip source-violation drop-limit 100
keepalive-interval
This command specifies the amount of time to wait before sending a Hello keepalive message.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables the generation of Hello keepalive messages on the tunnel.
seconds
Default: 60
Specifies the number of seconds to wait before sending a Hello keepalive message as an integer from 30
through 2147483648.
Usage Guidelines Use this command to set the amount of time to wait before sending a Hello keepalive message or disable the
generation of Hello keepalive messages completely. A keepalive mechanism is employed by L2TP in order
to differentiate tunnel outages from extended periods of no control or data activity on a tunnel. This is
accomplished by injecting Hello control messages after a specified period of time has elapsed since the last
data or control message was received on a tunnel. As for any other control message, if the Hello message is
not reliably delivered then the tunnel is declared down and is reset. The transport reset mechanism along with
the injection of Hello messages ensures that a connectivity failure between the LNS and the LAC is detected
at both ends of a tunnel.
Examples Use the following command to set the Hello keepalive message interval to 120 seconds:
keepalive-interval 120
Use the following command to disable the generation of Hello keepalive messages:
no keepalive-interval
local-receive-window
Specifies the number of control messages the remote peer LAC can send before waiting for an
acknowledgement.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 4
Specifies the number of control messages to send before waiting for an acknowledgement as an integer from
1 through 256.
Usage Guidelines Use this command to set the size of the control message receive window being offered to the remote peer
LAC. The remote peer LAC may send the specified number of control messages before it must wait for an
acknowledgment.
Examples The following command sets the local receive window to 10 control messages:
local-receive-window 10
max-retransmission
Sets the maximum number of retransmissions of a control message to a peer before the tunnel and all sessions
within it are cleared.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 5
Specifies the maximum number of retransmissions of a control message to a peer as an integer from 1 through
10.
Usage Guidelines Each tunnel maintains a queue of control messages to be transmitted to its peer. After a period of time passes
without acknowledgement, a message is retransmitted. Each subsequent retransmission of a message employs
an exponential backoff interval. For example; if the first retransmission occurs after 1 second, the next
retransmission occurs after 2 seconds has elapsed, then the next after 4 seconds. If no peer response is detected
after the number of retransmissions set by this command, the tunnel and all sessions within are cleared.
Use this command to set the maximum number of retransmissions that the LAC service sends before closing
the tunnel and all sessions within. it.
Examples The following command sets the maximum number of retransmissions of a control message to a peer to 7:
max-retransmissions 7
max-session-per-tunnel
Sets the maximum number of sessions that can be facilitated by a single tunnel at any time.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 512
Specifies the maximum number of sessions as an integer from 1 through 65535.
Usage Guidelines Use this command to set the maximum number of sessions you want to allow in a tunnel.
Examples The following command sets the maximum number of sessions in a tunnel to 5000:
max-sessions-per-tunnel 5000
max-tunnel-challenge-length
Sets the maximum length of the tunnel challenge in bytes. The challenge is used for authentication purposes
during tunnel creation.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
bytes
Default: 16
Specifies the number of bytes to set the maximum length of the tunnel challenge as an integer from 4 through
32.
Usage Guidelines Use this command to set the maximum length, in bytes, for the tunnel challenge that is used during tunnel
creation.
Examples The following command sets the maximum length of the tunnel challenge to 32 bytes:
max-tunnel-challenge-length 32
max-tunnels
The maximum number of tunnels that the current LNS service can support.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 32000
Specifies the maximum number of tunnels as an integer from 1 through 32000.
Usage Guidelines Use this command to set the maximum number tunnels that this LNS service can support at any one time.
Examples Use the following command to set the maximum number of tunnels for the current LNS service to 20000:
max-tunnels 20000
nai-construction domain
Designates the alias domain name to use for Network Access Identifier (NAI) construction.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Deletes the NAI construction domain alias.
domain_name { @ | % | - | \ | # | / }
Specifies the desired domain name alias followed immediately by a separator from the valid list. domain_name
must be an alphanumeric string of from 1 through 79 characters.
Usage Guidelines Use this command to specify the domain alias and separator to use for NAI construction. The specified domain
name must be followed by a valid separator (@ | % | - | \ | # | /).
Examples To specify a domain alias of mydomain@ with a separator of @, enter the following command:
nai-construction domain mydomain@
To delete the current setting for the NAI construction domain alias, enter the following command:
no nai-construction domain
newcall
Configures new call related behavior.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
default
Sets or restores default value assigned for specified parameter
duplicate-subscriber-requested-address
Configures how duplicate sessions with same address request are handled.
Examples The following command configures new call with duplicate address request to accept:
newcall duplicate-subscriber-requested-address accept
peer-lac
Adds a peer LAC address for the current LNS service. Up to eight peer LACs can be configured for each LNS
service.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
Syntax Description peer-lac { ip_address | ip_address/mask } [ encrypted ] secret secret [ description text ]
no peer-lac ip_address
no peer-lac ip_address
Deletes the peer LAC IP address specified by ip_address. ip_address must be entered using IPv4 dotted-decimal
notation.
ip_address
The IP address of a specific peer LAC for the current LNS service. ip_address must be entered using IPv4
dotted-decimal notation.
ip_address/mask
A network prefix and mask enabling communication with a group of peer LACs. ip_address is the network
prefix expressed in IPv4 dotted-decimal notation.
mask is the number of bits that defines the prefix.
encrypted
Specifies the encrypted shared key between the LAC and the LNS service.
This keyword is intended only for use by the system while saving configuration scripts. The system displays
the encrypted keyword in the configuration file as a flag that the variable following the secret keyword is the
encrypted version of the plain text secret. Only the encrypted secret is saved as part of the configuration file.
secret secret
Designates the secret which is shared between the current LNS service and the peer LAC. secret must ben
alphanumeric string of 1 through 127 characters that is case sensitive.
description text
Specifies the descriptive text to use to describe the specified peer LAC. text must be an alphanumeric string
of 0 through 79 characters.
Usage Guidelines Use this command to add a peer LAC address for the current LNS service.
Specific peer LACs can be configured by specifying their individual IP addresses. In addition, to simplify
configuration, communication with a group of peer LACs can be enabled by specifying a network prefix and
a mask.
Examples The following command adds a peer LAC to the current LNS service with the IP address of 10.10.10.100,
and specifies the shared secret to be 1b34nnf5d:
peer-lac 10.10.10.100 secret 1b34nnf5d
The following command enables communication with up to 16 peer LACs on the 192.168.1.0 network each
having a secret of abc123:
peer-lac 92.168.1.0/28 secret abc123
The following command removes the peer LAC with the IP address of 10.10.10.200 for the current LNS
service:
no peer-lac 10.10.10.200
proxy-lcp-authentication
Enables/disables proxy LCP authentication.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables the processing of proxy LCP authentication parameters from the LAC.
proxy-lcp-authentication
Default: Enabled
Enables the processing proxy LCP authentication parameters from the LAC.
Usage Guidelines When enabled, if proxy LCP authentication parameters are received from the LAC and are acceptable, the
LNS resumes the PPP session from the authentication phase and goes to the IPCP phase.
When disabled, PPP is always started from the LCP phase, ignoring and discarding any proxy LCP
authentication parameters received from the LAC. Disable this feature in situations where accept proxy LCP
Auth AVPs that the peer LAC sends should not be expected.
Examples Use the following command to disable the processing of proxy LCP authentication parameters from the LAC:
no proxy-lcp-authentication
Use the following command to re-enable the processing of proxy LCP authentication parameters from the
LAC:
proxy-lcp-authentication
retransmission-timeout-first
Configures the initial timeout for the retransmission of control messages to the peer LAC.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 1
Specifies the amount of time (in seconds) to wait before sending the first control message retransmission.
This value is an integer from 1 through 100.
Usage Guidelines Each tunnel maintains a queue of control messages to transmit to its peer. After a period of time passes without
acknowledgement, a message is retransmitted.
Examples The following command sets the initial retransmission timeout to 3 seconds:
retransmission-timeout-first 3
retransmission-timeout-max
Configures the maximum amount of time that can elapse before retransmitting control messages to the peer
LAC.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
integer
Default: 8
Specifies the maximum time (in seconds) to wait before retransmitting control messages. If this limit is reached,
the tunnel, and all sessions within it, is cleared. This value is an integer from 1 through 100.
Usage Guidelines Each tunnel maintains a queue of control messages to transmit to its peer. After a period of time passes without
acknowledgement, a message is retransmitted. Each subsequent retransmission of a message employs an
exponential backoff interval. For example; if the first retransmission occurs after 1 second, the next
retransmission occurs after 2 seconds has elapsed, then the next after 4 seconds. This continues until the limit
set by this command is reached. If this limit is reached, the tunnel, and all sessions within it, is cleared.
Examples Use the following command to set the maximum retransmission time-out to 10 seconds:
retransmission-timeout-max 10
setup-timeout
Configures the maximum amount of time, in seconds, allowed for session setup.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
seconds
Default: 60
Specifies the maximum time (in seconds) to wait for the setup of a session. seconds must be an integer from
1 through 1000000.
Usage Guidelines This command controls the amount of time allowed for tunnel establishment with a peer LAC. If this timer
is exceeded the tunnel setup is aborted.
Examples The following command configures a maximum setup time of 120 seconds:
setup-timeout 120
single-port-mode
When enabled, this command sets the LNS to use only the default local UDP port (port 1701) for the life of
a tunnel.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disable single port mode
Usage Guidelines Use this command to control the L2TP LNS tunnel local UDP port assignment mode. If single-port-mode is
enabled, the LNS-service uses the standard UDP port (port 1701) for the life of the incoming tunnel. Otherwise,
it assigns a new local UDP port number for a tunnel when it responds to a tunnel create request received on
the standard port number. This is done for load distributing the tunnel processing between multiple tasks
within the system to increase the capacity and performance. Even though all L2TP LACs are required to
support such dynamic port assignments during tunnel establishments, there exist some LACs that do not
support port assignment other than port 1701. This single-port-mode feature can be enabled to support such
LAC peers. This configuration must be applied for the LNS-Service before the bind command is executed.
Examples The following command enables single port mode for the current LNS service:
single-port-mode
trap
This command generates SNMP traps.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables SNMP traps.
Usage Guidelines Use this command to enable/disable all supported SNMP traps.
Examples To enable all supported SNMP traps, enter the following command;
trap all
tunnel-authentication
Enables/disables L2TP tunnel authentication for the LNS service.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disables tunnel authentication
Tunnel authentication is enabled by default.
Usage Guidelines When tunnel authentication is enabled, a configured shared secret is used to ensure that the LNS service is
communicating with an authorized peer LAC. The shared secret is configured by the peer-lac command, the
tunnel l2tp command in the Subscriber Configuration mode, or the Tunnel-Password attribute in the
subscribers RADIUS profile.
tunnel-switching
Enables or disables the LNS service from creating tunnels to another LAC for an existing tunnel.
Product PDSN
GGSN
Command Modes Exec > Global Configuration > Context Configuration > LNS Service Configuration
configure > context context_name > lns-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-lns-service)#
no
Disable tunnel switching.
Tunnel switching is enabled by default.
Usage Guidelines Tunnel switching is when the LNS has a tunnel connected to a LAC and creates a tunnel to a different LAC
and routes the data from the original LAC through the new tunnel to the other LAC.
Examples To disable tunnel switching in the LNS, enter the following command;
no tunnel-switching
The Local Policy Actiondef Configuration Mode is used to define the action definitions to be used for local
QoS policies.
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Actiondef Configuration
configure > local-policy-service service_name > actiondef actiondef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-actiondef)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
action
This command configures the action priority for an actiondef.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Actiondef Configuration
configure > local-policy-service service_name > actiondef actiondef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-actiondef)#
priority priority
Specifies a priority for the specified action.
priority must be a unique integer from 1 to 2048.
action_name arguments
The following actions are supported. arguments specify a set of parameters to be used when invoking the
action.
• activate-ambr uplink bandwidth downlink bandwidth
Sets the aggregated maximum bit rate (AMBR) for the APN.
Configures uplink and downlink bandwidth. bandwidth must be an integer from 1 to 1000000000.
• activate-flow-detection { initiation | termination } ruledef ruledef_name
Detects a flow and takes action.
initiation ruledef: Checks for flow initiation and adds a rule definition.
termination ruledef: Checks for flow termination and adds a rule definition.
ruledef_name must be an existing ruledef.
• activate-lp-rule name lprule_name
Activates a local-policy rule within service scheme when a subscriber is in the configured RAI or TAI
range.
lprule_name must be an existing local-policy rule within the service scheme expressed as an alphanumeric
string of 1 through 63 characters.
When the subscriber moves out of the configured RAI or TAI range, the local-policy rule is deactivated.
This option is added as part of Location Based QoS Override feature. For more information on this
feature, see the ECS Administration Guide.
• activate-rule name rule_name
Activates a rule within ECS rulebase for a subscriber.
rule_name must be an existing rule within this local QoS policy service expressed as an alphanumeric
string of 1 through 63 characters.
• activate-rulebase name rulebase_name
Associates the session with a specific rulebase.
rulebase_name must be an existing rulebase within this local QoS policy service expressed as an
alphanumeric string of 1 through 63 characters.
• allow-requested-qos
Allow a specific UE initiated QoS request.
• allow-session
Allows the session to continue.
• deactivate-flow-detection { initiation | termination } ruledef ruledef_name
Deactivates detection of flow and takes action.
initiation ruledef: Checks for flow initiation and adds a rule definition.
termination ruledef: Checks for flow termination and adds a rule definition.
ruledef_name must be an existing ruledef.
• deactivate-rule name rule_name
Deactivates a rule within ECS.
rule_name must be an existing rule within this local QoS policy service expressed as an alphanumeric
string of 1 through 63 characters.
• deactivate-rulebase name rulebase_name
Disassociates the rulebase from a session.
rulebase_name must be an existing rulebase within this local QoS policy service expressed as an
alphanumeric string of 1 through 63 characters.
• default-qos qci value arp value
Sets the default QoS parameters for the session
qci value must be an integer from 1 through 254.
arp value must be an integer from 1 through 15 (StarOS v12.1 and earlier) or 1 through 127 (StarOS
v12.2 and later).
• event-triggers { default-bearer-qos-change | qos-change | tai-change | uli-change }
This action specifies to enable the event triggers – Default EPS bearer QoS change event trigger and
QoS change event trigger.
The TAI-Change and ULI-Change event triggers are added as part of Location Based QoS Override
feature. For more information on this feature, see the ECS Administration Guide.
• reconnect-to-server [ send-usage-report ]
Reconnects to the PCRF server to handle fallback scenario. That is, when the session falls back to local
policy, this action specifies to retry connecting to the PCRF server.
send-usage-report: Triggers CCR-U with volume report immediately. The default behavior is that the
CCR-U will not be triggered immediately.
On timer-expiry, if the initial failure is due to CCR-U failure, and if send-ccru-immediate is configured,
then CCR-U will be sent with the usage report immediately.
• reject-requested-qos
Rejects UE QoS resource request.
• retry-count value
Retry action. This applies to start-timer/activate-rule/activate-ruledef.
value must be an integer from 0 through 65535.
• start-timer name duration value retry-count value
Starts a named timer. On expiry of this timer, the local policy engine is contacted to initiate the appropriate
action, such as termination of a session.
duration value: Enter a timer duration from 0 through 28800 seconds. A value of 0 can be used to leave
the local policy until the subscriber disconnects. Default timer value is 14400 (seconds).
retry-count specifies the maximum number of times the server will be retried before terminating the
call.
retry-count value must be an integer from 0 through 65535. Default retry count is 3.
• stop-timer name
Stops the designated timer.
• terminate-session
Terminates the session.
Usage Guidelines Use this command to enable the setting of parameters to be used when invoking actions. Actions are a series
of operations that are triggered by activated rules.
This command can be entered multiple times to configure multiple actions for an actiondef. The actions are
examined in priority order until a match is found and the corresponding action is applied.
Examples The following command creates an action to allow a session to continue with priority set to 125:
action priority 125 allow-session
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
The Local Policy Eventbase Configuration Mode is used to configure the events to be used for local QoS
policies.
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Eventbase Configuration
configure > local-policy-service service_name > eventbase eventbase_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-eventbase)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
rule
This command enables the setting of event rules. An event is something that occurs in the system which would
trigger a set of actions to take place, such as new-call or rat-change.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Eventbase Configuration
configure > local-policy-service service_name > eventbase eventbase_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-eventbase)#
Syntax Description rule priority priority [ event list_of_events ] ruledef ruledef_name actiondef actiondef_name [ continue ]
no rule priority priority
priority priority
Specifies a priority for the specified rule.
priority must be unique and an integer from 1 to 2048.
event list_of_events
The event is defined by any of the following events. Upon triggering the event, the rules specified in the
eventbase are executed.
• apn-ambr-mod-failure: This event is triggered as a result of a APN AMBR Modification failure.
• def-eps-bearer-qos-mod-failure: This event is triggered as a result of a Default EPS bearer QoS
Modification failure.
• default-qos-change: This event is triggered as a result of a default QoS change.
• fallback: This event is triggered as a result of fallback from PCRF.
• location-change: This event is triggered as a result of any change relating to location.
• new-call: This event is initiated when a new call is established.
• out-of-credit: This event is initiated on out of OCS credit.
ruledef ruledef_name
Associates the rule with a specific ruledef.
ruledef_name must be an existing ruledef within this local QoS policy service.
actiondef actiondef_name
Associates the rule with a specific actiondef.
actiondef_name must be an existing actiondef within this local QoS policy service expressed as an alphanumeric
string of 1 through 63 characters.
continue
Subsequent rules are also matched; otherwise, rule evaluation is terminated on first match.
Usage Guidelines Use this command to create, configure, or delete event rules.
The rules are executed in priority order, and if the rule is matched the action specified in the actiondef is
executed. If an event qualifier is associated with a rule, the rule is matched only for that specific event. If a
qualifier of continue is present at the end of the rule, the subsequent rules are also matched; otherwise, rule
evaluation is terminated on first match.
This command can be entered multiple times to configure multiple rules for an eventbase.
Important A maximum of 256 rules are suggested in an eventbase for performance reasons.
Examples The following command creates a rule with priority set to 2 and associated with ruledef rule5 and actiondef
action7:
rule priority 2 ruledef rule5 actiondef action7
The Local Policy Ruledef Configuration Mode is used to configure the rule definitions to be used for local
QoS policies.
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Ruledef Configuration
configure > local-policy-service service_name > ruledef ruledef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-ruledef)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
condition
This command is used to configure the conditions which trigger the ruledef event.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration > Local Policy Ruledef Configuration
configure > local-policy-service service_name > ruledef ruledef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-ruledef)#
Syntax Description condition priority priority { variable { eq | ge | gt | le | lt | match | ne | nomatch } regex | string_value |
int_value | set }
no condition priority priority
priority priority
Specifies a priority for the specified condition.
priority must be unique and an integer from 1 to 2048.
variable
The following variables are supported:
• 3g-uli mcc mcc_num mnc mnc_num tac
Configures 3G-ULI parameter with values for MCC, MNC and LAC. Operator takes specific action or
applies local-policy rule based on the 3G-ULI value in Change event notification from MME.
◦mcc mcc_num : MCC is a three digit number from 001 to 999. It is a string of size 3 to 3.
◦mnc mnc_num : MNC is a two or three digit number from 01 to 999. It is a string of size 2 to 3.
◦lac: LAC is a 4 byte field. It is a string of 4 hexadecimal values from 0x1 to 0xffff.
• apn
The APN associated with the current session expressed as an alphanumeric string of 1 through 63
characters.
• arp
The ARP value associated with the current session expressed as an integer from 1 through 15.
• bandwidth
Total bandwidth associated with the QCI and ARP value associated with the request, expressed as an
integer from 0 through 1000000000.
• bsid
Base Station Identifier associated with the subscriber expressed as an alphanumeric string of 1 through
63 characters.
• cause-code
Failure Cause Code associated with the subscriber expressed as an alphanumeric string of 1 through 63
characters.
• date
Date value to match. <Clock in format YYYY:MM:DD>
• day-of-month
The day of the month to match the rule to, expressed as an integer from 1 through 31.
• day-of-week
Sunday...Saturday, expressed as an integer from 1 to 7.
• final-unit-action { redirect | restrict-access | terminate } [ filter-id ] [ eq | ge | gt | le | lt | match | ne
| nomatch ] filter-id
This variable allows configuring different filter IDs and different Final-Unit-Action (FUA) actions for
the events like out-of-credit, etc. Based on the FUA and filter ID values, local policy engine will either
install pre-configured redirection rules/pre-configured rule that might drop all packets, or push a different
rule/policy.
When the FUA received from the session manager during out-of-credit scenario matches with the
configured FUA, then one of the following actions will be taken. If multiple filter-ids are configured,
then at least one filter-id should be matched.
◦redirect: Redirects the service
◦restrict-access : Restricts the service
◦terminate: Terminates the service
filter-id: This variable denotes the name of the filter list for the user. filter-id is a string of 1 through 128
characters. Note that match, nomatch, ne, and eq are more appropriate operators though other values
can also be used. Wild card values can be specified for string match.
Important This feature of supporting FUA in local policy will be active only when Gx Assume
Positive is active.
• imeisv
IMEISV of the user equipment expressed as an alphanumeric string of 1 through 63 characters.
• imsi
IMSI associated with the subscriber expressed as an alphanumeric string of 1 through 63 characters.
• local-policy-mode [ fallback | dual-mode | lp-only ]
This variable allows selecting different actions for different modes like local-policy only, dual-mode,
and fallback mode for the same event.
• fallback: This mode indicates that the action has to be taken only when the call is with local-policy
because of failure-handling.
• dual-mode: This mode indicates that action has to be taken if the call is in dual-mode wherein
both PCRF and local-policy co-exist.
• lp-only: This mode indicates that action has to be taken when only local-policy exists and PCRF
does not.
• meid
MEID associated with the subscriber expressed as an alphanumeric string of 1 through 63 characters.
• month-of-year
Jan, Feb....Dec, expressed as an integer from 1 through 12.
• msisdn
MSISDN associated with the session expressed as an alphanumeric string of 1 through 63 characters.
• nai
NAI associated with the session expressed as an alphanumeric string of 1 through 63 characters.
• pdn-type
Type of PDNs associated with the same APN.
• IPV4: IPv4 PDN Type
• IPV4V6: IPv4v6 PDN Type
• IPV6: IPv6 PDN Type
• qci
QCI associated with the current event expressed as an integer from 1 through 254.
• radio-access-technology
Radio access technology associated with the subscriber:
• cdma-1xrtt: CDMA 1X RTT radio access technology
• cdma-evdo: CDMA-EVDO radio access technology
• cdma-evdo-reva: CDMA EVDO REVA radio access technology
• cdma-other: Other CDMA radio access technologies
• ehrpd: EHRPD radio access technology
• eutran: EUTRAN radio access technology
• gan: GAN radio access technology
• serving-node-address
IP address associated with the current node serving the subscriber entered using IPv4 dotted-decimal or
IPv6 colon-separated-hexadecimal notation.
• serving-plmn
PLMN associated with the current node serving the subscriber expressed as an alphanumeric string of
1 through 63 characters.
• tai mcc mcc_num mnc mnc_num tac
Configures Tracking Area Identification associated with the subscriber. Operator takes specific action
or applies local-policy rule based on the TAI value in TAI-Change event notification from MME.
◦mcc mcc_num : MCC is a three digit number from 001 to 999. It is a string of size 3 to 3.
◦mnc mnc_num : MNC is a two or three digit number from 01 to 999. It is a string of size 2 to 3.
◦tac: TAC is a 4 byte field. It is a string of 4 hexadecimal values from 0x1 to 0xffff.
• time-of-day
Time associated with the change. <Clock in format HH:mm:ss or HH:mm >
eq | ge | gt | le | lt | match | ne | nomatch
eq: Operation equal to
ge: Operation greater than or equal to
gt: Operation greater than
le: Operation less than or equal to
lt: Operation less than
match: Operation match
ne: Operation not equal to
nomatch: Operation nomatch
Usage Guidelines Use this command to configure the conditions which trigger the ruledef event. A ruledef represents a set of
matching conditions.
This command can be entered multiple times to configure multiple conditions for a ruledef. The conditions
are examined in priority order until a match is found and the corresponding condition is applied.
Examples The following command creates a condition with priority set to 5 and configured match apn myapn*:
condition priority 5 apn match myapn*
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
The Local Policy Service Configuration Mode is used to configure the local QoS policy for one or more
services.
Command Modes Exec > Global Configuration > Local Policy Service Configuration
configure > local-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
actiondef
This command enables creating, configuring, or deleting action definitions for an event.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration
configure > local-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-service)#
no
Deletes the specified actiondef from the local QoS policy service.
actiondef_name
Specifies name of the actiondef.
actiondef_name must be unique within the service expressed as an alphanumeric string of 1 through 63
characters.
If the named actiondef does not exist, it is created, and the CLI mode changes to the Local Policy Actiondef
Configuration Mode wherein the actiondef can be configured.
If the named actiondef already exists, the CLI mode changes to the Local Policy Actiondef Configuration
Mode for that actiondef.
-noconfirm
Specifies that the command must execute without prompting for confirmation.
Usage Guidelines Use this command to create, configure, or delete an actiondef. The actiondef configuration is used to configure
the action definitions for an event. The event ruledef will have one or more rules and associated action(s).
This command can be entered multiple times to specify multiple actiondefs.
Important A maximum of 256 actiondefs are suggested in a local QoS policy service for performance reasons. An
actiondef can be referenced by multiple eventbases.
Local Policy Actiondef Configuration Mode commands are defined in the Local Policy Actiondef Configuration
Mode Commands chapter.
Examples The following command creates an actiondef named actiondef1 and enters the Local Policy Actiondef
Configuration Mode:
actiondef actiondef1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
eventbase
This command enables creating, configuring, or deleting an eventbase.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration
configure > local-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-service)#
no
Deletes the specified eventbase from the local QoS policy service.
eventbase_name
Specifies name of the eventbase.
eventbase_name must be unique within the service expressed as an alphanumeric string of 1 through 63
characters.
Important Currently, only one eventbase is supported, and it must be named "default".
If the named eventbase does not exist, it is created, and the CLI mode changes to the Local Policy Eventbase
Configuration Mode wherein the eventbase can be configured.
If the named eventbase already exists, the CLI mode changes to the Local Policy Eventbase Configuration
Mode for that eventbase.
-noconfirm
Specifies that the command must execute without prompting for confirmation.
Local Policy Eventbase Configuration Mode commands are defined in the Local Policy Eventbase Configuration
Mode Commands chapter.
Examples The following command creates an eventbase named default and enters the Local Policy Eventbase
Configuration Mode:
eventbase default
ruledef
This command enables creating, configuring, or deleting a rule definition.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Local Policy Service Configuration
configure > local-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-service)#
no
Deletes the specified ruledef from the local QoS policy service.
ruledef_name
Specifies name of the ruledef.
ruledef_name must be unique within the service expressed as an alphanumeric string of 1 through 63 characters.
If the named ruledef does not exist, it is created, and the CLI mode changes to the Local Policy Ruledef
Configuration Mode wherein the ruledef can be configured.
If the named ruledef already exists, the CLI mode changes to the Local Policy Ruledef Configuration Mode
for that ruledef.
-noconfirm
Specifies that the command must execute without prompting for confirmation.
Usage Guidelines Use this command to create, configure, or delete a ruledef. A ruledef represents a set of matching conditions.
This command can be entered multiple times to specify multiple ruledefs.
Important A maximum 256 ruledefs are suggested in a local QoS policy service for performance reasons.
Local Policy Ruledef Configuration Mode commands are defined in the Local Policy Ruledef Configuration
Mode Commands chapter.
Examples The following command creates a ruledef named rule5 and enters the Local Policy Ruledef Configuration
Mode:
ruledef rule5
suppress-cra
This command allows to suppress the Change Reporting Action (CRA) for event triggers enabled in local
policy configurations.
Product GGSN
P-GW
Command Modes Exec > Global Configuration > Local Policy Service Configuration
configure > local-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-local-policy-service)#
no
This variant is used to configure the default behavior. By default, the CRA notification is sent to MME if one
or a combination of these event triggers is installed.
Usage Guidelines Use this command to control the CRA notification towards MME based on the configured event triggers in
the local-policy configuration.
Examples The following command suppresses CRA if ECGI-Change event trigger is installed:
suppress-cra event-triggers ecgi-change
Command Modes Exec > Global Configuration > Context Configuration > Location Service Configuration
configure > context context_name > location-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-location-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate
Associates or disassociates supportive interfaces or services with this location service instance.
Product MME
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > Location Service Configuration
configure > context context_name > location-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-location-service)#
Syntax Description associate { diameter { dictionary standard | endpoint } | map-service map_svc_name | sls-service
sls_svc_name }
default associate diameter dictionary
no associate { diameter endpoint | map-service | sls-service }
default
Returns the command to its default value of 'standard'.
no
Disassociates a previously associated interface or service with this location service.
map-service map_svc_name
Associates a Mobile Application Part (MAP) service with this location service.
This keyword is applicable for SGSN only.
map_sv_svc_name specifies the name for a pre-configured MAP service to associate with this location service.
sls-service sls_svc_name
Associates an SLs service with this location service. The SLs service provides an interface between the MME
and Evolved Serving Mobile Location Center (E-SMLC).
This keyword is applicable for MME only.
sls_svc_name specifies the name for a pre-configured SLs service to associate with this location service.
Usage Guidelines Use this command to specify the Diameter dictionary and endpoint to be used for this location service, or
associate supportive services with this location service.
The location service provides SLg (MME) interface support or Lg (SGSN) interface support via the Diameter
protocol between the MME or SGSN and the GLMC.
Examples The following command associates a pre-configured Diameter endpoint named test12 to this location service:
associate diameter endpoint test12
destination-host
Configures the host name of the GLMC to be used for this Location service. When defined, this host name
is populated in the destination-host AVP.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > Location Service Configuration
configure > context context_name > location-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-location-service)#
no
Removes the configured destination host.
destination_host
Defines the host name to be used for be a string from 1 to 63 characters.
Usage Guidelines To comply with 3GPP TS29.172, the Destination-Host AVP is sent to the GMLC for all the Location Report
Request (LRR) messages initiated by MME.
Use this command to configure the destination-host AVP for this Location service.
If this command is not configured, the peer host name configured in the diameter endpoint is encoded as
destination-host AVP. Refer to the peer command in the Diameter Endpoint Configuration Mode.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
timeout
Configures the timers used to control various location service procedures.
Product SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > Location Service Configuration
configure > context context_name > location-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-location-service)#
default
Resets the specified timer to the default value.
area-event-invoke-timer aietimer_seconds
This timer, set in seconds, is used to guard the area event invoke procedure.
aietimer_seconds is an integer from 10 through 20. Default is 15.
lcsn lcns_seconds
Sets the NAS location service notification timer defining how long the SGSN will wait (in seconds) before
aborting the Location Service Request, and release all resources allocated for the transaction.
lcns_seconds is an integer from 10 through 20. Default is 15.
periodic-event-invoke-timer peitimer_seconds
Thi timer, set in seconds, is used to guard the period location invoke procedure.
peitimer_seconds is an integer from 10 to 20. Default is 15.
ue-available-guard-timer ueagtimer_seconds
This timer, set in seconds, is used to guard the packet-switched deferred location request (UE available event)
procedures.
Usage Guidelines Use this command to set the amount of time the SGSN waits to perform various location service procedures.
Examples The following command can be used to set the time the SGSN will wait, for example 12 seconds, before
aborting the Location Service Request:
timeout lcns 12
The following command can be used to set the timeout for the UE available guard timer to 460 seconds:
timeout ue-available-guard-timer 460
Important In Release 20.0, HeNBGW is not supported. Commands in this configuration mode must not be used in
Release 20.0. For more information, contact your Cisco account representative.
The Logical eNodeB configuration option enables the configuration of one or more logical eNodeBs within
the HeNB-GW. The Logical eNodeB configuration can be used to support load balancing within a pool of
TAIs.
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate mme-pool
Associates a previously configured MME pool to this logical eNodeB. An MME pool must be configured in
LTE Policy Configuration mode before using this configuration.
Product HeNB-GW
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
no
Removes the associated MME pool from this logical eNodeB configuration.
pool_name
Identifies the name of the pre-configured MME pool to associate with this logical eNodeB.
pool_name is an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to bind/associate a pre-configured MME pool to the this logical eNodeB. The MME pool
can be configured in LTE Policy configuration mode. The associate configuration is used to establish
associations with other helper services in general.
Each logical eNodeB can connect up to 8 MMEs. Since 8 logical eNodeBs can be configured per HeNB-GW
Network service, a total of 64 associations can be established between HeNB-GW and MME.
Examples Following command associates an MME pool named mme_pool with specific logical eNodeB:
associate mme-pool mme_pool
associate tai-list-db
Associates a previously configured TAI database name to this logical eNodeB. A TAI database name for TAI
configuration must be configured in LTE Policy Configuration mode before using this configuration.
Product HeNB-GW
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
no
Removes the associated TAI database from this logical eNodeB configuration.
tai_db_name
Identifies the name of the pre-configured TAI database to associate with this logical eNodeB.
tai_db_name is an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to bind/associate a pre-configured TAI database to the this logical eNodeB. The MME
pool can be configured in LTE Policy configuration mode. The associate configuration is used to establish
associations with other helper services in general.
A maximum number of 8 TAI databases are supported. Each TAI database can accommodate up to 256
configurations of Tracking Area Codes (TACs). Therefore a total of 2048 TACs are supported.
Examples Following command associates a TAI database named henb_tai_db with specific logical eNodeB:
associate tai-list-db henb_tai_db
bind s1-mme
Binds the pre configured Local SCTP IP Address for S1 association to MME.
Product HeNB-GW
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
no
Removes the binding of S1-MME interface from this logical eNodeB configuration.
ip_addr
Identifies the IP address of the S1-MME interface to associate with this HeNB-GW Network service.
addr_val must be entered in the IPv4 ( dotted decimal notation) or IPv6 (: / :: notation) .
Usage Guidelines Use this command to bind the pre configured IPv4 / IPv6 address of the S1-MME interface to the logical
eNodeB.
Examples Following command binds the S1-MME interface having 192:168:100:101IP address with specific logical
eNodeB.
bind s1-mme ipv6-address 192:168:100:101
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
s1-mme ip qos-dscp
This command configures the quality of service (Do's) differentiated service code point (DSCP) marking for
IP packets sent out on the S1-MME interface, from the HeNB-GW to the MME(s).
Product HeNB-GW
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
Syntax Description s1-mme ip qos-dscp { af11 | af12 | af13 | af21 | af22 | af23 | af31 | af32 | af33 | af41 | af42 | af43 | be | cs0
| cs1 | cs2 | cs3 | cs4 | cs5 | cs6 | cs7 | ef }
default s1-mme ip qos-dscp
qos-dscp { af11 | af12 | af13 | af21 | af22 | af23 | af31 | af32 | af33 | af41 | af42 | af43 | be | cs0 | cs1 | cs2 |
cs3 | cs4 | cs5 | cs6 | cs7 | ef }
Default: af11
Specifies the DSCP for the specified QoS traffic pattern. qos-dscp can be configured to any one of the
following:
af11: Assured Forwarding 11 per-hop-behavior (PHB)
af12: Assured Forwarding 12 PHB
af13: Assured Forwarding 13 PHB
af21: Assured Forwarding 21 PHB
af22: Assured Forwarding 22 PHB
af23: Assured Forwarding 23 PHB
af31: Assured Forwarding 31 PHB
af32: Assured Forwarding 32 PHB
af33: Assured Forwarding 33 PHB
af41: Assured Forwarding 41 PHB
Usage Guidelines DSCP levels can be assigned to specific traffic patterns to ensure that data packets are delivered according to
the precedence with which they are tagged. The diffserv markings are applied to the IP header of every
subscriber data packet transmitted over the S1-MME interface(s).
Examples The following command sets the DSCP-level for data traffic sent over the S1-MME interface to af12:
s1-mme ip qos-dscp af12
Product HeNB-GW
Command Modes Exec > Global Configuration > Context Configuration > HENBGW-NETWORK Service Configuration >
Logical eNodeB Configuration
configure > context context_name > henbgw-network-service service_name > logical-enb global-enb-id
plmn id mcc mcc_id mnc mnc_id { home-enb-id henb_id | macro-enb-id menb_id }
Entering the above command sequence results in the following prompt:
[context_name]host_name(logical-enb)#
default
Sets the SCTP port to the default value of 36412 to communicate with the MMEs using S1-MME interface.
port_num
Specifies the SCTP port number to communicate with the HeNBs using S1-MME interface as an integer from
1 through 65535. Default: 36412
Usage Guidelines Use this command to assign the SCTP port with SCTP socket to communicate with the HeNB using S1AP.
Only one SCTP port can be associated with one MME service.
Examples The following command sets the default SCTP port number 699 for to interact with Home eNodeB using
S1AP on S1-MME interface:
default s1-mme sctp port
The Loopback Interface Configuration Mode is used to create and manage an internal IP network address.
The address must be configured with a 32-bit mask.
Command Modes Exec > Global Configuration > Context Configuration > Ethernet Interface Configuration
configure > context context_name > interface interface_name loopback
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-loopback)#
Important Available commands or keywords/variables vary based on platform type, product version, and installed
license(s).
crypto-map
Applies the specified IPSec crypto-map to this interface.
Product All
no
Deletes the application of the crypto map on this interface.
map_name
Specifies the name of the crypto map being applied as an alphanumeric string of 1 through 127 characters
that is case sensitive.
secondary-address sec_ip_addr
Applies the crypto map to the secondary address for this interface. sec_ip_addr must be specified using the
IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal notation.
Usage Guidelines In order for ISAKMP and/or manual crypto maps to work, they must be applied to a specific interface using
this command. Dynamic crypto maps should not be applied to interfaces.
The crypto map must be configured in the same context as the interface.
Examples To apply the IPSec crypto map named cmap1 to this interface, use the following command:
crypto-map cmap1
description
Sets the descriptive text for the current interface.
Product All
no
Clears the description for the interface.
text
Specifies the descriptive text as an alphanumeric string of 0 through 79 characters.
Usage Guidelines Set the description to provide useful information on the interface's primary function, services, end users, etc.
Any information useful may be provided.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip address
Specifies the primary and optional secondary IPv4 addresses and subnets for this interface.
Product All
no
Removes the IPv4 address from this interface.
Important For IPv4 addresses, 31-bit subnet masks are supported per RFC 3021.
secondary ip_address
Configures a secondary IPv4 address on the interface.
Important You must configure the primary IPv4 address before you will be allowed to configure a secondary address.
srp-activate
Activates the IP address for Interchassis Session Recovery (ICSR). Enable this IPv4 address when the Service
Redundancy Protocol (SRP) determines that this chassis is ACTIVE. Requires an ICSR license on the chassis
to activate.
Usage Guidelines The following command specifies the primary IP address and subnets for this interface.
Examples The following example configures an IPv4 address for this interface:
ip address 192.154.3.5/24
ip ranged-address
Specifies an IPv4 address and subnet; all addresses in the subnet are local. Configures the range or group of
IP address for the loopback interface. This command enables support for multiple Enterprise HAs in one HA
service.
Product All
no
Removes the IPv4 ranged address from this interface.
Important This interface configuration is allowed only for IPv4 addresses and must be bound to the HA Service.
Usage Guidelines This command provide Enterprise HA support on HA service for multiple enterprise nodes. Refer HA
Administration Guide for more information.
This IP address range configuration must meet the following criteria:
• The ranged address must be a primary address.
• The ranged address must be unique across the interface configuration.
• The ranged address must be unique across the context.
• The IP address specified in the ranged address must not be part of any other interface.
• The ranged-address can be a n SRP-activated address.
Examples The following example configures a ranged IPv4 address for this interface:
ip ranged-address 192.154.3.5/24
ip vrf
Associates this interface with a specific Virtual Routing and Forwarding (VRF) table.
Product All
no
Removes the specified VRF table from this interface.
vrf_name
Specifies the name of an existing VRF table as an alphanumeric string of 1 through 63 characters.
Use the Context Configuration mode ip vrf forwarding command to preconfigure the VRF name.
Usage Guidelines The following command specifies a ranged IP address for this interface.
Examples The following example elasticities this interface with VRF named vrf_012:
ip vrf forwarding vrf_012
ipv6 address
Specifies an IPv6 address and subnet mask.
Product PDSN
HA
no
Removes the IPv6 address from this interface.
ipv6_address/mask
Specifies an individual host IP address to add to this host pool in IPv6 colon-separated-hexadecimal CIDR
notation.
Important On the ASR 5000, routes with IPv6 prefix lengths less than /12 and between the range of /64 and /128 are
not supported.
Usage Guidelines Configures the IPv6 address and subnet mask for a specific interface.
Examples The following example configures an IPv6 address for this interface:
ipv6 address 2002:0:0:0:0:0:c014:101/128
ipv6 ospf
Enables Open Shortest Path First Version 3 (OSPFv3) functionality on this IPv6 a interface.
Product PDSN
HA
GGSN
Syntax Description [ no ] ipv6 ospf [ area { integer | ipv4-address } | cost cost-value | dead-interval dead-intrv | hello-interval
hello-intrvl | priority p-value | retransmit-interval retx-interval | transmit-delay td-interval ]
no
Removes a previously configured access group association.
cost cost-value
Specifies a link cost as an integer from 1 through 65535. The link cost is carried in the LSA updates for each
link. The cost is an arbitrary number.
dead-interval dead-intrv
Specifies the interval (in seconds) after which a neighbor is declared dead when no hello packets as an integer
from 1 through 65535.
hello-interval hello-intrvl
Specifies the interval (in seconds) between hello packets that OSPFv3 sends on an interface as an integer from
1 through 65535.
priority p-value
Specifies the of the interface as an integer from 0 through 255.
retransmit-interval retx-interval
Specifies the time (in seconds) between link-state advertisement (LSA) retransmissions for adjacencies
belonging to the OSPFv3 interface as an integer from 1 through 65535.
transmit-delay td-interval
Specifies the estimated time (in seconds) required to send a link-state update packet on the interface as an
integer from 1 through 65535.
Examples ipv6 ospf area 334 cost 555 dead-interval 40 hello-interval 10 priority 10 retransmit-interval 5
transmit-delay 10
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> Custom TAI List Configuration
configure > lte-policy > tai-mgmt-db db_name tai-custom-list tac value
Entering the above command sequence results in the following prompt:
[local]hostname{tai-cstm-list}#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
tai
Configures a Tracking Area Identifier (TAI) for this custom TAI list.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> Custom TAI List Configuration
configure > lte-policy > tai-mgmt-db db_name tai-custom-list tac value
Entering the above command sequence results in the following prompt:
[local]hostname{tai-cstm-list}#
no
Removes a configured TAI from the TAI management object.
mcc number
Specifies the mobile country code (MCC) portion of a PLMN identifier as an integer from 100 through 999.
mnc number
Specifies the mobile network code (MNC) portion of a PLMN identifier as a 2- or 3-digit integer from 00
through 999.
tac value+
Specifies the Tracking Area Code portion of the TAI as an integer from 1 through 65535. Up to 16 TAC
values can be entered on a single line.
Usage Guidelines Use this command to configure one or more TAIs for this custom TAI list.
A maximum of 15 TAIs can be configured per Custom TAI List.
A TAC can be added in this custom TAI list only if it has already configured in any of the TAI management
objects within this TAI Management Database.
All the TAIs configured within a Custom TAI List are assumed to use same S-GW, time-zone, zone-code,
and other configurations within the TAI Management Object. If a Custom TAI List includes TAIs from
different objects then those objects should be configured with same S-GW address, time-zone, zone-code,
etc.
If the TAU/Attach comes with a TAI that matches a Custom TAI List, the resulting
ATTACH_ACCEPT/TAU_ACCEPT will include all the TAIs present in Custom TAI List as well as the
received TAI.
If the Custom TAI List is configured without any TAIs, the ATTACH_ACCEPT/TAU_ACCEPT will include
all the TAIs from TAI Management object in which received TAI is present.
Examples The following set of commands show a Custom TAI List with TAC 3024, which includes TACs 3022, 3023,
3025, and 3026:
tai-custom-list tac 3024
tai mcc 311 mnc 480 tac 3022
tai mcc 311 mnc 480 tac 3023
tai mcc 311 mnc 480 tac 3025
tai mcc 311 mnc 480 tac 3026
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
ambr
Configures the maximum aggregated uplink and downlink bitrate values for this profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
no
Removes the aggregated maximum bitrate (AMBR) configuration for this emergency profile.
max-ul bitrate
Configures the maximum aggregated bitrate value for the uplink as an integer from 0 through 1410065408.
max-dl bitrate
Configures the maximum aggregated bitrate value for the downlink as an integer from 0 through 1410065408.
Usage Guidelines Use this command to configure uplink and downlink maximum aggregated bitrate values to be shared across
all non-guaranteed bitrate bearers established for the emergency session.
Examples The following example configures the uplink AMBR value to 2000 bps and the downlink AMBR value to
2000 bps:
ambr max-ul 2000 max-dl 2000
apn
Configures the name and PDN type of the access point name (APN) used for emergency PDN connections.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
Syntax Description apn apn_name pdn-type { ipv4 | ipv4v6 | ipv6 } [ restoration-priority priority_value ]
no apn
no
Removes the selected APN from the profile.
apn_name
apn_name specifies the name of a pre-configured APN profile to associate with this LTE emergency profile.
For more information on APNs, refer to the apn command in the Context Configuration Mode Commands
chapter.
restoration-priority priority_value
Configures the APN restoration priority value for emergency sessions for this APN profile. The reactivation
of emergency PDNs after a P-GW restart notification is processed in the order of this priority.
priority_value
The priority value is an integer value from 1 through 16. Where "1" is the highest priority and "16" is the
lowest priority. Default: 16 (lowest priority).
To define the APN restoration priority for non-emergency sessions, refer to the APN Profile > apn-restoration
command.
Usage Guidelines Use this command to select the APN to be used for emergency bearer services. APNs are configured through
the APN Configuration mode. For more information regarding APN configuration mode commands, refer to
the APN Configuration Mode Commands chapter.
Examples The following command specifies that the APN named apn-3.com is to be used for emergency bearer services
and that the PDN supports IPv4 traffic only:
apn apn-3.com pdn-type ipv4
The following command configures the APN Restoration Priority for APN profile "eap" with restoration
priority value 1.
apn eap pdn-type ipv4 apn-restoration-priority 1
associate
Associates a location service with this LTE emergency profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
no
Disassociates a previously associated location service with this LTE emergency profile.
location-service location_svc_name
Associates a location service with this LTE emergency profile. Only one location service can be associated
with an LTE emergency profile.
location_svc_name specifies the name for a pre-configured location service to associate with the LTE emergency
profile as an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to associate a pre-configured location service with an LTE emergency profile. This enables
the associated location service to provide location information of emergency calls to the GMLC.
For more information about Location Services (LCS), refer to the location-service command in the Context
Configuration Mode Commands chapter as well as the Location Services Configuration Mode Commands
chapter.
Further details can be found in the Location Services chapter of the MME Administration Guide.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
lcs-qos
Configures the required Location service (LCS) Quality of Service (QoS) settings for this emergency profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
no
Removes the configured LCS QoS settings for this emergency profile.
horizontal-accuracy variable
Defines the horizontal (longitude and latitude) accuracy of the LCS request.
variable must be entered as an integer from 0 to 127, where 0 is the most accurate.
vertical-accuracy variable
Defines the vertical (altitude) accuracy of the LCS request.
variable must be entered as an integer from 0 to 127, where 0 is the most accurate.
Usage Guidelines Use this command to define the location service QoS settings to be used for this emergency profile.
Configuration of these settings is optional.
For Emergency Services, the MME will always set the Response Time to Low Delay. If QoS is configured,
the horizontal accuracy is mandatory. If a vertical accuracy is specified in this command, the MME will set
the Vertical Requested flag.
Refer to 3GPP TS 29.171 and 3GPP TS 23.032 for more details about these settings.
Examples The following command sets the LCS QoS horizontal accuracy to 20, which represents an accuracy of 57.3
meters. No vertical accuracy is specified.
lcs-qos horizontal-accuracy 20
local-emergency-num
This command configures Local Emergency Numbers to be sent in Attach/TAU responses.
Product MME
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
no
Removes the specified Local Emergency Number from the list.
emergency_number
The emergency number is a number assigned to a type of emergency number (ambulance, marine, and so on)
with a string of size 1 to 10.
custom_number
Is specific to the custom local emergency number . custom_number is an hexadecimal number from 0x1 to
0xFF
Usage Guidelines This command allows the subscriber to download a list of local emergency numbers used by the serving
network. This list is downloaded by the network to the User Equipment (UE) at successful registration as well
as subsequent registration updates.
Examples The following configuration allows the operator to assign a emergency number for ambulance:
local-emergency-num 123 ambulance
The following configuration allows the operator to remove the emergency number assigned for ambulance:
no local-emergency-num 123 ambulance
local-emergency-num-ie
This command is used to configure local emergency numbers to be sent only over TAU messages.
Product MME
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
inter-mme-tau
The local-emergency-num-ie keyword with inter-mme-tau option allows the configured local emergency
number list to be sent in a TAU Accept during Inter-MME-TAUs, that is, when the UE switches from a 2G
network to 4G network, from a 3G network to 4G network or from a 4G network to 4G network handover
(for both idle and connected mode).
tau
The local-emergency-num-ie keyword with tau option allows the configured local emergency number list to
be sent in a TAU Accept message during all TAUs (for example, periodic TAUs and so on).
Usage Guidelines This command configuration of the local emergency numbers to be sent only over TAU messages.
Examples The following configuration allows the operator to send the emergency number list over Inter-MME-TAU
messages:
local-emergency-num-ie inter-mme-tau
The following configuration allows the operator to send the emergency number list over all TAU messages:
local-emergency-num-ie tau
The following command removes the configuration of local emergency numbers sent over TAU messages:
no local-emergency-num-ie
pgw
Statically configures a P-GW to support emergency bearer services for this profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
no fqdn
Removes the specified P-GW Fully Qualified Domain Name (FQDN) or IP address from this profile.
fqdn fqdn
Specifies the domain name of the P-GW as an alphanumeric string of 1 through 256 characters.
ip-address address
Specifies the IP address for the P-GW in IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal notation.
Important A maximum of four P-GW IP address configurations are allowed per profile.
• both: Specifies that both the GTP and PMIP protocols are supported.
• gtp: Specifies that only GTP is supported.
• pmip: Specifies that only PMIP is supported.
weight value
Assigns a weight to P-GW IP address to use as a preferred P-GW as an integer from 1 through 100. Lowest
value has the least preference.
Usage Guidelines Use this command to statically configure the P-GW used to support emergency bearer services through this
profile.
Examples The following command configures the P-GW supporting emergency bearer services for this profile as
pdn-911.gov:
pgw fqdn P-GW FQDN
The following command configures the P-GW supporting emergency bearer services for this profile as having
an IPv4 address of 10.2.3.4, supporting GTP only, and having a weight of 10:
pgw ip-address 10.2.3.4 protocol gtp weight 10
qos
Configures the quality of service (QoS) parameters for the emergency bearer service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
Syntax Description qos qci qci arp arp_value preemption-capability { may | shall-not } vulnerability { not-preemptable |
preemtable }
no qos
qci qci
Specifies the QoS Class Identifier (QCI) for the emergency bearer profile as an integer from 0 through 255.
arp arp_value
Defines the address retention priority value as an integer from 1 through 15.
Usage Guidelines Use this command to set the QoS ARP and QCI parameters for the emergency bearer configuration.
Examples The following command sets the QCI number to 7, the ARP value to 2 the preemption capability to "may",
and the vulnerability flag to "pre-emptable":
qos qci 7 arp 2 preemption-capability may vulnerability preemptable
ue-validation-level
Configures the type of user equipment (UE) that can use the emergency bearer service through the profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Emergency Profile Configuration
configure > lte-policy > lte-emergency-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(lte-emergency-profile)#
default
Returns the command to its default setting of "none".
Usage Guidelines Use this command to indicate which UEs can use the emergency bearer service through this profile.
Examples The following command configures the type of UE allowed to use the emergency bearer service to "IMSI
required, authentication optional":
ue-validation-level imsi
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration > LTE
Forbidden Location Area Configuration
configure > lte-policy > ho-restrict-list list_name > forbidden location-area plmnid plmn_id
Entering the above command sequence results in the following prompt:
[local]host_name(forbidden_la)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
lac
Configures a 3G location area code (LAC) or area codes where a UE, associated with this LTE policy, is
restricted from participating in a handover scenario.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration > LTE
Forbidden Location Area Configuration
configure > lte-policy > ho-restrict-list list_name > forbidden location-area plmnid plmn_id
Entering the above command sequence results in the following prompt:
[local]host_name(forbidden_la)#
no
Removes a configured forbidden handover area code or area codes from this policy. If no location area code
is specified, then all location area codes are removed.
area_code
Specifies an area code or area codes from which UEs are restricted from participating in a handover as an
integer from 0 through 65535. Multiple area codes can be entered (up to 128 in a single line, separated by
spaces).
Usage Guidelines Use this command to configure 3G location-based area codes that will be forbidden to UEs associated with
this LTE policy.
Examples The following command configures eight location-based area codes (1, 2, 3, 4, 5, 6, 7, 8) where a UE, associated
with this LTE policy, is restricted from participating in a handover scenario:
lac 1 2 3 4 5 6 7 8
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration > LTE
Forbidden Tracking Area Configuration
configure > lte-policy > ho-restrict-list list_name > forbidden tracking-area plmnid plmn_id
Entering the above command sequence results in the following prompt:
[local]host_name(forbidden_ta)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
tac
Configures a tracking area code (TAC) or area codes where a UE, associated with this LTE policy, is restricted
from participating in a handover scenario.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration > LTE
Forbidden Tracking Area Configuration
configure > lte-policy > ho-restrict-list list_name > forbidden tracking-area plmnid plmn_id
Entering the above command sequence results in the following prompt:
[local]host_name(forbidden_ta)#
no
Removes a configured forbidden handover area code or area codes from this policy. If no tracking area code
is specified, then all tracking area codes are removed.
area_code
Specifies a tracking area code or area codes from which UEs are restricted from participating in a handover
as an integer from 0 to 65535. Multiple area codes can be entered (up to 128 in a single line, separated by
spaces).
Usage Guidelines Use this command to configure tracking area codes that will be forbidden to UEs associated with this LTE
policy.
Examples The following command configures two tracking area codes (1, 2, 3, 4, 5, 6, 7, 8) where a UE, associated with
this LTE policy, is restricted from participating in a handover scenario:
tac 1 2 3 4 5 6 7 8
Command Modes Exec > Global Configuration > LTE Policy Configuration > Foreign PLMN GUTI Management Database
configure > lte-policy > foreign-plmn-guti-mgmt-db
Entering the above command sequence results in the following prompt:
[local]host_name(foreign-plmn-guti-mgmt-db)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
plmn
Configures a foreign Public Land Mobile Network (PLMN) entry in the Foreign PLMN GUTI management
database. This optional configuration is used to control the acceptance or immediate reject of Attach Requests
and TAU Requests containing a GUTI from this PLMN.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > Foreign PLMN GUTI Management Database
configure > lte-policy > foreign-plmn-guti-mgmt-db
Entering the above command sequence results in the following prompt:
[local]host_name(foreign-plmn-guti-mgmt-db)#
Syntax Description plmn mcc { mcc_value | any } mnc { mnc_value | any } [ allow | reject ]
plmn mcc mcc_value mnc mnc_value
no
Removes the specified PLMN entry from the Foreign PLMN GUTI management database.
allow
Configures the MME to allow foreign GUTIs from this PLMN.
reject
Configures the MME to reject foreign GUTIs from this PLMN.
Usage Guidelines Use this command to create and configures a foreign Public Land Mobile Network (PLMN) entry in the
Foreign PLMN GUTI management database. This optional configuration is used to control the acceptance or
immediate reject of Attach Requests and TAU Requests containing a GUTI from this PLMN.
If the configured action is Reject, the MME takes the following actions:
• Attach Requests: a NAS Identity Request is sent to the UE to determine its IMSI, and no DNS lookup
is performed to find a peer MME or SGSN.
• TAU Requests: a TAU Reject message is sent immediately with cause code 9 (UE Identity cannot be
derived by the network), and no DNS lookup is performed to find a peer MME or SGSN.
If the configured action is Allow, the MME continues processing the Attach Request or TAU Request, and a
DNS request may be made.
A maximum of 16 foreign PLMN entries can be added to a Foreign PLMN GUTI management database.
Examples The following command creates a plmn entry in the foreign PLMN GUTI management database. The entry
specifies that GUTIs from PLMNs with the MCC of 123 and any MNC be rejected.
plmn mcc 123 mnc any reject
Important In Release 20.0, HeNBGW is not supported. Commands in this configuration mode must not be used in
Release 20.0. For more information, contact your Cisco account representative.
The MME Pool configuration is used to configure one or more MMEs to which the HeNB-GW is to
communicate. This configuration is available under lte-policy configuration mode. Adding or modifying an
MME pool instance puts the user into the MME Pool configuration mode.
Command Modes Exec > Global Configuration > LTE Policy Configuration > HeNBGW MME Pool Configuration
configure > lte-policy > henbgw mme-pool mme_pool_name
Entering the above command sequence results in the following prompt:
[local]host_name(mme-pool)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
mme
Configures a specific MME to HeNBGW.
Product HeNB-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > HeNBGW MME Pool Configuration
configure > lte-policy > henbgw mme-pool mme_pool_name
Entering the above command sequence results in the following prompt:
[local]host_name(mme-pool)#
no
Removes the configured MME to HeNBGW.
mme_name
It is the MMe name, a string of size 1 through 63.
ipv4-address ipv4_address
Specifies the remote SCTP IP Address for S1 assocation to MME. This must be followed by ipv4_address,
the IPv4 address using dotted-decimal notation.
ipv6-address ipv6_address
Specifies the remote SCTP IP Address for S1 assocation to MME. This must be followed by ipv6_address,
the IPv6 address using ( : / :: ).
sctp
Configure the S1-MME SCTP parameters.
port port_val
Designates SCTP port.
Examples The following command configures the mme with name my_mme, ipv4-address 123.456.234.444 with SCTP
port 302.
mme my_mme ipv4-address 123.456.234.444 sctp port 302
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration
configure > lte-policy > ho-restrict-list list_name
Entering the above command sequence results in the following prompt:
[local]host_name(ho-restrict-list)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
forbidden
Configures the handover restriction lists provided to eNodeBs where handover restrictions are enforced for
UEs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE HO Restriction List Configuration
configure > lte-policy > ho-restrict-list list_name
Entering the above command sequence results in the following prompt:
[local]host_name(ho-restrict-list)#
Syntax Description [ no ] forbidden { inter-rat { all | cdma2000 | geran | utran } | location-area plmnid id | tracking-area
plmnid id }
default forbidden inter-rat
no
Removes the forbidden configuration from the LTE policy.
location-area plmnid id
Specifies that handovers to 3G location area codes defined through this keyword and subsequent configuration
mode are to be prohibited for UEs associated with the LTE policy. Enters the LTE Forbidden Location Area
Configuration Mode. id must be a valid PLMN ID expressed as an integer comprised of an MCC (Mobile
Country Code) and MNC (Mobile Network Code) [five-digit minimum, six-digit maximum].
Important Up to 16 forbidden location area PLMN IDs can be configured per handover restriction list.
LTE Forbidden Location Area Configuration Mode commands are defined in the LTE Forbidden Location
Area Configuration Mode Commands chapter.
tracking-area plmnid id
Specifies that handovers to 4G tracking area codes defined through this keyword and subsequent configuration
mode are to be prohibited for UEs associated with the LTE policy. Enters the LTE Forbidden Tracking Area
Configuration Mode. id must be a valid PLMN ID and be an integer value comprising an MCC and MNC
(five-digit minimum, six-digit maximum).
Important Up to 16 forbidden tracking area PLMN IDs can be configured per handover restriction list.
LTE Forbidden Tracking Area Configuration Mode commands are defined in the LTE Forbidden Tracking
Area Configuration Mode Commands chapter.
Usage Guidelines Use this command to create the list of restricted handover types that apply to all UEs associated with the LTE
policy.
Examples The following command prohibits UEs associated with this LTE policy from participating in a handover to
a GERAN network type:
forbidden inter-rat geran
The following command prohibits UEs, associated with this LTE policy and a mobile network with a PLMN
ID of 12345, from participating in a handover to location area codes defined in the Location Area Configuration
Mode:
forbidden location-area plmnid 12345
Important In Release 20.0, HNBGW is not supported. Commands in this configuration mode must not be used in
Release 20.0. For more information, contact your Cisco account representative.
The LTE MME HeNB-GW Management Database Mode is used to create and manage a list of HeNB-GWs.
The HeNB-GWs defined in this database are used by the MME during S1-based handovers to Home eNodeBs
when the HeNBs are connected to the MME via HeNB-GWs.
Command Modes Exec > Global Configuration > LTE Policy Configuration > MME HeNB-GW Management Database
configure > lte-policy > mme henbgw mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(henbgw-mgmt-db)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
henbgw-global-enbid
This command configures the Global eNodeB Id and TAI of a Home eNodeB within the HeNB-GW
management database.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > MME HeNB-GW Management Database
configure > lte-policy > mme henbgw mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(henbgw-mgmt-db)#
no
Removes a configured entry in the MME HeNB-GW management database.
mcc number
Specifies the mobile country code (MCC) portion of a PLMN identifier as an integer from 100 through 999.
mnc number
Specifies the mobile network code (MNC) portion of a PLMN identifier as a 2- or 3-digit integer from 00
through 999.
enbid number
Specifies the Global eNodeB ID for this HeNB-GW as an integer value from 1 through 1048575.
Usage Guidelines Use this command to configure the global eNodeB ID and TAI of one or more HeNB-GWs within the
HeNB-GW management database.
A maximum of 8 HeNB-GWs can be configured within the HeNB-GW management database.
Examples This following command configures the Global eNodeB ID and TAI for an HeNB-GW entry within the
HeNB-GW management database:
henbgw-global-enbid mcc 123 mnc 456 enbid 789
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Network Global MME ID Management
Database Configuration
configure > lte-policy > network-global-mme-id-mgmt-db
Entering the above command sequence results in the following prompt:
[local]host_name(network-global-mme-id-mgmt-db)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
plmn
Configures associations between public land mobile network (PLMN) IDs and ranges of MME group IDs.
On the S4-SGSN, this command allows the operators to configure a custom list of MME group IDs if networks
have been configured with LACs in the 32768-65535 range for UMTS and GPRS coverage.
Product MME
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Network Global MME ID Management
Database Configuration
configure > lte-policy > network-global-mme-id-mgmt-db
Entering the above command sequence results in the following prompt:
[local]host_name(network-global-mme-id-mgmt-db)#
Syntax Description [ no ] plmn mcc mcc_value mnc mnc_value mme-group-id-range first id last id
no
Removes the selected PLMN ID to MME group ID range from the MME ID management database.
Usage Guidelines Use this command to create associations between PLMN IDs and a range of MME group IDs.
On the S4-SGSN, use this command to create a custom list of MME group IDs on the S4-SGSN if networks
have been configured with LACs in the 32768-65535 range for UMTS and GPRS coverage. The S4-SGSN
will use this custom list to identify whether the received LAC is a native LAC or a LAC mapped from a
globally unique temporary identifier (i.e., an MME group code part of GUTI). Once the plmn configuration
is completed, operators must associate the configuration with the GPRS and/or SGSN services configured on
the S4-SGSN using the associate network-global-mme-id-mgmt-db command. Refer to the SGSN Service
Configuration Mode and GPRS Service Configuration Mode chapters in the GPRS/UMTS Command Line
Reference for a description of this command.
Examples The following command creates an association between a PLMN ID of 12323 and a set of MME group IDs
with a range of 500 through 575:
plmn mcc 12323 mnc 23 mme-group-id-range first 500 last 575
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Paging Map Configuration
configure > lte-policy > paging-map map_name
Entering the above command sequence results in the following prompt:
[local]host_name(paging-map)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
precedence
Enables the operator to apply a priority for different paging-profiles based on traffic type. When the MME
service is associated with a paging map, the system checks the profile map to determine which paging-profile
to adopt for a given paging trigger.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Paging Map Configuration
configure > lte-policy > paging-map map_name
Entering the above command sequence results in the following prompt:
[local]host_name(paging-map)#
Syntax Description precedence priority traffic-type { cs [ voice | sms | other ] | ps [ arp arp_value | qci qci_value ] | signaling
[ detach | idr | lcs | node-restoration ] } paging-profile paging_profile_name
no precedence priority
no
Remove the paging map from the system.
precedence priority
precedence: For StarOS releases 16.5 and higher, enter an integer from 1 to 19, where 1 is the highest priority
and 19 is the lowest priority. For StarOS releases prior to 16.5, enter an integer from 1 to 7, where 1 is the
highest priority and 7 is the lowest priority. For releases 20.0 onwards enter an integer from 1 to 35, where
1 is the highest priority and 35 is the lowest priority. The numbers of paging-profiles supported are increased
from 8 to 16.
traffic-type { cs [ voice | sms | other ] | ps [ qci qci_value ] | signaling [ detach | idr | lcs | node-restoration
]}
Defines the type of traffic of the incoming call.
• cs (Circuit Switched) - All data and control activities that involve CSFB. Paging requests from the MSC
for mobile-terminated calls alone are treated as CS type. Paging requests for SMS are treated as PS type.
Optionally, define the CS traffic sub-type:
• voice: Paging requests from MSC for mobile terminated voice calls.
• sms: Paging requests from MSC for mobile terminated SMS requests.
• other: MM Information Request messages coming from MSC can also trigger paging if UE is in
IDLE state. These requests are mapped to 'other' sub-traffic type.
If a sub-traffic-type is not configured then paging-profile configured for CS (with no sub traffic-type
qualification) is applied. If no such entry exists, then default heuristics based paging behavior is applied.
• ps (Packet Switched) - All data and control activities that involve packet services. SRVCC is also mapped
to this traffic-type as the voice is carried using PS service.
Optionally, define the QoS Class Identifier (QCI) value for this PS traffic:
qci qci_value
where qci_value is an integer from 1 through 9. (standard values). From 128 up to 255 are non-standard
values.
QCI qualified entries can only be used for paging triggered by Downlink Data Notifications received
on S11. If the incoming DDN contains EPS Bearer ID (EBI) information, the QCI corresponding to that
PDN is used to find the appropriate 'ps qci xx' entry and its configured paging-profile.
If there are multiple EBIs included in the DDN the mapping entry with highest precedence is selected.
If no QCI specific mapping exists, or if the incoming DDN does not have the EBI information then 'PS'
specific mapping (with no qci qualifier) is selected.
Optionally, define the ARP priority based paging for PS traffic type in the paging-map:
arp arp_value
The allowed ARP value " arp_value " is an integer from 1 through 15.
• signaling [ detach | idr | lcs | node-restoration ]: UE level signaling requests. This traffic can be
optionally qualified according to the following sub-traffic types:
detach: Paging requests triggered due to UE getting detached.
idr: Paging triggered in response to an IDR event, such as receiving an IDR Request.
lcs: (Location Services) – Paging requests triggered due to Positioning Requests coming from SMLC
over SLs interface. Mobile Terminated Location Requests arriving on SLg interface can also trigger
paging if UE is in IDLE state, and are included in this sub-traffic type.
node-restoration: Paging requests triggered due to node restoration (for example, due to P-GW Restart
Notification (PRN)). By default, no precedence is assigned to node restoration signaling traffic. The
MME treats node restoration paging with the least priority.
If a sub-traffic-type is not configured then paging-profile configured for signaling (with no sub traffic-type
qualification) shall be applied. If no such entry exists then default-heuristics based paging behavior is
applied.
paging-profile paging_profile_name
The paging-profile to apply for paging UE.
Important If the associated paging-map does not exist or does not have an entry matching the paging-trigger, the
MME performs paging as outlined in 'Default Heuristics Paging Behavior' in the Heuristic and Intelligent
Paging chapter in the MME Administration Guide.
Usage Guidelines Use this command to apply different paging-profiles based on traffic types.
The command defines the order (1 - highest, 35 - lowest) in which the MME checks the entries in this
paging-map. If the paging trigger (like Downlink Data Notification or MSC request) matches the traffic-type
of that entry, then the corresponding paging-profile is used for paging the UE. If the paging trigger does not
match, then the next entry in the precedence order is picked and checked for a match. If no match is found in
the entire paging-map table then default heuristic paging profile is adopted.
If the MME receives another paging trigger (for example from the MSC for CSFB) while paging is already
in progress, the MME checks whether a higher precedence paging profile can be applied. If the new trigger
has a paging-map entry with a higher precedence, the MME restarts the paging process using the paging-profile
associated with the new map entry.
Paging is typically triggered when either the MSC indicates that there is an incoming call to the UE (Call
Service, CS), or when the S-GW sends a Downlink Data Notification (Packet Service, PS) to the MME, or
when there is a bearer/PDN request coming from the P-GW/S-GW.
Refer to the Heuristic and Intelligent Paging chapter in the MME Administration Guide for more information.
Related Commands
Refer to the paging-profile command in the LTE Policy Configuration Commands chapter to create the paging
profiles used in this command.
Examples The following example specifies a special paging-profile for IMS-Voice and a default paging-profile for the
rest of PS paging triggers:
precedence 1 traffic-type ps qci 1 paging-profile profile-voice
precedence 2 traffic-type ps paging-profile profile-default
In the following example, Mobile Terminated voice triggered paging requests will use profile-voice. All other
CS traffic types like MM-InformationRequest and MT-SMS use profile-cs:
precedence 1 traffic-type cs voice paging-profile profile-voice
precedence 2 traffic-type cs paging-profile profile-cs
In the following example, signaling paging requests due to a node restoration (P-GW Restart Notification
(PRN)) will use the prnpaging map, and is assigned a lower precedence of 3:
precedence 3 traffic-type signaling node-restoration paging-profile prn
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Paging Profile Configuration
configure > lte-policy > paging-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(paging-profile)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
paging-stage
Enables the operator to configure different stages of paging in the order of desired execution with parameters
that control the pace, volume, and behavior of a given paging stage.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Paging Profile Configuration
configure > lte-policy > paging-profile profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(paging-profile)#
Syntax Description [ no ] paging-stage level match-criteria { ue-contact-time seconds | all } action { last-n-enb-last-tai
max-n-enb value | all-enb-last-tai | all-enb-all-tai } t3413-timeout seconds max-paging-attempts attempts
no
Remove the paging-stage from the system.
level
Defines different levels of paging-stages, each with a different match-criteria and different action. level must
be a value 1 through 5.
match-criteria
Specifies the criteria for selecting a given paging stage.
• ue-contact-time seconds - Number of seconds elapsed since the MME last heard from UE. This time,
if set, acts as an upper time limit to consider a given paging-stage for paging purposes. seconds must be
a value from 0 through 86400.
• all - No criteria. Operator can use this match-criteria for the final paging stage.
action
Defines how the paging request should be formulated.
• last-n-enb-last-tai max-n-enb value - Send paging request to the last known number of eNodeBs
(configured using max-n-enb value ) and to the last known TAI. value must be a value from 1 through
5.
• all-enb-last-tai - Send paging request to all eNodeBs and to last known TAI.
• all-enb-all-tai - Send paging request to all eNodeBs and to all TAIs.
t3413-timeout seconds
Defines the time-interval in seconds between paging requests. The MME uses this timer for retransmission
of an S1 Paging request to UE for PS paging. timeout must be a value from 0 through 20.
CS triggered S1 Paging requests are transmitted only once by the MME (no retransmission). For a CS paging
to be sent again, another SGs paging request needs to be sent by MSC/VLR towards MME.
max-paging-attempts attempts
Defines the number of paging requests to be sent out during this paging-stage.
attempts must be a value from 0 through 5.
Usage Guidelines Use this command to configure paging procedure stages, which in turn control the pace, volume, and behavior
of paging for each stage.
Related Commands Refer to the paging-map command in the LTE Policy Configuration Commands chapter to assign a priority
for this paging profile based on traffic type.
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Peer Map Configuration
configure > lte-policy > peer-map map_name
Entering the above command sequence results in the following prompt:
[local]host_name(peer-map)#
Important Available commands or keywords/variables vary based on platform type, product version, and installed
license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
precedence
Configures the matching criteria and precedence for mapping an LTE Policy with a peer profile.
Product P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Peer Map Configuration
configure > lte-policy > peer-map map_name
Entering the above command sequence results in the following prompt:
[local]host_name(peer-map)#
Syntax Description precedence priority match-criteria { all peer-profile-name profile_name | peer-ip-address { ip_address |
ip_address/mask } [ serving-plmnid mcc mcc mnc mnc ] peer-profile-name profile_name | serving-plmnid
mcc mcc mnc mnc [ peer-ip-address { ip_address | ip_address/mask } ] peer-profile-name profile_name
}
no precedence priority
no
Removes the selected precedence priority number from the peer map.
priority must be an integer from 1 to 1024.
priority
priority must be an integer from 1 to 1024. Precedence 1 has highest priority.
match-criteria
Defines the criteria to be used to match peer nodes.
all
Specifies that all peer nodes are to be associated with the peer map.
To map a peer to a profile when there is no specific criteria required, use the all keyword.
peer-profile-name profile_name
Sets the peer profile with which the matching criteria is associated.
profile_name must be an existing peer profile expressed as an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to map LTE Policy to a peer profile based on matching criteria and precedence for the
criteria.
A maximum of 1024 precedence entries can be configured.
Examples The following command associates the peer profile named pp5 with peers associated with a serving node
PLMN ID MCC of 111 and an MNC of 222:
precedence 100 match-criteria serving-plmnid mcc 111 mnc 222 peer-profile-name pp5
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
Important Available commands or keywords/variables vary based on platform type, product version, and installed
license(s).
cause-code-group
Creates a new cause code group, or specifies an existing cause code group and enters the Cause Code Group
Configuration Mode.
Product MME
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified cause code group with all related configuration from the LTE Policy.
group_name
Specify a name of a cause-code-group to create, modify, or remove. This name must be an alphanumeric
string of 1 through 16 characters. Each new cause-code-group must have a unique name.
A maximum of 4 cause code groups can be defined across all services (MME+GPRS+SGSN).
protocol
Specifies the protocol for the cause code group being created/accessed. Options include:
• BSSGP for 2G.
• RANAP for 3G
• S1-AP
Usage Guidelines Use this command to create or modify a group of cause codes.
Entering this command results in a prompt, with the protocol ID included, similar to the following:
[context_name]hostname(s1ap-cause-code)#
Depending upon the protocol you have selected, the Cause Code Group configuration commands are defined
in the
• BSSGP Cause Code Configuration Mode Commands chapter of this guide.
• RANAP Cause Code Configuration Mode Commands chapter of this guide.
• S1AP Cause Code Configuration Mode Commands chapter of this guide.
Examples The following command creates an S1-AP cause code group named move-ue-to-idle.
cause-code-group move-ue-to-idle protocol s1ap
congestion-action-profile
Creates an action profile for MME or ePDG or HeNBGW critical, major and minor congestion thresholds.
The profile defines the action to be taken when these thresholds are exceeded.
Important In Release 20.0, HeNBGW is not supported. This command must not be used for HeNBGW in Release
20.0. For more information, contact your Cisco account representative.
Product ePDG
HeNBGW
MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified profile from the system.
profile_name
Specifies the name of the action profile. If the entered name does not refer to an existing profile, a new profile
is created. profile_name is an alphanumeric string of 1 through 64 characters.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to establish MME or ePDG action profiles to be associated with critical, major and minor
congestion thresholds. This command is also used to remove an existing profile.
Note This command is part of a licensed feature and requires a proper license to function: MME Resiliency
Bundle.
For information on setting the action to be taken within this profile, see the Congestion Action Profile
Configuration Mode Commands chapter in this guide, and the Congestion Control chapter in the System
Administration Guide.
Examples The following command creates a major congestion action profile named mme_major_profile and moves to
the Congestion Action Profile Configuration mode:
congestion-action-profile mme_major_profile
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
foreign-plmn-guti-mgmt-db
Creates a new, or enters the existing Foreign PLMN GUTI management database.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Remove the specified management database from the system.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
db-name
Specifies the name of the management database. If the name does not refer to an existing database, a new
database is created.
db_name is an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to create a new, or enter the existing Foreign PLMN GUTI management database. This
management database allows for the optional configuration of foreign PLMNs for which Attach Requests or
TAU Requests containing a GUTI from such a PLMN can either be allowed or immediately rejected.
A maximum of four separate Foreign PLMN GUTI management databases can be configured.
Entering this command results in the following prompt:
[context_name]hostname(foreign-plmn-guti-mgmt-db)#
Global MME ID management database commands are defined in the LTE Foreign PLMN GUTI Management
Examples Database
The Configuration
following command Mode
createsCommands chapter.GUTI management database named fguti-db1.
a Foreign PLMN
henbgw mme-pool
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Creates or configures a specified MME or MME pool to which the HeNB-GW is to communicate and enters
the MME pool configuration mode.
Product HeNB-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes an already configured MME or MME pool from the system.
pool_name
Specifies the name of the MME pool being created or accessed. If the pool name does not refer to an existing
profile, a new pool is created. pool_name is an alphanumeric string of 1 through 63 characters.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to enter the MME Pool Configuration Mode for an existing pool or for a newly defined
MME pool. This command is also used to remove an existing MME pool.
MME Pool Configuration Mode commands are defined in the MME Pool Configuration Mode Commands
chapter.
Examples The following command helps entering the MME Pool Configuration Mode for a new or existing profile
named henb_mme_pool:
mme pool henb_mme_pool
henbgw overload-control
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Product HeNBGW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
default
Sets/Restores the default values assigned to HeNBGW overload control parameters. The default value of load
Reduction indication is 99 percent.
load-reduction-indicator percentage_value
Designates the percentage of HeNBs to relay overload start message if traffic load Reduction indication IE
is not present in the overload start message.
percentage_value is the percentage value, which is an integer between 1 and 99.
ochl-guard-time minutes
Designates the guard timeout value for sending overload stop messages if overload stop message is not received
from MME to all the HeNBs in corresponding MME's overload control HeNBs list.
minutes is the number of minutes, which is an integer between 0 and 2147483647.
Usage Guidelines Use this command to configure HeNBGW overload control parameters.
Examples The following command configures HeNBGW overload control parameter ochl-guard-time as 45 minutes:
henbgw overload-control ochl-guard-time 45
henbgw qci-dscp-mapping-table
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Use this command to configure qci-dscp-mapping-table for HENBGW. The maximum limit for the tables
that can be configured is 32.
Product HeNB-GW
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the qci-dscp-mapping-table for HENBGW.
Important This command on execution will open a new mode HeNBGW QCI DSCP Mapping Table mode.
table_name
It is the qci-dscp-mapping-table for HENBGW, a string of size between 1 and 63.
henbgw s1-reset
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Product HeNBGW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
s1-reset Removes sending s1-reset/partial-reset messages from HeNBGW.
s1-reset
Configures option to enable or disable sending s1-reset/partial-reset messages from HeNBGW.
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Configures HENBGW session recovery. The HENBGW session recovery is valid only when require session
recovery is configured. Base session recovery feature will enable recovery of IP-Sec tunnels when integrated
IP-Sec is used. Enhanced HENBGW session recovery feature will enable recovery of SCTP/UE sessions in
HENBGW. This feature should be enabled if henb(s) have the capability to retain UE S1AP state across SCTP
connection restarts.
Product HeNBGW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the configured HENBGW session recovery.
idle-timeout secs
Configures the idle timeout.
Designates the maximum duration of the session recovered without any activity, in seconds, before system
automatically terminates the session. Zero indicates function is disabled.
secsis the number of seconds (0 is disabled), which is an integer between 0 through 2147483647.
Usage Guidelines Use this command to configure HENBGW session recovery with idle timeout.
Examples The following command configures HeNBGW session recovery with idle timeout 45 seconds:
henbgw session-recovery idle-timeout 45
ho-restrict-list
Creates a handover (HO) restriction list or specifies an existing HO restriction list and enters the Handover
Restriction List Configuration Mode.
Product MME
ePDG
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified restriction list from the system.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
list_name
Specifies the name of the HO restriction list. If the entered list name does not refer to an existing list, a new
list is created. list_name is an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to enter the Handover Restriction List Configuration Mode for an existing list or for a
newly defined list. This command is also used to remove an existing list.
Entering this command results in the following prompt:
[context_name]hostname(ho-restrict-list)#
Handover Restriction List Configuration Mode commands are defined in the LTE Handover Restriction List
Configuration Mode Commands chapter.
Examples The following command enters the Handover Restriction List Configuration Mode for a new or existing list
named ho_restricit_list1:
ho-restrict-list ho_restrict_list1
imei-tac-group
Creates an IMEI-TAC group and provides commands to configure up to 25,000 IMEI-TACs (international
mobile equipment identity - type allocation code (IMEI-TAC) that can be used as the filtering criteria for
Operator Policy selection.
Important This functionality is available on releases 18.6, 19.4, and 20.0 and higher.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified IMEI-TAC group with all related configuration from the LTE Policy.
group_name
Specify a name for the IMEI-TAC group to create, modify, or remove the group. This name must be an
alphanumeric string of 1 through 64 characters. Each new IMEI-TAC group must have a unique name.
A maximum of 50 IMEI-TAC groups can be defined on the MME.
Usage Guidelines Use this command to create, modify, or delete an IMEI-TAC group. Create up to 50 IMEI-TAC groups. Each
group can contain up to 500 unique IMEI-TAC values and/or up to 20 IMEI-TAC ranges, which can be
overlapping.
This command is used as part of the configuration required to enable operator policy selection based on
IMEI-TAC. Including the type allocation code (TAC) in the operator policy selection process supports network
access restrictions being applied to UEs based on the type of wireless device identified by the IMEI-TAC.
For details about this feature and all the other commands required for its configuration, refer to the Operator
Policy Selection Based on IMEI-TAC chapter in the MME Administration Guide.
Examples The following command creates an S1-AP cause code group named move-ue-to-idle.
cause-code-group move-ue-to-idle protocol s1ap
lte-emergency-profile
Creates an LTE emergency profile or specifies an existing emergency profile and enters the LTE Emergency
Profile Configuration Mode.
Product MME
ePDG
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes an LTE emergency profile from the system.
profile_name
Specifies the name of the LTE emergency profile being created or accessed. If the profile name does not refer
to an existing profile, a new profile is created. profile_name is an alphanumeric string of 1 through 64 characters.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to enter the LTE Emergency Profile Configuration Mode for an existing profile or for a
newly defined profile. This command is also used to remove an existing profile.
Important A maximum of four LTE emergency profiles are allowed per system.
LTE Emergency Profile Configuration Mode commands are defined in the LTE Emergency Profile
Configuration Mode Commands chapter.
Examples The following command enters the LTE Emergency Profile Configuration Mode for a new or existing profile
named emergency_profile3:
lte-emergency-profile emergency_profile3
Important In Release 20.0, HeNBGW is not supported. This command must not be used for HeNBGW in Release
20.0. For more information, contact your Cisco account representative.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified management database from the system.
[ -noconfirm ]
Executes the command without any additional prompt and confirmation from the user.
db_name
Specifies the name of the management database. If the name does not refer to an existing database, a new
database is created. db_name is an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to create a new, or enter the existing MME HeNB-GW management database. This command
enables configuration for the MME to distinguish between an HeNB-GW and an eNodeB.
In case of S1-based handovers to Home eNodeBs served by a HeNB-GW, the lookup at the MME for the
target eNodeB based on global eNodeB id cannot be completed, as the MME is aware of only the HeNB-GW.
In those cases, an additional lookup needs to be performed based on the TAI, in order to find the HeNB-GW
serving the Home eNodeB. The S1 Handover request message will then be sent to the HeNB-GW and forwarded
to the correct Home eNodeB in order to prepare the target RAN node for handover.
One HeNB-GW management database is allowed per LTE Policy.
A maximum of 8 HeNB-GWs can be configured within this management database.
Entering this command results in the following prompt:
[context_name]hostname(henbgw-mgmt-db)#
MME HeNB-GW management database commands are defined in the LTE MME HeNB-GW Management
Database Configuration Mode Commands chapter.
Examples The following command enters the existing LTE MME HeNB-GW Managment Database Configuration Mode
(or creates it if it does not already exist) for the database named henbgw_db1:
mme henbgw mgmt-db henbgw_db1
Product MME
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
default
Resets either the paging cache size or the paging cache timer to default settings.
size cache_size
cache_size: Enter an integer from 0 to 10000 to specify the maximum number of Tracking Area Code (TAC)
entries to be cached.
Entering the '0' value disables caching and should be followed by use of the mme paging cache clear command
(in the Exec mode).
Default cache size = 5000 TAC entries per SessMgr.
timeout time
time: Enter an integer from 1 to 1440 to specify the number of minutes that each Tracking Area Code (TAC)
entry remains valid.
A lower cache timeout helps to refresh the cache frequently and enables this functionality to quickly adapt to
changes in the network. We recommend the timeout value be less than the expected eNodeB flap frequency;
for example, if the eNodeBs connected to the MME are expected to disconnect and reconnect every 10 minutes
(due to network issues), then the timeout configuration should be less than 10 minutes.
Default timeout = 5 minutes.
Usage Guidelines Both size and timeout must be configured to enable paging cache optimization. The mme paging cache
command must be entered twice, once for each parameter.
Examples Set the paging cache timeout to match the eNodeB flap frequency of 10 minutes:
mme paging cache timeout 10
Set the paging cache size to disable caching:
mme paging cache size 0
end
mme paging cache clear { all | instance sessmgr_instance }
network-global-mme-id-mgmt-db
Creates a new, or enters the existing MME group ID management database.
Product MME
ePDG
SAEGW
S-GW
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Remove the global MME ID management database from the system.
Usage Guidelines Use this command to create a new, or enter the existing MME group ID management database. The MME
group ID management database is used to create associations between PLMN IDs and ranges of MME group
IDs.
On the S4-SGSN, this command enables operators to create a MME group ID management database that can
be associated with an SGSN service.
Important Only one MME group ID management database can be created per system.
Global MME ID management database commands are defined in the LTE Network Global MME ID
Management Database Configuration Mode Commands chapter.
paging-map
Creates a paging map or specifies an existing paging map and enters the Paging Map Configuration Mode.
Product MME
ePDG
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Remove the paging map from the system.
paging_map_name
Specifies the name of the paging map being created or accessed. If the map name does not refer to an existing
map, a new map is created. paging_map_name must be an alphanumeric string of 1 through 64 characters.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
Usage Guidelines Enter the Paging Map Configuration Mode for an existing or newly defined map. This command is also used
to remove an existing map.
Entering this command results in the following prompt:
hostname(paging-map)#
Paging Map Configuration Mode commands are defined in the LTE Paging Map Configuration Mode
Commands chapter.
Refer to the Heuristic and Intelligent Paging chapter in the MME Administration Guide for more information
about Paging Maps.
Examples The following command enters the existing Paging Map Configuration Mode (or creates it if it does not already
exist) for the map named map1:
paging-map map1
paging-profile
Creates a paging profile or specifies an existing paging profile and enters the Paging Profile Configuration
Mode.
Product MME
ePDG
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Remove the paging map from the system.
paging_profile_name
Specifies the name of the paging profile being created or accessed. If the profile name does not refer to an
existing profile, a new profile is created. paging_profile_name must be an alphanumeric string of 1 through
64 characters.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
Usage Guidelines Enter the Paging Profile Configuration Mode for an existing or newly defined paging profile. This command
is also used to remove an existing profile.
Entering this command results in the following prompt:
hostname(paging-profile)#
Paging Profile Configuration Mode commands are defined in the LTE Paging Profile Configuration Mode
Commands chapter.
Refer to the Heuristic and Intelligent Paging chapter in the MME Administration Guide for more information
about Paging Profiles.
Examples The following command enters the existing Paging Profile Configuration Mode (or creates it if it does not
already exist) for the profile named profile1:
paging-profile profile1
peer-map
Creates a peer map and enters the LTE Peer Map Configuration mode.
Product P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified peer map from the LTE policy.
map_name
Specify a name of a peer map to create, modify, or remove. This name must be an alphanumeric string of 1
through 64 characters. Each new peer map must have a unique name.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
LTE Peer Map Configuration Mode commands are defined in the LTE Peer Map Configuration Mode
Commands chapter of this guide.
Examples The following command creates a peer map named map5 and enters the LTE Peer Map Configuration mode:
peer-map map5
sgsn-mme
This command is used to enable or disable subscriber data optimization in a SGSN-MME combo node.
Product SGSN
MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Disables the configured optimization in a SGSN-MME combo node.
subscriber-data-optimization
Enables subscriber data optimization in a SGSN-MME combo node.
Usage Guidelines This command is used to configure data optimization in a SGSN-MME combo node. When this command is
configured in a co-located SGSN and MME node, lower memory or CPU utilization and reduced signaling
towards other nodes in network is achieved. This feature is supported by both the S4-SGSN and the Gn-SGSN.
For the feature to apply to a Gn-SGSN, the Gn-SGSN must be configured to connect to an HSS. The is a
licensed Cisco feature. A separate feature license is required. Contact your Cisco account representative for
detailed information on specific licensing requirements. For information on installing and verifying licenses,
refer to the Managing License Keys section of the Software Management Operations chapter in the System
Administration Guide
Examples The following command is used to enable subscriber data optimization in a SGSN-MME combo node.
sgsn-mme subscriber-data-optimization
subscriber-map
Creates a subscriber map or specifies an existing subscriber map and enters the Subscriber Map Configuration
Mode.
Product MME
ePDG
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified subscriber map from the system.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
map_name
Specifies the name of the subscriber map. If the map name does not refer to an existing map, a new map is
created. map_name must be an alphanumeric string of 1 through 64 characters.
Usage Guidelines Enter the Subscriber Map Configuration Mode for an existing or newly defined map. This command is also
used to remove an existing map.
Entering this command results in the following prompt:
hostname(subscriber-map)#
Subscriber Map Configuration Mode commands are defined in the LTE Subscriber Map Configuration Mode
Commands chapter.
Examples The following command enters the existing Subscriber Map Configuration Mode (or creates it if it does not
already exist) for the map named map1:
subscriber-map map1
tai-list-db
Important In Release 20.0, HeNBGW is not supported. This command must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Creates or configures a list of Tracking Area Information (TAI). This list is called TAI List Database.
Product HeNB-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified TAI list database from the system.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
db_name
Specifies the name of the TAI list database. If the name does not refer to an existing database, a new database
is created. db_name is an alphanumeric string of 1 through 64 characters.
Usage Guidelines Enters the TAI List Database Configuration Mode for an existing or newly defined database. This command
is also used to remove an existing database.
Important Maximum of eight TAI List Database instances can be created in a system.
TAI List Database Configuration Mode commands are defined in the TAI List Database Configuration Mode
Commands chapter.
Examples The following command enters the existing TAI List Database configuration mode (or creates it if it does not
already exist) for the database named tai_db1:
tai-list-db tai_db1
tai-mgmt-db
Creates a Tracking Area Identifier (TAI) Management Database or specifies an existing database and enters
the TAI Management Database Configuration mode. On an S4-SGSN, this command is used as part of
configuring S-GWs and their associated RAIs to bypass DNS resolution of RAI FQDN for obtaining the
S-GW address.
Product MME
ePDG
SAEGW
S-GW
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration
configure > lte-policy
Entering the above command sequence results in the following prompt:
[local]host_name(lte-policy)#
no
Removes the specified management database from the system.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
db_name
Specifies the name of the management database. If the name does not refer to an existing database, a new
database is created. db_name is an alphanumeric string of 1 through 64 characters.
Usage Guidelines Enter the TAI Management Database Configuration Mode for an existing or newly defined database. This
command is also used to remove an existing database.
On the S4-SGSN, once you have created a new, or accessed an existing, TAI Management database, a TAI
Management Object must be created or accessed and the S-GWs and their associated RAIs configured using
the rai and sgw-address commands. Refer to the LTE TAI Object Configuration Mode chapter for details on
these two commands.
Entering this command results in the following prompt:
[context_name]hostname(tai-mgmt-db)#
TAI Management Database Configuration Mode commands are defined in the TAI Management Database
Configuration Mode Commands chapter.
Examples The following command enters the existing TAI Management Database configuration mode (or creates it if
it does not already exist) for the database named tai_db1:
tai-mgmt-db tai_db1
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Subscriber Map Configuration
configure > lte-policy > subscriber-map subscriber_map_name
Entering the above command sequence results in the following prompt:
[local]host_name(subscriber-map)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
precedence
Sets the order of precedence, the matching criteria and the association to an operator policy for subscribers
meeting the match criteria.
Product MME
SAEGW
SaMOG
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE Subscriber Map Configuration
configure > lte-policy > subscriber-map subscriber_map_name
Entering the above command sequence results in the following prompt:
[local]host_name(subscriber-map)#
no
Removes the selected precedence number definition from the subscriber map. number must be an integer
value from 1 to 1024.
number
Specifies the order of precedence for the subscriber map. 1 (the lowest number) takes the highest precedence.
number must be an integer value from 1 to 1024.
match-criteria
Specifies that the keyword following this keyword is the criteria to be used to match a UE.
all
Specifies that all UEs are to be associated with the operator policy.
So with release 19.4 the behavior of the syntax has modified slightly so that if imei-tac-group is the selected
matching criteria, then the command syntax would be similar to:
precedence precedence_value match-criteria imei-tac group group_name [ imsi mcc mcc mnc mnc |
msin { first start_msin_value last end_msin_value } ] [ service-plmnid plmn_id] operator-policy
policy_name
imsi mcc mcc_num mnc mnc_num [ msin first start_range last end_range | service-plmnid id ]
Specifies that UEs with criteria matching the International Mobile Subscriber Identifier (IMSI) information
(MCC and MNC) are to be associated with a specified operator policy.
mcc mcc_num: Specifies the mobile country code (MCC) portion of the IMSI identifier as an integer value
between 100 and 999.
mnc mnc_num: Specifies the mobile network code (MNC) portion of the IMSI identifier as a 2- or 3-digit
integer value between 00 and 999.
msin first start_range last end_range: Optionally specifies a range of Mobile Subscriber Identification
Numbers that further narrows the match criteria for the IMSI match configuration. start_range and end_range
must each be an integer value of 10 digits.
service-plmnid plmn_id: Optionally specifies a local service PLMN ID number used further narrow the
IMSI-based operator policy selection. plmn_id must be an integer value of five digits minimum and six digits
maximum (the combination of the MCC and MNC).
service-plmnidplmn_id
Specifies a local service PLMN ID number used for PLMN ID-based operator policy selection. id must be
an integer value of five digits minimum and six digits maximum (the combination of the MCC and MNC).
ssid id
Specifies a local SSID used for SSID-based operator policy selection to support the SaMOG Local Break Out
(LBO) feature. The operator policy and associated call control profile are selected based on the SSID received
in the Called-Station-Id attribute in Access-Request.
id must be an alphanumeric string of 1 through 32 characters.
The SaMOG LBO feature is license dependant. Contact your Cisco account representative for more information.
operator-policy-name policy_name
Sets the operator policy with which the matching criteria is associated. policy_name must be an existing
operator policy expressed as an alphanumeric string of 1 through 64 characters. Operator policies are configured
in the Operator Policy Configuration Mode. For more information about operator policies, refer to the Operator
Policy Configuration Mode Commands chapter.
Usage Guidelines The operator policy provides mechanisms to fine tune the behavior of subsets of subscribers above and beyond
the behaviors described in the user profile. It is also used to control the behavior or visiting subscribers in
roaming scenarios, enforce roaming agreements, and provide a measure of local protection against foreign
subscribers.
Examples The following command associates the operator policy named op_pol1 with UEs associated with an IMSI
MCC of 111, an MNC of 222, and a service PLMN ID of 123456:
precedence 100 match-criteria imsi mcc 111 mnc 222 service-plmnid 123456 operator-policy-name
op_pol1
The following command gives the match-criteria configuration a precedence of 155 and instructs the MME
to select and apply the nokia1 operator policy for UEs with IMEI-TAC that matches one of the following
IMEI-TAC: 35850000 or 01124500
precedence 155 match-criteria imei-tac value 35850000 01124500 operator-policy-name nokia1
The following command identifies this as having the highest precedence and sets the matching criteria for the
operator policy selection to based on IMEI-TAC + MCC-MNC of UE + Serving PLMNID:
precedence 1 match-criteria imei-tac-group myGroup imsi mcc 123 mnc 234 service-plmnid 56789
operator-policy-name BESTpol
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
configure > lte-policy > tai-mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-db)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
network-name
This command configures the long (full) and short network name used in the Long and Short network name
IE in the EMM Information message that is sent to the UE from the MME.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
configure > lte-policy > tai-mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-db)#
no
Removes a configured network-name.
full full_name
Defines the full (long) network name. full_name is an alphanumeric string of 1 through 251 characters.
short short_name
Defines the short network name. short_name is an alphanumeric string of 1 through 251 characters.
Usage Guidelines Use this command to configure the full and short network name on the tai-db object.
This configuration affects the actions of the MME when the MME is configured to set the UE time. If this
configuration exists, and there is no interaction with an MSC that sends a EMM INFORMATION message,
the MME uses the above configuration while sending an EMM INFORMATION message.
There is no default for this configuration. If this configuration does not exist at the tai-db level and
tai-mgmt-object level, no network name will be signaled in a EMM Information message.
tai-custom-list
Creates a new custom TAI list, and/or enters the tai-cstm-list configuration mode.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
configure > lte-policy > tai-mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-db)#
tac value
Specifies the Tracking Area Code portion of the TAI as an integer from 1 through 65535.
A maximum of 1000 Custom TAI Lists can be configured per TAI Management Database.
Usage Guidelines Use this command to enter the Custom TAI List Configuration Mode for an existing object or for a newly
defined object.
Prior to 17.0, the MME could have a tracking area in only one tracking area list (TAI List). Consequently,
the tracking area list assigned to subscribers attaching from different TAIs will be same, even if the adjacency
of these tracking areas is not same. This resulted in the MME getting TAUs even as subscribers moved to the
adjacent area.
With this functionality, you can configure adjacency lists as TAI Lists, thus reducing the Tracking Area
Updates (TAU) received by MME. This feature enables the MME to send configured customized TAI List
in ATTACH_ACCEPT/TAU_ACCEPT when a request is received from the custom or border TAIs.
Entering this command results in the following prompt:
[local]hostname{tai-cstm-list}#
Custom TAI List Configuration Mode commands are defined in the LTE Custom TAI List Configuration Mode
Commands chapter.
Examples The following command creates a Custom TAI List for TAC 2325 and enters the Custom TAI List Configuration
Mode:
custom-tai-list tac 2325
tai-mgmt-obj
Creates new, or removes/enters existing, LTE Tracking Area Identifier (TAI) object configurations. On the
S4-SGSN, this command is required as part of configuring S-GWs and their associated RAIs to bypass DNS
resolution of RAI FQDN for obtaining the S-GW address.
Product MME
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
configure > lte-policy > tai-mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-db)#
no
Removes a configured TAI management object from the TAI management database.
-noconfirm
Executes the command without any additional prompt and confirmation from the user.
object_name
Specifies the name of the TAI management object and enters the LTE TAI Management Object Configuration
Mode as an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to enter the LTE TAI Management Object Configuration Mode for an existing object or
for a newly defined object. This command is also used to remove an existing object.
On the S4-SGSN, after creating the TAI Management Object and entering TAI Management Object
Configuration Mode, the rai and sgw-address commands are used to complete the S-GW for RAI configuration.
Refer to the LTE TAI Management Object Configuration mode chapter for details on these commands.
Entering this command results in the following prompt:
[context_name]hostname(tai-mgmt-obj)#
LTE TAI Management Object Configuration Mode commands are defined in the LTE TAI Management Object
Configuration Mode Commands chapter.
Examples The following command creates a TAI management object called tai-obj3 and enters the LTE TAI Management
Object Configuration Mode:
tai-mgmt-obj tai-obj3
timezone
Configures the timezone to be used for the UE timezone in S11 and Non Access Stratum (NAS) messages.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
configure > lte-policy > tai-mgmt-db db_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-db)#
+|-
Specifies the offset direction from the Coordinated Universal Time (UTC).
hours value
Specifies the offset from UTC in hours. value must be an integer from 0 through 14.
minutes { 0 | 15 | 30 | 45 }
Optionally specifies the offset minutes added to the hours value.
daylight-savings-time-increment { 0 | 1 | 2 }
Specifies the number of hours the timezone should be offset due to daylight savings time. This allows the
MME to serve areas that have daylight savings time different than that of the MME. This keyword is available
in release 14.0 and higher.
If the TAI management database/object is configured for daylight savings using this keyword, the daylight
savings time adjustment is applied in these messages only if the system time is within a daylight savings
period.
Usage Guidelines Use this command to configure the timezone to be used for the UE timezone in S11 and NAS messages.
Important Time zone configurations at the TAI Management Object level take precedence over time zone
configurations at the TAI Management Database level. If neither is configured, the system defaults to the
MME (system) time zone.
Examples The following command sets the timezone mapping for this management database to plus-3 hours and 15
minutes from UTC:
timezone + hours 3 minutes 15
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
emergency-services-not-supported
This command disables emergency services at a TAI object management level per TAC basis.
Product MME
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
For the emergency-services-not-supported command, the no command prefix enables emergency services
at TAI management object level.
Usage Guidelines In a shared RAN network, there are several TACs contolled by several operators connected to the same MME,
and disabling emergency services for a particular TAC becomes a challenge. To over come this, MME provides
the emergency-services-not-supported command, which disables emergency services at a TAI management
object level per TAC basis.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ims-voice-over-ps
Configures support for IMS Voice over Packet-Switched information element for this TAI List.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
Disables support for IMS Voice over PS for this TAI List.
Usage Guidelines Use this command to enable support for IMS Voice over PS for this TAI List.
When enabled, the IMS Voice Over PS Indicator is sent in the S6a Insert-Subscriber-Data Answer message.
This indicates whether the TAI supports 'IMS Voice over PS session'.
If IMS Voice over PS support is configured globally within the Call Control Profile (using the
network-feature-support-ie ims-voice-over-ps command) as well as on a per TAI basis, the global
configuration (from call control profile) is honored.
Examples The following command enables the MME to send the IMS Voice over PS indicator in the S6a
Insert-Subscriber-Data Answer message for this TAI list.
ims-voice-over-ps
lai
Configures a Location Area Identifier (LAI) for this TAI management object.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
Removes a configured LAI from the TAI management object.
mcc number
Specifies the mobile country code (MCC) portion of a PLMN identifier as an integer from 100 through 999.
mnc number
Specifies the mobile network code (MNC) portion of a PLMN identifier as a 2- or 3-digit integer from 00
through 999.
lac area_code
Specifies the Location Area Code portion of the TAI as an integer from 1 through 65535.
Usage Guidelines Use this command to configure an LAI for this management object.
Examples The following command adds an LAI to this management object with an MCC of 111, an MNC of 22, and a
LAC of 101:
lai mcc 122 mnc 22 lac 101
network-name
This command configures the long (full) and short network name used in the Long and Short network name
IE in the EMM Information message that is sent to the UE from the MME.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
Removes a configured network-name.
full full_name
Defines the full (long) network name. full_name is an alphanumeric string of 1 through 251 characters.
short short_name
Defines the short network name. short_name is an alphanumeric string of 1 through 251 characters.
Usage Guidelines Use this command to configure the full and short network name on the tai-mgmt-obj.
This configuration affects the actions of the MME when the MME is configured to set the UE time. If this
configuration exists, and there is no interaction with an MSC that sends an EMM INFORMATION message,
the MME uses the above configuration while sending a EMM INFORMATION message.
There is no default for this configuration. If this configuration does not exist at the tai-db level and
tai-mgmt-object level, no network name will be signaled in a EMM Information message.
rai
Configures a Routing Area Identifier (RAI) for an associated S-GW for this TAI management object.
Product SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
Syntax Description rai mcc number mnc number lac area_code rac rac_value
no rai
no
Removes a configured LAI from the TAI management object.
mcc number
Specifies the mobile country code (MCC) portion of a PLMN identifier as an integer from 100 through 999.
mnc number
Specifies the mobile network code (MNC) portion of a PLMN identifier as a 2- or 3-digit integer from 00
through 999.
lac area_code
Specifies the Location Area Code portion of the TAI as an integer from 1 through 65535.
rac
Specifies the Routing Area Code portion of the TAI as an integer from 1 to 255.
Usage Guidelines On the S4-SGSN, use this command as part of the configuration of the selection of an SGW for RAI on the
S4-SGSN for operators wishing to bypass the DNS resolution of RAI FQDN for obtaining the SGW address.
Once the RAI is configured, the SGW address that serves this RAI must be configured with the sgw-address
command. For details on this command, refer to the description of sgw-address in this chapter.
Examples The following command adds a RAI to this management object with an MCC of 111, an MNC of 22, and a
LAC of 110:
rai mcc 111 mnc 22 lac 101 rac 110
sgw-address
Configures an IP address for a Serving Gateway (S-GW), a supported S5/S8 protocol type, and selection
weight used in a pool for S-GW selection. On an S4-SGSN, this command is used as part of the configuration
to bypass DNS resolution of RAI FQDN for an S-GW.
Product MME
SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
Syntax Description sgw-address ipv4_or_ipv6_address s5-s8-protocol { both | gtp | pmip } weight number
no sgw-address ipv4_or_ipv6_address s5-s8-protocol { both | gtp | pmip }
ipv4_or_ipv6_address
Specifies the IP address of the S-GW in the selection pool in IPv4 dotted-decimal or IPv6
colon-separated-hexadecimal notation.
Up to 32 S-GW addresses can be configured per TAI management object.
weight number
Specifies the priority or weight of the S-GW address used during weighted round-robin selection within this
TAI management object. number must be an integer from 1 through 100.
Usage Guidelines Use this command to configure a pool of S-GW addresses used for S-GW selection.
On the S4-SGSN, use this command to complete the configuration of bypassing DNS resolution of RAI FQDN
to obtain the S-GW address. This command is not valid on the S4-SGSN until the following commands have
been executed:
• tai-mgmt-db in LTE Policy Configuration Mode
• tai-mgmt-obj in LTE Management Database Configuration Mode
• rai in LTE TAI Management Object Configuration Mode
Examples The following command configures an S-GW with an IPv4 address of 10.2.3.4, a supported S5/S8 protocol
type of GTP, and a selection weight of 3:
sgw-address 10.2.3.4 s5-s8-protocol gtp weight 3
sgw-address-resolution-mode
This command specifies the address resolution mode of the SGW address(s) configured in this object. This
command is applicable only for S4-SGSN.
Product SGSN
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
default
Resets the configuration to the default value, that is fallback-for-dns.
fallback-for-dns
Instructs the system to try DNS resolution. If the DNS query fails, the SGSN will use locally configured
addresses. The S4-SGSN will use locally configured SGW address on DNS failure
Default: enabled
local
Instructs the system to only use locally configured S-GW addresses and not to use DNS query.
Default: disabled
Usage Guidelines Use this command to specify the DNS query or local address resolution for this LTE TAI Management Object.
The addresses will be valid only for lac and rac defined under tai-mgmt-object.
Examples The following command sets the address resolution mode to use local addresses only if the DNS query fails:
sgw-address-resolution-mode fallback-for-dns
tai
Configures a Tracking Area Identifier (TAI) for this TAI management object.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
Removes a configured TAI from the TAI management object.
mcc number
Specifies the mobile country code (MCC) portion of a PLMN identifier as an integer from 100 through 999.
mnc number
Specifies the mobile network code (MNC) portion of a PLMN identifier as a 2- or 3-digit integer from 00
through 999.
tac value +
Specifies the Tracking Area Code portion of the TAI as an integer from 1 through 65535. Up to 16 TAC
values can be entered on a single line.
Usage Guidelines Use this command to configure one or more TAIs for this management object. Up to 16 TAIs can be configured
per management object.
Examples The following command adds a TAI to this management object with an MCC of 111, an MNC of 22, and a
TAC value of 1001:
tai mcc 122 mnc 22 tac 1001
timezone
Configures the timezone to be used for the UE timezone in S11 and Non-Access Stratum (NAS) messages.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no timezone
Removes the timezone configuration from the management object.
+|-
Specifies the offset direction from Coordinated Universal Time (UTC).
hours value
Specifies the offset from UTC in hours as an integer from 0 through 14.
minutes { 0 | 15 | 30 | 45 }
Optionally specifies the offset minutes added to the hours value.
daylight-savings-time-increment { 0 | 1 | 2 }
Specifies the number of hours the timezone should be offset due to daylight savings time. This allows the
MME to serve areas that have daylight savings time different than that of the MME. This keyword is available
in release 14.0 and higher.
If the TAI management database/object is configured for daylight savings using this keyword, the daylight
savings time adjustment is applied in these messages only if the system time is within a daylight savings
period.
Usage Guidelines Use this command to configure the timezone to be used for the UE timezone in S11 and NAS messages.
Important Time zone configurations at the TAI Management Object level take precedence over time zone
configurations at the TAI Management Database level. If neither is configured, the system defaults to the
MME (system) time zone.
Examples The following command sets the timezone mapping for this management object to plus-3 hours and 15 minutes
from UTC:
timezone + hours 3 minutes 15
zone-code
Configures a zone code for the management object.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > LTE Policy Configuration > LTE TAI Management Database Configuration
> LTE TAI Management Object Configuration
configure > lte-policy > tai-mgmt-db db_name > tai-mgmt-obj obj_name
Entering the above command sequence results in the following prompt:
[local]host_name(tai-mgmt-obj)#
no
Removes the configured zone-code from the management object.
zc_id
Identifies the zone code configuration instance as an integer from 1 through 65535.
Usage Guidelines
Important While there is no limit to the number of zone codes that can be created, only 10 LACs per zone code can
be defined.
Use this command to define zone code restrictions. Regional subscription data at the home subscriber service
(HSS) is used to determine the regional subscription area in which the subscriber is allowed to roam. The
regional subscription data consists of a list of zone codes which are comprised of one or more location areas
(identified by a LAC) into which the subscriber is allowed to roam. Regional subscription data, if present in
the Insert-Subscriber-Data-Request (IDR) and the Update-Location-Answer (ULA) from the HSS, defines
the subscriber's subscription area for the addressed MME. It contains the complete list (up to 10 zone codes)
that apply to a subscriber in the currently visited PLMN. During the Location Update procedure, the zone
code list is received in the ULA from the HSS. The zone code list is validated against the configured values
in this command. If matched, the Location Update procedure is allowed to proceed. If not matched, the response
is that the Network Node Area is restricted and the Location Update procedure fails.
Examples The following command sets the zone code for this management object to 1:
zone-code 1
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
bind
Binds the service to a logical IP interface serving as the S2a (HSGW, SaMOG) or S5/S8 (S-GW) interface
and specifies the maximum number of subscribers that can access this service over the configured interface.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
no
Removes the interface binding from this service.
address ipv6_address
Specifies the IPv6 address of the interface configured as the S5/S8 interface.
ipv6_address is specified in IPv6 colon-separated-hexadecimal notation.
ipv4-address ipv4_address
Specifies the IPv4 address of the interface configured as the S2a or S5/S8 interface.
Important The SaMOG PMIPv6-based S2a interface currently supports IPv4 bind address only.
max-subscribers num
Default: 1500000
Specifies the maximum number of subscribers that can access this service on this interface.
num must be an integer from 0 through 3000000.
Important The maximum number of subscribers supported depends on the installed license key and the number of
active packet processing cards in the system. A fully loaded system can support 3,000,000 total subscribers.
Refer to the license key command and the Usage section (below) for additional information.
Usage Guidelines Associate the MAG service to a specific logical IP address. The logical IP address or interface takes on the
characteristics of an S2a or S5/S8 interface that provides the session connectivity to/from a P-GW. Only one
interface can be bound to a service. The interface should be configured prior to issuing this command.
This command also sets a limit as to the number of simultaneous subscribers sessions that can be facilitated
by the service/interface at any given time.
When configuring the max-subscribers option, be sure to consider the following:
• The total number of S2a or S5/S8 interfaces you will configure
• The total number of subscriber sessions that all of the configured interfaces may handle during peak
busy hours
• An average bandwidth per session multiplied by the total number of sessions
• The type of physical port to which these interfaces will be bound
Taking these factors into account and distributing your subscriber session across all available interfaces will
allow you to configure your interfaces to optimally handle sessions without degraded performance.
Examples The following command would bind the logical IP interface with the address of
4551:0db8:85a3:08d3:3319:8a2e:0370:1344 to the MAG service and specifies that a maximum of 300,000
simultaneous subscriber sessions can be facilitated by the interface/service at any given time:
bind address 4551:0db8:85a3:08d3:3319:8a2e:0370:1344 max-subscribers 300000
encapsulation
Configures data encapsulation type to be used for specific MAG service.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Resets the encapsulation type to be used by this service to the default option of GRE.
gre
gre: Specifies that GRE encapsulation is to be used for PMIPv6 tunnel data between the MAG and the Local
Mobility Anchor (LMA). This is the default for this command.
Important The SaMOG PMIPv6-based S2a interface currently supports GRE encapsulation only.
ipip
ipip: Specifies that IP-in-IP encapsulation is to be used for PMIPv6 tunnel data between the MAG and the
LMA.
Usage Guidelines Use this command to select the encapsulation type to be used for PMIPv6 tunnel data between the MAG and
the LMA.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
heartbeat
Configures the PMIPv6 heartbeat message interval, retransmission timeout, and max retransmission for the
MAG Service.
Product HSGW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
Syntax Description heartbeat { interval seconds | retransmission { max number | timeout seconds } }
default heartbeat { interval | retransmission { max | timeout } }
no heartbeat
no
Disables the PMIPv6 heartbeat functionality. The HSGW starts sending heartbeat request to peers when the
heartbeat interval is configured.
default
Resets the specified parameter to the system default value.
interval seconds
The interval in seconds at which heartbeat messages are sent.
seconds is an integer from 30 through 3600.
Default: 60
Usage Guidelines Proxy Mobile IPv6 (PMIPv6) is a network-based mobility management protocol to provide mobility without
requiring the participation of the mobile node in any PMIPv6 mobility related signaling. The Mobile Access
Gateway (MAG) service sets up tunnels dynamically to manage mobility for a mobile node.
This command provides configuration of heartbeat messages between the MAG and LMA services to know
the reachability of the peers, to detect failures, quickly inform peers in the event of a recovery from node
failures, and allow a peer to take appropriate action.
Examples The following command enables PMIPv6 heartbeat messaging to known MAG service peers and sets the
heartbeat interval to 160 seconds.
heartbeat interval 160
information-element-set
Identifies the information element set of mobility options to be used in Proxy Binding Update (PBU) messages
sent by the MAG to the LMA.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Resets the command to the default value of "standard".
custom2 [ disable-ie hardware-ie ]: Specifies that a custom set of mobility options will be used in proxy
binding update messages that are sent in Vendor Specific Mobility Options. When enabled, BSID will be sent
in S2a PMIPv6 message.
If the disable-ie hardware-ie keyword is included with this command, then the information elements in PBU
and the hardware ID in PBU are disabled. When information element custom2 is enabled, all the supported
mobility options are enabled including hardware ID for a specific customer. The disable-iekeyword only
disables the hardware ID mobility option in PBU. The hardware-id mobility option format is supported like
custom1.
custom3: Specifies that a custom set of mobility options will be used in proxy binding update messages that
are sent in Vendor Specific Mobility Options.
standard: Specifies that a standard set of mobility options are to be used in proxy binding update messages.
3GPP specification 29.275 defines these as Protocol Configuration Options.
Important The information element set of mobility options for SaMOG PMIPv6-based S2a interface must be set to
standard. Any other configuration may result in a call setup failure.
Usage Guidelines Use this command to identify the type of information element set of mobility options to be used in PBU
messages sent form the MAG to the LMA. The mobility options can be either standards-based (3GPP 29.275)
or custom (vendor-specific as defined by 3GPP 29.275).
Examples The following command identifies the information element set of mobility options to use in PBU messages
as custom:
information-element-set custom1
max-retransmissions
Configures maximum number of retransmissions of Proxy MIP control messages to the Local Mobility Anchor
(LMA).
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Rests the maximum number of allowed retransmissions to the default value of 5.
num
Default: 5
Specifies the maximum number of times the MAG service will attempt to communicate with the LMA before
it marks it as unreachable.
count must be an integer from 0 through 4294967295.
Usage Guidelines Use this command to limit the number of retransmissions to LMA before marking it as unreachable. If the
value configured is reached, the call is dropped.
Examples The following command configures the maximum number of retransmissions for the MAG service to 3:
max-retransmissions 3
mobility-header-checksum
Switches between RFC3775 and RFC6275 for the "next header" value in the mobility header.
Product HSGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Sets the command to the default value of rfc3775.
rfc3775
Configures the "next header" value to 2, as defined in RFC3775.
rfc6275
Configures the "next header" value to 135, as defined in RFC6275.
Usage Guidelines Use this command to switch between RFC3775 and RFC6275 for the "next header" value in the mobility
header. This value is used for appending and calculating the checksum for outbound mobility messages from
MAG to LMA. For inbound messages from LMA to MAG, either of the two values are acceptable for verifying
the checksum.
Examples The following command configures the "next header" value to 2, as defined in RFC3775:
mobility-option-type-value rfc3775
mobility-option-type-value
Changes the mobility option type value used in mobility messages.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Sets the command to the default value of custom1.
custom1
(Default) Non-standard type values used before they were defined by IANA.
standard
Standard type values as defined by IANA. In addition, standard option uses type values defined in RFC 5844
for home address (HoA) options for the PMIPv6 PBU/PBA/revocation message.
Usage Guidelines Use this command to change the mobility option type value used in mobility messages.
Examples The following command changes the mobility option type value to standard:
mobility-option-type-value standard
policy
Configures policies applied to MAG service.
Product HSGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
Syntax Description policy include-bsid-binding-update { all | none { dereg | handoff | init | renew } }
[ default ] policy include-bsid-binding-update
default
Restores the command to the default values of init and handoff.
Examples The following command configures the MAG Service to include BSID in the PBU during initalization,
'registration renewal', and deregistration.
policy include-bsid-binding-update init renew dereg
reg-lifetime
Configures the Mobile IPv6 session registration lifetime for this service.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Resets the command value to the default setting of 600.
seconds
Default: 600
Sets the time value (in seconds) for session lifetimes for this service.
seconds must be an integer from1 through 262140.
Usage Guidelines Use this command to limit PMIPv6 lifetime on this service. If the Proxy Binding Acknowledge (PBA) from
the LMA contains a lifetime shorter or longer than what is specified, it is used instead.
Examples The following command sets the registration lifetime for Mobile IPv6 sessions using this service to 1200
seconds (20 minutes):
reg-lifetime 1200
renew-percent-time
Configures percentage of lifetime at which a registration renewal is sent to the Local Mobility Anchor (LMA).
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Resets the command to the default value of 75.
percent
Default: 75
Specifies the time percentage when the registration renewal is sent to the LMA. percent is a percentage value
of the registration lifetime expressed as an integer from 1 through 100.
Usage Guidelines Use this command to specify when a registration renewal is sent to the LMA for subscribers using this service.
If the registration lifetime is 600 seconds (10 minutes) and this command is set to 75 (percent), then the
registration renewal message is sent after 450 seconds of the registration lifetime has expired.
Examples The following command sets the registration renewal time for subscribers using this service to 90 percent of
the registration lifetime:
renew-percent-time 90
retransmission-policy
Configures the retransmission policy for Proxy MIP control message retransmissions.
Product HSGW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Returns the command to its default setting of exponential-backoff.
{ exponential-backoff | normal }
Sets the retransmission timeout behavior for this service.
exponential-backoff: Specifies that the Proxy Binding Update (PBU) retransmission uses an exponential
backoff to increase the retransmission timeout for each retry.
normal: Specifies that the PBU retransmission uses the configured retransmission timeout value for all PBU
retransmission retries.
Usage Guidelines Use this command to specify the retransmission policy for PMIP control messages.
Examples The following command sets the retransmission timeout policy for PMIP control packets to "normal":
retransmission-policy normal
retransmission-timeout
Configures the maximum allowable time for the MAG service to wait for a response from the LMA before
it attempts to communicate with the LMA again (if the system is configured to retry the LMA), or marks the
LMA as unreachable.
Product HSGW
S-GW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Resets the timeout setting to the default value of 3,000 milliseconds.
no
Deletes a previously configured timeout value.
time
Default: 3000
Specifies the maximum allowable time (in milliseconds) for the MAG service to wait for a response from the
LMA before it: (a) attempts to communicate with the LMA again (if the system is configured to retry the
LMA) or (b) marks the LMA as unreachable.
time must be an integer from 100 through 100000.
Usage Guidelines Use the retransmission timeout command in conjunction with the max-retransmissions command in order
to configure the MAG services behavior when it does not receive a response from a particular LMA.
Examples The following command configures a retransmission timeout value of 5000 milliseconds:
retransmission-timeout 5000
signalling-packets
Enables the QoS Differentiated Services Code Point (DSCP) marking feature for IP headers carrying outgoing
signalling packets.
Product HSGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAG Service Configuration
configure > context context_name > mag-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mag-service)#
default
Restores the specified parameter to its default setting of 0x0.
no
Disables the specified functionality.
ip-header-dscp value
Used to configure the QoS Differentiated Services Code Point (DSCP) marking for IP header encapsulation.
value: Represents the DSCP setting as the first six most-significant bits of the ToS field. It can be configured
to any hexadecimal value from 0x0 through 0x3F. Default is 0x0.
Usage Guidelines Use this command to enable or disable the DSCP marking feature for IP headers carrying outgoing signalling
packets. DSCP marking is disabled by default.
Examples The following command configures the HSGW service to support DSCP marking for IP headers carrying
outgoing signalling packets:
signalling-packets ip-header-dscp 0x21
The MAP Service Configuration Mode is used to configure properties for Mobile Application Part (MAP)
service.
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
access-protocol
Configures access protocol parameters for the MAP service as defined for a specific SCCP network instance.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
sccp-network sccp_id
Specifies the ID number of the SCCP network to use for the SGSN connection.
sccp_id : Must be an integer from 1 to 16.
ssn subsys_num
Identifies the subsystem number for the destination.
subsys_num: Enter an integer from 1 through 255.
no
Removes the access protocol SCCP network instance ID from the configuration.
Usage Guidelines Use this command to associate access protocol parameters to a specific instance of the MAP service for an
SCCP network.
Examples The following command associates the access protocols with the SCCP network ID #10:
access-protocol sccp-network 10
application-context-name
Configure the operation timer(s) for one or more MAP application contexts.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
default
Resets the operation timers for all applications to system defaults.
application
Select one of the following applications to enable the application:
• authentication-failure-report : Sets the reporting operation timer for authentication failure. The setting
range for this timer is 15 to 30 seconds and the default setting is 15 seconds.
• cancel-location : Sets the cancel location operation timer. The setting range for this timer is 15 to 30
seconds and the default setting is 15 seconds.
• check-imei : Sets the check-IMEI operation timer. The setting range for this timer is 15 to 30 seconds
for releases 8.0 and 8.1 and 1 to 30 seconds for releases 9.0 and higher. The default setting is 15 seconds.
• delete-subscriber-data : Sets the delete subscriber data operation timer. The setting range for this timer
is 15 to 30 seconds and the default setting is 15 seconds.
• mo-fwd-sm : Sets the operation timer for fowarding mobile-originated SMS. The setting range for this
timer is 1 to 10 minutes and the default setting is 1 minute (60 seconds).
• ms-purge : Sets the operation timer for MS-purge function. The setting range for this timer is 15 to 30
seconds and the default setting is 15 seconds.
• mt-fwd-sm : Sets the operation timer for fowarding mobile-terminated SMS. The setting range for this
timer is 1 to 10 minutes and the default setting is 1 minute (60 seconds).
• ready-for-sm : Sets the operation timer for the ready for SMS operation. The setting range for this timer
is 15 to 30 seconds and the default setting is 15 seconds.
• send-authentication-info : Sets the operation timer for the sending authentication information operation.
The setting range for this timer is 15 to 30 seconds and the default setting is 15 seconds.
• stand-alone-insert-subscriber-data : Sets the operation timer for the standalone insert subscriber data
operation. The setting range for this timer is 15 to 30 seconds and the default setting is 15 seconds.
• ugl-insert-subscriber-data : Sets the operation timer for the insert subscriber data portion of the update
GPRS location operation. TThe setting range for this timer is 15 to 30 seconds and the default setting
is 15 seconds.
• update-gprs-location : Sets the operation timer for the update GPRS location operation. The setting
range for this timer is 15 to 30 seconds and the default setting is 15 seconds.
operation-timervalue
Configures the operation timer for the selected application. Timer values are indicated above.
Usage Guidelines Repeat this command entering a different application each time to enable multiple applications.
auth-vectors
Configures the number of authorization vectors to be requested from the home location register (HLR) during
call setup to provide subscriber authentication.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
default
Resets the number of vectors requested from the HLR to the system default.
number-to-request number
number: Must be an integer from 1 to 5 to define the number of authorization vectors be requested from the
HLR.
Default is 5.
Usage Guidelines Set the number of requests to be received from the HLR.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
equipment-identity-register
Defines the information relevant to the equipment-identity-register (EIR) used by the SGSN.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
Syntax Description equipment-identity-register { isdn E.164_num | point code pt_code } [ check-imei-every-n-events times|
source-ssn ssn] [ include-imsi | map-include-imsi ]
no equipment-identity-register { isdn E.164_num | point codept_code }] [ include-imsi | map-include-imsi
]
no
Deletes the EIR configuration.
isdn number
Enter the E.164 number of the EIR.
number: must be a string of 1 to 15 digits.
source-ssn ssn
Identifies the subsystem number (SSN) to be used.
check-imei-every-n-events times
Configures the frequency with which a 'check IMEI' message is sent to the EIR. When set, the SGSN skips
sending the 'check IMEI' message for the first N-1 where IMDI/IMEISV is received.
times :
• For releases 8.0 and 8.1, the value must be an integer from 1 to 15.
• For releases 15.0 and higher, the value must be an integer from 1 to 15.
Important This feature requires the enabling of verify-equipment-identity for IMEI or IMEISV as specified with
the the gmm retrieve-equipment-identity imei command of the call-control-profile configuration mode.
include-imsi
Enables inclusion of IMSI checking during the IMEI check procedure. By default this function is not included.
map-include-imsi
Enables the inclusion of IMSI parameter in the CHECK_IMEI Request. By default, IMSI is not included in
the CHECK_IMEI Request.
Usage Guidelines Configure the identity of the EIR that the SGSN uses and the interaction parameters.
Increasing the check-imei-every-n-events frequency enables the EIR to avoid overload as the number of
data-only devices attaching to the network increases.
Examples Configure EIR with point code 1.255.1 to perform IMEI check after every 61st received Attach Request
message:
equipment-identity-register point code 1.255.1 check-imei-every-n-events 62
Configure IMSI to be included in the MAP-CHECK-IMEI operation:
equipment-identity-register point code 1.255.1 check-imei-every-n-events 62 include-imsi
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
gmlc
This command identifies the gateway mobile location center (GMLC) associated with the Location Service
functionality.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
Syntax Description gmlc { isdn E.164_number | point-code point-code } [ gsn-address ipv4 address ] [ source-ssn SSN_value
]
no gmlc { isdn | point-code }
no
Deletes the GMLC configuration. When the no form of the command is used the SGSN aborts all active
Location Requests towards that GMLC and stops the Location Service functions.
isdn number
Enter the E.164 number of the GMLC.
number: must be a string of 1 to 15 digits.
point-code pt_code
Enter SS7 point code address of the GMLC in dotted-decimal format according to variant settings:
• ITU Range 0.0.1 to 7.255.7
• ANSI Range 0.0.1 to 255.255.255
• TTC Range 0.0.1 to 15.31.255
• or a string of 1 to 11 characters
Important Even though gsn-address is optional per the CLI grammar, this keyword is a mandatory parameter when
configuring the GMLC for the location service feature in the SGSN.
source-ssn ssn
Identifies the subsystem number (SSN) to be used.
ssn must be an integer from 1 to 255.
Usage Guidelines This command identifies the GMLC associated with the MAP Service configuration in support of the Location
Services functionality enabled on the SGSN.
Only one GMLC can be defined per MAP Service configuration.
Related Commands:
• associate map-service in the Location Service configuration mode associates this MAP service when
configuration Location Services functionality.
Examples Use a command similar to the following to define a gateway mobility location center (GMLC) with an ISDN
ID of 491720499, a GSN address of 192.168.64.1, and an SSN of 131:
gmlc isdn 491720499 gsn-address 192.168.64.1 ssn 131
hlr
This command enters the configuration mode for the home location register (HLR). The HLR is a database
containing the subscriber profile information for all mobile stations (MS) / user equipment (UE) connecting
to a specific GPRS or UMTS core network.
Important The commands and options for this mode are documented in the HLR Configuration Mode chapter.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
policy
This command configures the Transaction Capabilities Application Part (TCAP) -specific MAP policy for
either ANSI or ITU SS7 variants.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
use-received-destination-address
Selecting this keyword overwrites stored CG and CD addresses with a new address received in first TC CNT
msg
use-received-source-address
Selecting this keyword instructs the MAP service to use the received source address for the dialog.
Usage Guidelines Use this command to determine how TCAP will handle MAP messages.
short-message-service
This command enables and disables the short message service (SMS service) and provides access to the SMS
Service configuration mode.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
no
Disables the SMS service.
Usage Guidelines Enter the command to access the SMS service configuration mode to fine tune the SMS functionality.
Examples short-message-service
timeout
Use this command to configure the m1 timeout value for the LCS procedure..
Product SGSN
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MAP Service Configuration
configure > context context_name > map-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-map-service-service_name)#
m1 seconds
This keyword sets the expiry value for the SGSN's m1 timer, which sets the time the SGSN waits to send a
negative PSL Response and clear the location request.
seconds is an integer from 60 to 600. Default is 120.
Usage Guidelines The m1 timer is used for location service (LCS) specific map message PSL (provide subscriber location). The
gateway mobile location center (GMLC), associated with the LCS functionality, also maintains an m1 timer
when it sends a PSL Request to the SGSN. If the GMLC's timer expires before receiving a response from the
SGSN, then the GMLC aborts the location procedure.
This command enables the operator to determine the amount of time the SGSN should wait before sending a
negative PSL Response and cleaing the location request to complete the LCS procedure.
The Mobile IP HA Assignment Table Configuration Mode is used to assign specific Home Agent (HA) IP
addresses to ranges of Mobile Node (MN) IP addresses.
Command Modes Exec > Global Configuration > Context Configuration > MIP HA Assignment Table
configure > context context_name > mobile-ip ha assignment-table table_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mobile-ip-ha-assignment)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
hoa-range
This command assigns ranges of Mobile Node (MN) IP addresses to specific Home agent IP addresses.
Product HA
Command Modes Exec > Global Configuration > Context Configuration > MIP HA Assignment Table
configure > context context_name > mobile-ip ha assignment-table table_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mobile-ip-ha-assignment)#
no
Removes the specified Home Agent assignment from the assignment table.
ip_address ip_address2
Specifies a range of MN IP addresses. ip_address and ip_address2 must be specified in IPv4 dotted-decimal
or IPv6 colon-separated notation.
ha ip_address3
Specifies the IP address of the Home Agent to assign to MNs that are within the specified range. ip_address3
must be specified in IPv4 dotted-decimal or IPv6 colon-separated notation.
Usage Guidelines Use this command to assign ranges of MN IP addresses to specific HAs.
Important A maximum of eight MIP HA assignment tables can be configured per context with a maximum of eight
MIP HA assignment tables across all contexts.
Important A maximum of 256 non-overlapping hoa-ranges can be configured per MIP HA Assignment table with
a maximum of 256 non-overlapping hoa-ranges across all MIP HA Assignment tables.
Examples The following command assigns any MN IP address that falls in the range of 192.168.100.0 through
192.168.101.0 to the HA with the IP address of 192.168.200.10:
hoa-range 192.168.100.0 192.168.101.0 ha 192.168.200.10
The MPLS-LDP Configuration Mode is used to configure Label Distribution Protocol (LDP) specific
parameters for MPLS-IP forwarding.
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
advertise-labels
Configures the Label Advertisement parameters.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
no
Disables the label advertisement parameters.
default
Advertises the labels from the label space allocated for LDP protocol.
explicit-null
Advertises the Explicit NULL label for all the prefixes.
implicit-null
Advertises the Implicit NULL label for all the prefixes.
Usage Guidelines Use this to configure advertisement of the Implicit NULL or Explicit NULL label for all the prefixes advertised
by the system in this context.
Examples The following command configures the MPLS-IP forwarding to advertise the Explicit NULL label for all the
prefixes:
advertise-labels explicit-null
The following command configures the MPLS-IP forwarding to advertise the Implicit NULL label for all the
prefixes:
advertise-labels implicit-null
discovery
Configures the Label Distribution Protocol (LDP) neighbor discovery parameters.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
Syntax Description discovery { hello { hello-interval integer_value | hold-interval integer_value } | transport-address ipv4_addr
}
default discovery hello
no discovery transport-address
default
Sets the LDP discovery hello interval at 5 seconds and hold interval at 15 seconds.
no
Disables the LDP neighbor discovery.
transport-address ipv4_addr
Configures the LDP transport address as an IPv4 address entered in dotted-decimal notation. Transport address
is the same as the LDP router ID.
Usage Guidelines This is an optional command that is used to configure LDP peer discovery parameters. The LDP discovery
hold-interval is always set to three times the LDP discovery hello-interval. Transport address is the address
used for the TCP session over which LDP is running. If the transport address is not configured, the LDP
router-id is used as transport address. Any update to transport address will take effect only if LDP is disabled
and re-enabled. The "default" option sets the hello intervals to the default values.
Examples The following command sequence configures the LDP peer discovery parameters:
discovery hello hello-interval 10
discovery hello hold-interval 30
discovery transport-address 10.2.3.4
enable
Enables the Label Distribution Protocol (LDP).
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
no
Disables the LDP protocol.
Usage Guidelines This command is used to enable or disable the LDP protocol. By default the LDP protocol is disabled.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
router-id
Configures the Label Distribution Protocol Router ID.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
no
Disables the router ID.
ipv4_addr
Must be an IPv4 address entered in dotted-decimal notation.
Usage Guidelines This command is used to configure the LDP router-id. This is an optional parameter. If the ID is not configured,
the largest operational loopback address is selected as the LDP router ID. If LDP has started, any change will
take effect only after disabling and enabling LDP.
session
Configures the LDP session parameters.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration > MPLS-LDP Configuration
configure > context context_name > mpls-ip > protocol ldp
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ldp)#
default
Configures the default values for hold-interval parameter at 45 and keepalive-interval parameter at 15.
timers
Configures the LDP session keepalive parameters.
hold-interval integer_value
Configures the session hold time interval in seconds.
integer_value is an integer from 45 through 65535.
Default: 45
keepalive-interval integer_value
Configures the frequency of sending keepalive packets in seconds.
integer_value is an integer from 15 through 21845.
Default: 15
Usage Guidelines This optional command is used to configure LDP session timers. LDP session hold-interval is always set to
three times the LDP session keepalive-interval. The "default" option sets the session keepalive and hold
intervals to the default values.
Examples The following command sequence configures the LDP session parameters:
session timers keepalive-interval 30
session timers hold-interval 45
default session timers
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
aaa accounting
Configures the sending of subscriber session AAA accounting by the Home Agent (HA) service.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
no
Disables AAA accounting for the HA service.
Usage Guidelines Enabling the HA service will send all accounting data (start, stop, and interim) to the configured AAA servers.
The default is AAA accounting enabled.
Important In order for this command to function properly, AAA accounting must be enabled for the context in which
the HA service is configured using the aaa accounting subscriber radius command.
AAA accounting for the HA service can be disabled using the no version of the command.
Examples The following command disables AAA accounting for the HA service:
no aaa accounting
bind
Designates the address of the MIPv6HA service and specifies the maximum number of subscribers that can
access this service over the interface.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
no
Removes the bound address from the HA service.
address
Specifies the IPv6 address of the MIPv6HA service using IPv6 colon-separated-hexadecimal notation.
max-subscribers count
Default: 3000000
Specifies the maximum number of subscribers that can access this service on this interface.
count is an integer from 0 through 4000000.
Important The maximum number of subscribers supported depends on the installed license key and the number of
active packet processing cards installed in the system. Refer to the license key command for additional
information.
Usage Guidelines Use this command to associate the HA service with a specific logical IP address. The logical IP address or
interface takes on the characteristics of a Pi interface. Only one interface can be bound to a service. The
interface should be configured prior to issuing this command.
This command also sets a limit as to the number of simultaneous subscribers sessions that can be facilitated
by the service/interface at any given time.
When configuring the max-subscribers option, be sure to consider the following:
• The total number of interfaces that you will configuring for use as Pi interfaces
• The maximum number of subscriber sessions that all of these interfaces may handle during peak busy
hours
• The average bandwidth for each of the sessions
• The type of physical port to which these interfaces will be bound
Taking these factors into account and distributing your subscriber session across all available interfaces will
allow you to configure your interfaces to optimally handle sessions without degraded performance.
Examples The following command binds the logical IP interface with the address of 2001:4A2B::1f3F to the HA service
and specifies that a maximum of 600 simultaneous subscriber sessions can be facilitated by the interface/service
at any given time.
bind address 2001:4A2B::1f3F max-subscribers 600
The following command disables a binding that was previously configured:
no bind address
default
Restore default values assigned for specified parameter.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
aaa
Restores the AAA setting configured by the aaa command to its default of enabled.
refresh-advice-option
Restores the refresh-advice-option setting to its default of disabled.
refresh-interval-percent
Restores the refresh-interval-percent setting to its default of 75.
reg-lifetime
Restores the Mobile IP session registration lifetime setting configured by the reg-lifetime command to its
default: 600 seconds.
sequence-number-validate
Restores the sequence-number-validate setting to its default of enabled.
setup-timeout
Restore the maximum amount of time allowed for setting up a session to the default: 60 seconds.
simul-bindings
Restores the simultaneous bindings setting to its default: 1.
subscriber
Configures settings for the default subscriber.
timestamp-replay-protection
Restores the timestamp-replay-protection scheme according to RFC 4285.
Usage Guidelines After the system has been modified from its default values, this command is used to set or restore specific
parameters to their default values.
Examples The following command is used to return the simultaneous bindings setting parameter to it's default value:
default simul-bindings
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
refresh-advice-option
Configures inclusion of refresh advice option in the Binding Acknowledgement sent by the Home Agent
(HA).
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
Usage Guidelines Includes the refresh advice option in the binding acknowledgements sent by the home agent. Default is disabled.
refresh-interval-percent
Configures the percentage of the granted lifetime to be used in the refresh interval mobility option in the
Binding Acknowledgement sent by the Home Agent (HA).
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
value
value represents a percentage expressed as an integer from 1 through 99. Default is 75.
Usage Guidelines Use this command to configure the amount of the granted lifetime to be used in the refresh interval mobility
option in the Binding Acknowledgement sent by the Home Agent (HA).
reg-lifetime
Specifies the longest registration lifetime that the HA service will be allowed in any Registration Request
message from the mobile node.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
no
Sets the registration lifetime to infinite.
time
Specifies the registration lifetime in seconds. time is an integer from 1 through 262140. Default is 600.
Usage Guidelines Use to limit a mobile nodes' lifetime. If the mobile node requests a shorter lifetime than what is specified, it
is granted. However, Per RFC 2002, should a mobile node request a lifetime that is longer than the maximum
allowed by this parameter, the HA service will respond with the value configured by this command as part
of the Registration Reply.
Examples The following command configures the registration lifetime for the HA service to be 2400 seconds:
reg-lifetime 2400
The following command configures an infinite registration lifetime for MIPv6 calls:
no reg-lifetime
sequence-number-validate
Configures sequence number validation of the received MIPV6 control packet by the Home Agent (HA)
according to RFC 3775.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
Usage Guidelines Use this command to enable sequence number validation of the received MIPV6 control packet by the Home
Agent (HA) as per RFC 3775. Default is enabled.
setup-timeout
The maximum amount of time allowed for session setup.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
seconds
Default: 60 seconds
The maximum amount of time (in seconds) to allow for setup of a session expressed as an integer from 1
through 1000000. Default is 60 seconds.
Usage Guidelines Use this command to set the maximum amount of time allowed for setting up a session.
Examples To set the maximum time allowed for setting up a session to 5 minutes (300 seconds), enter the following
command:
setup-timeout 300
simul-bindings
Specifies the maximum number of "care-of" addresses that can be simultaneously bound for the same user as
identified by NAI and Home address.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
number
Configures maximum number of "care of" addresses that can be simultaneously bound for the same user as
identified by their NAI and home address. number is an integer from 1 through 3. Default is 1.
Usage Guidelines Per RFC 2002, the HA service creates a mobile binding record (MBR) for each subscriber session it is
facilitating. Each MBR is associated with a care-of address. As the mobile node roams, it is possible that the
session will be associated with a new care-of address.
Typically, the HA service will delete an old binding and create a new one when the information in the
Registration Request changes. However, the mobile could request that the HA maintains previously stored
MBRs. This command allows you to configure the maximum number of MBRs that can be stored per subscriber
if the requested.
Examples The following command configures the service to support up to 2 addresses per subscriber:
simul-bindings 2
timestamp-replay-protection tolerance
Designates timestamp replay protection scheme as per RFC 4285.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > MIPv6HA Service Configuration
configure > context context_name > mipv6ha-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mipv6ha-service)#
tolerance seconds
Defines the acceptable difference in timing (between timestamps) before rejecting packet, in seconds. seconds
must be an integer from 0 through 65535. The default is 7.
Usage Guidelines Use this command to define the acceptable difference in timing (between timestamps) before rejecting packet.
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate
Associates or disassociates supportive services and templates with the MME-eMBMS service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
no
Disassociates a previously associated service or template from this MME-eMBMS service configuration.
egtp-service service_name
Specifies the name for a pre-configured eGTP service to associate with the MME-eMBMS service. The eGTP
service provides eGTP-C protocol interface support between EPS nodes. Only one eGTP service can be
associated with an MME-eMBMS service.
The eGTP service should be configured prior to issuing this command. For more information about the eGTP
service, refer to the egtp-service command in the Context Configuration Mode Commands chapter and the
eGTP Service Configuration Mode Commands chapter.
service_name is an alphanumeric string of 1 through 63 characters.
Important Service names must be unique across all contexts within a chassis.
context ctxt_name
Identifies a specific context name where the named eGTP service is configured. If this keyword is omitted,
the eGTP service must exist in the same context as the MME-eMBMS service.
ctxt_name is an alphanumeric, case-sensitive string of 1 through 63 characters.
sctp-param-template template_name
Associates a Stream Control Transmission Protocol (SCTP) parameter template with this MME-eMBMS
service. For more information on the SCTP parameter template, refer to the sctp-param-template command
in the Global Configuration Mode Commands chapter and the SCTP Parameter Template Configuration Mode
Commands chapter.
template_name specifies the name for a pre-configured SCTP parameter template to associate with this
MME-eMBMS service. The name entered must be an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to associate a pre-configured eGTP service or SCTP parameter template with the
MME-eMBMS service.
Examples The following command disassociates a previously configured association between the MME-eMBMS service
and the already configured egtp1 eGTP service :
no associate egtp-service egtp1
bind
Binds the MME-eMBMS service to a logical IP interface serving as the M3 interface.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
Syntax Description bind { ipv4-address address [ ipv4-address secondary_address ] | ipv6-address address [ ipv6-address
secondary_address ] }
no bind
no
Removes a previously configured IP address used for binding the SCTP (local bind address) to communicate
with the eNodeBs using an S1-MME interface.
Usage Guidelines Use this command to associate the MME-eMBMS service with a specific logical IP address that will be used
for binding the SCTP socket that communicates with the eNodeB using M3. Only one IP address can be
configured with this command for one MME service.
Examples The following command would bind the logical IP interface with the address of 192.168.3.1 to the
MME-eMBMS service to interact with eNodeB:
bind ipv4-address 192.168.3.1
The following command disables a binding that was previously configured:
no bind
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
mmemgr-recovery
Sets the action the MME takes regarding the peers (MCEs) upon recovery after an MME Manager crash/failure.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
no-reset
Configures the MME-eMBMS service not to have the MME perform a reset of peer associations upon recovery
of the MMEMgr after a manager crash/failure.
This is the default setting.
reset-peers
Configures the MME-eMBMS service to have the MME perform a reset of peer associations upon recovery
of the MMEMgr after a manager crash/failure.
Usage Guidelines If the MMEMgr crashes or fails, the configuration (defined with this command) instructs the MME what
actions to take at the time of recovery in reference to the peer association.
Examples The following command instructs the MME not to reset associations :
mmemgr-recovery no-reset
plmn-id
Configures the carrier's Public Land Mobile Network (PLMN) identifier to associate with this eMBMS service
area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
mcc mcc
Specifies the mobile country code (MCC) portion of the PLMN identifier. The mcc must be an integer from
100 through 999.
mnc mnc
Specifies the mobile network code (MCC) portion of the PLMN identifier. The mnc must be a 2- or 3-digit
integer from 00 through 999.
Usage Guidelines Use this command to specify the PLMN identifier to associate with the eMBMS area for this MME's eMBMS
service.
Examples The following command configures the PLMN identifier with an MCC of 462 and MNC of 02:
plmn id mcc 462 mnc 02
sctp port
Configures the SCTP port number to be associated with the M3AP interface of the eMBMS service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
port_number
Enter an integer from 1 to 65535. The default is 36412.
Usage Guidelines Use this command to identify the SCTP port for the M3AP interface.
Examples The following command configures sctp port 34414 as the sctp port number associated with the M3AP interface
:
sctp port 34414
setup-timeout
Specifies the maximum amount of time allowed for session setup.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME-eMBMS Service
configure > context context_name > mme-embms-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-embms-service)#
seconds
The maximum amount of time, in seconds, to allow for setup of a session. Where seconds must be an integer
from 1 through 10000. The default is 60.
Usage Guidelines Use this command to set the maximum amount of time allowed for setting up a session.
Examples The following command sets the maximum setup time as 120 seconds:
setup-timeout 120
Command Modes Exec > Global Configuration > Context Configuration > SGs Service Configuration > MME LAC Pool Area
Configuration
configure > context context_name > sgs-service service_name > pool-area pool_area_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-pool-area)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
hash-value
Configures the Visitor Location Register (VLR) hash value mapping for this pool area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > SGs Service Configuration > MME LAC Pool Area
Configuration
configure > context context_name > sgs-service service_name > pool-area pool_area_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-pool-area)#
Syntax Description hash-value { value | non-configured-values | range value to value } use-vlr vlr_name
no hash-value { value | non-configured-values | range value to value }
no
Removes the configured hash-value from the pool-area configuration.
value
Specifies the VLR hash value to be used with the configured VLR. value must be an integer from 0 through
999.
non-configured-values
Specifies that the VLR configured in this command is to be used non-configured hash values.
use-vlr
Specifies the VLR to be used with the hash value configuration in this command.
Usage Guidelines Use this command to configure hash values to be used with VLRs.
In Release 12.2 and later, a maximum of 48 hash lists can be created per pool area. In older releases, a total
of 32 hash lists can be created per pool area.
In a pool configuration, the MME selects the VLR that corresponds to the hash of the UE's IMSI. If that VLR
is inactive, the MME will use the default VLR (as defined by the non-configured-value option). If no default
VLR has been configured, or if the default VLR is inactive, the MME selects any other available VLR from
the pool.
If the chosen VLR is active at the time of selection and then subsequently becomes inactive when the request
is sent to it, the current request fails. On the next request from the UE, the VLR selection mechanism is applied
again. A VLR that failed previously will only be selected again if it became active since the earlier failure.
Examples The following command configures all hash values within a range of 0 to 500 to use a VLR named vlr1:
hash-value range 0 to 500 use-vlr vlr1
The following command configures hash values of 501 to use a VLR named vlr2:
hash-value 501 use-vlr vlr2
The following command configures all non-configured hash values to use a VLR named vlr3:
hash-value non-configured-values use-vlr vlr3
lac
Configures a 3G location area code or area codes that define this pool area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > SGs Service Configuration > MME LAC Pool Area
Configuration
configure > context context_name > sgs-service service_name > pool-area pool_area_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-pool-area)#
no
Removes a configured forbidden handover area code or area codes from this policy. If no location area code
is specified, then all location area codes are removed.
area_code
Specifies an area code or area codes used to select a VLR for the pool area as an integer from 0 through 65535.
Multiple area codes can be entered (up to 16 in a single line, separated by spaces).
Usage Guidelines Use this command to configure 3G location-based area codes that define this pool area.
In Release 12.2 and later, a maximum of 96 areas can be added per pool area (in a single line, or separately).
In older releases, a total of 16 area codes can be added (in a single line, or separately).
Examples The following command configures eight location-based area codes (1, 2, 3, 4, 5, 6, 7, 8) that define this pool
area:
lac 1 2 3 4 5 6 7 8
plmnid
Configures the Public Land Mobile Network (PLMN) identifier for the LAC pool area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > SGs Service Configuration > MME LAC Pool Area
Configuration
configure > context context_name > sgs-service service_name > pool-area pool_area_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-pool-area)#
no
Removes the configured PLMN identifier for the LAC pool area.
any
This keyword specifies any PLMN identifier can be configured for the LAC pool area.
mcc mcc_value
Specifies the mobile country code (MCC) portion of the PLMN identifier as an integer from 100 through 999.
mnc mnc_value
Specifies the mobile network code (MNC) portion of the PLMN identifier as a 2- or 3-digit integer from 00
through 999.
Usage Guidelines Use this command to set the PLMN identifier for the LAC pool area. Any PLMN identifier can be can be
configured for the LAC pool area or a specific PLMN identifier can be configured by providing the MCC and
MNC of the PLMN identifier.
Examples The following command configures the PLMN identifier with MCC value as 102 and MNC value as 20 for
this MME service:
plmnid mmc 102 mnc 20
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
hash-value
Configures the selection of MSC in a MSC pool area based on the hash value derived from the IMSI.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service Configuration > Pool Area Configuration
configure > context context_name > mme-service service_name > pool-area pool_area_name type hash-value
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-pool-area-hash-value)#
no
Removes the configured hash value for this pool area.
hash-value
Specifies the specific hash value for this pool area.
hash_value must be an integer from 0 through 999.
use-msc msc_name
Specifies the MSC to use when this pool area is selected.
msc_name is the name of the MSC as configured in the MME Service using the msc command; msc_name
must be an alphanumeric string of 1 to 39 characters.
Usage Guidelines This command associates an MSC with this pool. It also assigns an the MSC to use based on the hash value
as computed from the IMSI digits [(IMSI div 10) modulo 1000].
A maximum of 24 hash values can be configured within each pool area.
If no matching MSC is found, the SRVCC handover fails.
Examples The following command configures hash values from 111 to 222 to use the MSC named mscwest1 in this
pool.
hash-value range 111 to 222 use-msc mscwest1 arg1
plmn-id
Associates a Public Land Mobile Network (PLMN) identifier with a Mobile Switching Center (MSC) pool
area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service Configuration > Pool Area Configuration
configure > context context_name > mme-service service_name > pool-area pool_area_name type hash-value
or
configure > context context_name > mme-service service_name > pool-area pool_area_name type
round-robin
Entering the above command sequences result in the following prompts, respectively:
[context_name]host_name(config-mme-pool-area-hash-value)#
[context_name]host_name(config-mme-pool-area-round-robin)#
no
Removes the configured plmn-id assigned to this MSC pool area.
mcc code
Specifies the Mobile Country Code for this mobile access network. code must be a three-digit integer from
200 to 999.
mnc code
Specifies the Mobile Network Code for this mobile access network. code must be a two or three-digit integer
from 00 to 999.
Usage Guidelines Use this command to associate a PLMN with an MSC pool area. This PLMN is used to select an MSC pool
area based on the target PLMN as specified in the SRVCC handover request.
When configured, the MME attempts to select an MSC using the following selection order:
Examples The following command identifies the mobile network with a MCC of 123 and a MNC of 12.
plmn-id mcc 123 mnc 12
use-msc
Associates a Mobile Switching Center (MSC) with the pool area.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service Configuration > Pool Area Configuration
configure > context conext_name > mme-service service_name > pool-area pool_area_name type
round-robin
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-pool-area-round-robin)#
no
Removes the associated MSC name from this pool area.
use-mscmsc_name
Associates an MSC name with this pool area.
msc_name is the name of the MSC as configured in the MME Service using the msc command; msc_name
must be an alphanumeric string of 1 to 39 characters.
Usage Guidelines This command associates an MSC with this pool area. With a round-robin pool area selection, the MME
selects the next MSC within the pool based on a round-robin scheme.
A maximum of 24 MSC associations can be defined within each round-robin pool area.
Examples The following command associates the MSC named mscsouth1 to this pool.
use-msc mscsouth1
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate
Associates or disassociates a Stream Control Transmission Protocol (SCTP) parameter template with the SGs
service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
no
Removes the SCTP parameter template association.
template_name
Associates an SCTP parameter template with the SGs service.
template_name specifies the name for a pre-configured SCTP parameter template to associate with this SGs
service as an alphanumeric string of 1 through 63 characters. For more information on the SCTP parameter
template, refer to the sctp-param-template command in the Global Configuration Mode Commands chapter
and the SCTP Parameter Template Configuration Mode Commands chapter.
Usage Guidelines Use this command to associate a pre-configured SCTP parameter template with the SGs service.
Caution This is a critical configuration. Any change to this configuration will cause the SGs service to restart.
Important If no SCTP parameter template is specified, all default settings for the configurable parameters in the
SCTP Parameter Template Configuration Mode apply.
Examples The following command associates a pre-configured SCTP parameter template called sctp-3 to the SGs service:
associate sctp-param-template sctp-3
bind
Binds the service to a logical IP interface serving as the SGs interface.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Syntax Description bind { ipv4-address ipv4_address [ ipv4-address ipv4_address ] | ipv6-address ipv6_address [ ipv6-address
ipv6_address ] }
no bind
no bind
Removes the interface binding from this service.
Usage Guidelines Associate the SGs service to a specific logical IP address. The logical IP address or interface takes on the
characteristics of an SGs interface that provides the session connectivity for circuit switched fallback (CSFB)
to/from a Mobile Switching Center/Visitor Location Register (MSC/VLR). Only one interface can be bound
to a service. The interface should be configured prior to issuing this command
Caution This is a critical configuration. Any change to this configuration will cause the SGs service to restart.
Removing or disabling this configuration will stop the SGs service.
Examples The following command binds the logical IP interface with the IPv4 address of 192.68.10.22 to the SGs
service:
bind ipv4-address 192.68.10.22
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip
This command configures the IP parameters on the SGs interface.
Product MME
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
no
Removes IP parameter configuration from the SGs service/interface.
qos-dscp dscp_value
The qos-dscp keyword designates the Quality of Service - Differentiated Services Code Point value to the
packet leaving through the SGs interface.
dscp_value is a value assigned to the packet for DSCP marking. The value can be a pre-defined DSCP value
or an arbitrary value ranging from 0x01 to 0x3F.
Usage Guidelines SGs interface allows Differentiated Services Code Point (DSCP) marking functionality. DSCP marking helps
in packet traffic management. DSCP marking can be performed on both IPv4 and IPv6 packets leaving the
SGs interface.
Either the pre-defined DSCP values can be used for marking, or any arbitrary value ranging from 0x01 to
0x3F can be assigned. The default DSCP value is 0x00 or be (Best Effort). The default DSCP value is
automatically set when the configuration is disabled.
Examples The following command shows the IP configuration for DSCP marking on the SGs service.
ip qos-dscp ef
non-pool-area
Configures a non-pool area where a group of Location Area Code (LAC) values use a specific visitor Location
Register (VLR).
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Syntax Description non-pool-area name use-vlr vlr_name [ lac value(s) ] [ plmnid { any | mcc mcc_value mnc mnc_value } ]
no non-pool-area name [ lac value(s) ]
name
Specifies the name of the non-pool area as an alphanumeric string of 1 through 63 characters.
use-vlr vlr_name
Specifies the VLR to be used in this non-pool area configuration as an alphanumeric string of 1 through 63
characters.
lac value(s)
Specifies the location area code or codes to be used with the configured VLR in this non-pool area configuration.
value(s) is an integer(s) from 1 through 65535.
In Release 12.2 and later, a maximum of 96 areas can be added per non pool area (in a single line, or separately).
In older releases, a total of 16 area codes can be added (in a single line, or separately).
Usage Guidelines Use this command to configure a non-pool area where LAC values and/or PLMN IDs are associated with a
specific VLR.
In Release 12.2 and later, a maximum of 48 combined non pool areas and pool areas can be created. In older
releases, a maximum of 8 combined non pool areas and pool areas can be created.
Examples The following command creates a non-pool area named svlr1 associated with a VLR named vlr1 and containing
LAC values of 1, 2, 3, 4, 5, 6, 7, and 8:
non-pool-area svlr1 use-vlr vlr1 lac 1 2 3 4 5 6 7 8
pool-area
Creates a location area code (LAC) pool area configuration or specifies an existing pool area and enters the
LAC Pool Area Configuration Mode.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
no
Removes the selected pool area configuration from the SGs service.
pool_name
Specifies the name of the LAC pool area configuration. If pool_name does not refer to an existing pool, a
new pool is created. pool_name must be an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to enter the LAC Pool Area Configuration Mode for an existing pool area configuration
or for a newly defined pool area configuration. This command is also used to remove an existing pool area
configuration.
In Release 12.2 and later, a maximum of 48 combined pool areas and non pool areas can be created. In older
releases, a maximum of 8 combined pool areas and non pool areas can be created.
Entering this command results in the following prompt:
[context_name]hostname(config-sgs-pool-area)#
LAC Pool Area Configuration Mode commands are defined in the MME LAC Pool Area Configuration Mode
Commands chapter.
Examples The following command enters the LAC Pool Area Configuration Mode for a new or existing pool area
configuration named pool1:
pool-area pool1
sctp
Configures the Stream Control Transmission Protocol (SCTP) port number for this service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
no
Removes the SCTP configuration for this service.
port port_number
Specifies the SCTP port number used to communicate with the MSC/VLR using the SGs interface as an
integer from 1 through 65535.
Usage Guidelines Use this command to assign the SCTP port with SCTP socket to communicate with the MSC/VLR through
the SGs interface. A maximum of one SCTP port can be associated with one SGs service.
Examples The following command sets the SCTP port to 29118 for this service:
sctp port 29118
tac-to-lac-mapping
Maps any Tracking Area Code (TAC) value or a specific TAC value to a LAC value.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Usage Guidelines Use this command to map TAC values to LAC values.
Enter up to 8 mappings per line.
In Release 12.2 and later, a maximum of 64 mapping lists can be created. In older releases, a maximum of 32
mapping lists can be created.
If no mapping is entered, the default behavior is TAC equals LAC.
Examples The following command maps a TAC value of 2 to a LAC value of 3, a TAC value of 4 to a LAC value of 5,
and a TAC value of 6 to a LAC value of 7:
tac-to-lac-mapping tac 2 map-to lac 3 tac 4 map-to lac 5 tac 6 map-to lac 7
timer
Configures the SGs-AP timer values.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Syntax Description timer { ts6-1 value | ts10 value | ts12-1 value | ts12-2 value | ts13 value | ts8 value | ts9 value }
default timer { ts10 | ts12-1 | ts12-2 | ts13 | ts6-1 | ts8 | ts9 }
ts10 value
Specifies the Ts10 timer value (in seconds) as an integer from 1 through 30. This timer is used to guard the
Implicit IMSI detach from non-EPS services procedure.
Default: 4.
ts12-1 value
Specifies the Ts12-1 timer value (in seconds) as an integer from 8 through 23048. This timer is used to control
the reset of the 'MME-Reset' variable. It is expected to take a value greater than the longest periodic tracking
area update timer running on the MME, plus the transmission delay on the radio interface.
Default: 36000.
ts12-2 value
Specifies the Ts12-2 timer value (in seconds) as an integer from 1 through 120. This timer is used to guard
the MME reset procedure. There is one Ts12-2 timer per VLR for which the MME has an SGs association.
Default: 4.
ts13 value
Specifies the Ts13 timer value (in seconds) as an integer from 1 through 30. This timer configures the
retransmission interval for sending SGs message SGsAP-EPS-DETACH-INDICATION to MSC/VLR due
to an Implicit IMSI detach from EPS services. If no SGsAP-EPS-DETACH-ACK is received, the MME will
resend SGsAP-EPS-DETACH-INDICATION message upon expiry of this timer.
Default: 4.
ts6-1 value
Specifies the Ts6-1 timer value (in seconds) as an integer from 10 through 90. This timer is used to guard the
Location Update procedure. It is expected to take a value greater than 2 times the maximum transmission time
in the SGs interface, plus the supervision timer of the Update Location procedure (as defined in 3GPP TS
29.002 [15]).
Default: 15.
ts8 value
Specifies the Ts8 timer value (in seconds) as an integer from 1 through 30. This timer is used to guard the
Explicit IMSI detach from EPS services procedure.
Default: 4.
ts9 value
Specifies the Ts9 timer value (in seconds) as an integer from 1 through 30. This timer guards the Explicit
IMSI detach from non-EPS services procedure.
Default: 4.
Examples The following command sets the SGs-AP Ts6-1 timer to 20 seconds:
timer ts6-1 20
vlr
Configures the Visitor Location Register (VLR) to be used by this service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Syntax Description vlr vlr_name { ipv4-address ipv4_address [ ipv4-address ipv4_address ] | ipv6-address ipv6_address [
ipv6-address ipv6_address ] } port port_number
no vlr vlr_name
no
Removes the configured VLR from this service.
vlr_name
Specifies the name of the VLR as an alphanumeric string of 1 through 63 characters.
port port_number
Specifies the SCTP port number as an integer from 1 through 65535.
Usage Guidelines Use this command to configure the VLR used by this SGs service.
In Release 12.2 and later, a maximum of 48 separate VLRs can be created. In older releases, a maximum of
32 separate VLRs can be created.
Each individual VLR can be defined with up to 10 separate associations to a single MSS pool. Each of these
associations support SCTP multi-homing by defining a primary/secondary IP address. Application layer
messages are transmitted to the first available association for a particular VLR. If a complete failure of the
underlying SCTP layer for a given association (for example, both SCTP paths in a multi-homed configuration)
occurs, the VLR association is removed as a candidate for application message transmission until it recovers.
A given MMS (VLR) will remain available as long as at least one related association remains available.
Examples The following command configures a VLR to be used by this service with a name of vlr1, with an SCTP
multi-homed primary IPv4 address of 10.2.3.4, a secondary IPv4 address of 10.2.10.1, and a port number of
29118:
vlr vlr1 ipv4-address 10.2.3.4 ipv4-address 10.2.10.1 port 29118
vlr-failure
Configures automatic VLR failure handling for the SGs service
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME SGs Service Configuration
configure > context context_name > sgs-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-sgs-service)#
Syntax Description [ no ] vlr-failure duration minutes backoff-timer seconds detach-rate number [ -noconfirm ]
no
Removes the configuration from this service, which disables automatic detection and offload of VLRs when
an SGs association failure occurs.
duration minutes
Specifies the amount of time in minutes during which all qualifying UEs will be detached.
The MME splits this duration into n intervals, 5 seconds apart. For example, a setting of 2 minutes with 100
subscribers would result in the MME processing all subscribers in the first 2 intervals (10) seconds. Any
subscribers remaining at the expiry of the duration will not be processed.
If no detach rate is defined using the detach-rate keyword, a maximum of 50 subscribers are processed per
interval. Any remaining UEs will remain attached until detached by other means (UE/network detach, etc).
minutes must be an integer from 1 through 3000.
backoff-timer seconds
Specifies the period of time the MME will wait following the detection of a VLR condition before starting
the controlled release of affected UEs. The MME begins offloading UEs following the expiry of this backoff
timer. If the VLR has recovered before the backoff timer expires, no offloading is performed.
seconds must be an integer from 1 to 3000.
detach-rate number
This optional keyword specifies a maximum number of detaches to perform per 5 second cycle.
For example, if 12,000 subscribers are to be detached during a 5 minute window (duration = 5 minutes), the
MME calculates 60 cycles (5 minutes / 5-second cycles) which results in 200 UEs to detach per cycle.
If the detach-rate is configured to 100, the MME will only detach 100 per 5 second cycle for, resulting in a
total of 6000 detaches. Any remaining UEs will remain attached until detached by other means (UE/network
detach, etc).
number must be an integer from 1 to 2000.
Usage Guidelines This command requires that a valid MME Resiliency license key be installed. Contact your Cisco Account
or Support representative for information on how to obtain a license.
This command configures the MME to automatically initiate the VLR offload feature when a SGs association
failure is detected.
This command provides equivalent functionality to the sgs vlr-failure Exec Mode command. The differences
are that the Exec Mode command must be applied manually, while this Config Mode command is applied
automatically when a failure condition is detected.
Both commands cannot be enabled simultaneously. An error message is reported to the operator if this is
attempted.
Examples The following command enables automatic SGs failure handling functionality. After detecting an SGs
association failure, the MME will wait 180 seconds before starting to detach UEs over a 60 minute window,
without exceeding a detachment rate of 100 UEs per 5-second cycle.
vlr-failure duration 60 backoff-timer 180 detach-rate 100
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
Caution Restarting the MME service leads to termination of UE sessions at the MME, purge of subscriber data
and closure of all connections towards peer nodes such as eNodeB, HSS, S-GW, etc. It may also lead to
termination of other services associated with the MME. It is strongly advised to make any configuration
changes that restarts the service only while in maintenance mode or at startup.
associate
Associates or disassociates supportive services and policies, such as an Evolved GPRS Tunnelling Protocol
(eGTP) service, an HSS peer service, or an MME policy subscriber map with an MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Disassociates a previously associated service with this MME service.
egtp-service egtp_svc_name
Associates an eGTP service with MME service.
egtp_svc_name specifies the name for a pre-configured eGTP service to associate with the MME service. The
eGTP service provides eGTP-C protocol interface support between EPS nodes. For more information on the
eGTP service, refer to the egtp-service command in the Context Configuration Mode Commands chapter and
the eGTP Service Configuration Mode Commands chapter.
Only one eGTP service can be associated with a service. The eGTP service should be configured prior to
issuing this command.
egtp-sv-service egtp_sv_svc_name
Associates an eGTP Sv service with this MME service.
egtp_sv_svc_name specifies the name for a pre-configured eGTP Sv service to associate with the MME service.
For more information on the eGTP Sv service, refer to the egtp-service command in the Context Configuration
Mode Commands chapter.
foreign-plmn-guti-mgmt-db db_name
Associates a Foreign PLMN GUTI management database with this MME service.
db_name specifies the name for a pre-configured foreign PLMN GUTI management database to associate
with the MME service. For more information on the Foreign PLMN GUTI management database, refer to the
foreign-plmn-guti-mgmt-db command in the LTE Policy Configuration Mode Commands chapter.
Only one Foreign PLMN GUTI management database can be associated to an MME service. The Foreign
PLMN GUTI management database should be configured prior to issuing this command.
Multiple MME services can be associated to the same Foreign PLMN GUTI management database.
henbgw-mgmt-db db_name
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release 20.0. For more
information, contact your Cisco account representative.
Associates the specified HeNB-GW management database with the MME service.
db_name specifies the name for an LTE MME HeNB-GW Management Database to associate with the MME
service as an alphanumeric string of 1 through 64 characters. This is required to support S1 HANDOVERs
to Home eNodeBs connected via a HeNB-GW.
hss-peer-service hss_svc_name
Associates an HSS peer service with this MME service.
hss_svc_name specifies the name for a pre-configured HSS peer service to associate with the MME service
as an alphanumeric string of 1 through 64 characters. The HSS peer service provides S6a and S13 interface
support via the Diameter protocol between the MME and an HSS (S6a) or EIR (S13). For more information
about the HSS peer service, refer to the hss-peer-service command in the Context Configuration Mode
Commands chapter and the HSS Peer Service Configuration Mode Commands chapter.
Only one HSS peer service can be associated to a service. The HSS peer service should be configured prior
to issuing this command.
ipne-service ipne_svc_name
Associates an IPNE service with this MME service.
ipne_svc_name must be an alphanumeric string of 1 to 63 characters to identify a pre-configured,
uniquely-named IPNE service. For more information about the IPNE service, refer to the sections for the
IPNE Service Configuration Mode Commands and the IPNE Endpoint Configuration Mode Commands.
location-service location_svc_name
Associates a location service with this MME service. Only one location service should be associated with an
MME Service.
location_svc_name specifies the name for a pre-configured location service to associate with the MME service
as an alphanumeric string of 1 through 64 characters. For more information about Location Services (LCS),
refer to the location-service command in the Context Configuration Mode Commands chapter and the Location
Services Configuration Mode Commands chapter.
lte-emergency-profile profile_name
Associates an LTE emergency profile with this MME service.
profile_name specifies the name for a pre-configured LTE emergency profile to associate with the MME
service as an alphanumeric string of 1 through 64 characters. For more information about the LTE emergency
profile, refer to the lte-emergency-profile command in the LTE Policy Configuration Mode Commands
chapter and the LTE Emergency Profile Configuration Mode Commands chapter.
network-global-mme-id-mgmt-db
Associates the configured global MME ID management database with this MME service. The global MME
ID management database is configured through the LTE Policy Configuration Mode using the
network-global-mme-id-mgmt-db command.
sbc-service sbc_svc_name
Important Beginning with Release 18.4, this keyword is only accessible or configurable if a valid SBc license key
is installed. For information about obtaining such a license, contact your Cisco Representative.
For more information about the SBc service, refer to the sbc-service command in the Global Configuration
Mode Commands chapter, the SBc Service Configuration Mode Commands chapter, and the Cell Broadcast
Center - SBc Interface feature chapter in the MME Administration Guide.
sctp-param-template template_name
Associates a Stream Control Transmission Protocol (SCTP) parameter template with this MME service.
template_name specifies the name for a pre-configured SCTP parameter template to associate with this MME
service as an alphanumeric string of 1 through 63 characters. For more information on the SCTP parameter
template, refer to the sctp-param-template command in the Global Configuration Mode Commands chapter
and the SCTP Parameter Template Configuration Mode Commands chapter.
sgs-service sgs_svc_name
Associates an SGs service with this MME service.
sgs_svc_name specifies the name for a pre-configured SGs service to associate with the MME service as an
alphanumeric string of 1 through 64 characters. For more information on the SGs service, refer to the sgs-service
command in the Context Configuration Mode Commands chapter and the MME SGs Service Configuration
Mode Commands chapter.
sgtpc-service sgtpc_svc_name
Associates an SGTPC service with this MME service.
sgtpc_svc_name specifies the name for a pre-configured SGTPC service to associate with the MME service
as an alphanumeric string of 1 through 64 characters.
Important When co-locating an SGSN and MME, the MME Service cannot be associated with the same SGTP service
that is used by the SGSN.
For more information on the SGTPC service, refer to the sgtp-service command in the Context Configuration
Mode Commands chapter and the SGTP Service Configuration Mode Commands chapter.
context ctx_name
Identifies a specific context name where the named service is configured. If this keyword is omitted, the named
service must exist in the same context as the MME service.
ctx_name is name of the configured context of the named service expressed as an alphanumeric string of 1
through 63 characters that is case sensitive.
subscriber-map map_name
Associates this MME service with a pre-configured subscriber map.
map_name specifies the name of a pre-configured subscriber map to associate with the MME service as an
alphanumeric string of 1 through 64 characters. For more information on subscriber maps, refer to the
subscriber-map command in the LTE Policy Configuration Mode Commands chapter and the LTE Subscriber
Map Configuration Mode Commands chapter.
tai-mgmt-db database_name
Associates this MME service with a pre-configured TAI Management Database.
database_name specifies the name of a pre-configured TAI Management Database to associate with the MME
service as alphanumeric string of 1 through 64 characters. For more information on subscriber maps, refer to
the tai-mgmt-db command in the LTE Policy Configuration Mode Commands chapter and the LTE TAI
Management Database Configuration Mode Commands chapter.
Usage Guidelines Use this command to associate a pre-configured service or policy with an MME service.
Caution This is a critical configuration. The MME service cannot be started without this configuration. Any change
to this configuration will cause the MME service to be restarted. Removing or disabling this configuration
will stop the MME service.
Examples The following command associates a pre-configured eGTP service called egtp1 in the dst_ctx context to an
MME service:
associate egtp-service egtp1 context dst_ctx
The following command associates a pre-configured HSS peer service called hss1 in the same context as
MME service to an MME service:
associate hss-peer-service hss1
bind s1-mme
Binds the MME service to a logical IP interface serving as the S1-MME interface.
Important Before modifying this bind configuration using the no bind s1-mme command, we recommend that the
MME Administrator use the clear mme-service db record command, under the Exec mode, to empty
the MME records database.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description bind s1-mme { ipv4-address address [ ipv4-address secondary_address ] | ipv6-address address [
ipv6-address secondary_address ] } [ crypto-template name ] [ max-subscribers number ]
no bind s1-mme
no
Removes a previously configured IP address used for binding the SCTP (local bind address) to communicate
with the eNodeBs using an S1-MME interface.
crypto-template name
Specifies an existing crypto template name used when implementing IP Security (IPSec) on the S1-MME
interface. name is an alphanumeric string of 1 through 104 characters.
max-subscribers number
Specifies the maximum number of subscribers that can access this service on this interface as an integer from
0 through 8000000.
For Release 15.0, the ASR 5500 platform supports up to 10,000,000 MME UE sessions.
Usage Guidelines Use this command to associate the MME service with a specific logical IP address that will be used for binding
the SCTP socket that communicates with the eNodeB using S1AP. Only one IP address can be configured
with this command for one MME service.
The MME passes the IP address during setting up the SCTP association with the eNodeB.
Caution This is a critical configuration. The MME service can not be started without this configuration. Any change
to this configuration will cause the MME service to be restarted. Removing or disabling this configuration
will stop the MME service.
Important Up to two IPv4 or IPv6 addresses can be configured to support SCTP multi-homing.
Examples The following command would bind the logical IP interface with the address of 192.168.3.1 to the MME
service to interact with eNodeB:
bind s1-mme ipv4-address 192.168.3.1
The following command disables a binding that was previously configured:
no bind s1-mme
csg-change-notification
This command enables or disables the Closed Subscriber Group (CSG) Information reporting (notification)
mechanism on the MME. When enabled, the MME includes the CSG Information Reporting Action IE with
the appropriate Action field for subscribers.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
By default, this feature is disabled. Using the default command prefix causes the MME to reset the configuration
for this parameter to the default so that the feature is disabled.
no
Disables the feature.
Usage Guidelines Use this command to enable or disable CSG change notification to the SGW/PGW.
By default csg-change-notification is disabled; the MME does not send CSG notification to the SGW/PGW.
dns
Specifes the context where the Domain Name System (DNS) client service is configured for DNS query to
select an MSC, P-GW, S-GW, peer SGSN or peer MME for this MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description dns { msc | peer-mme | peer-sgsn | pgw | sgw } [ context ctx_name ]
no dns { msc | peer-mme | peer-sgsn | pgw | sgw }
no
Removes a previously specified context having a DNS client service configured for DNS query to select a
MSC, peer MME, peer SGSN, P-GW or S-GW with this MME service.
msc
Specifies the context where a DNS client service is configured for DNS queries for selecting a Mobile Switching
Center (MSC) for SRVCC.
peer-mme
Specifies the context where a DNS client service is configured for DNS queries for selecting a peer MME.
peer-sgsn
Specifies the context where a DNS client service is configured for DNS queries for selecting a peer SGSN
for inter-RAT handovers.
pgw
Specifies the context where a DNS client service is configured for DNS queries for selecting a P-GW.
sgw
Specifies the context where a DNS client service is configured for DNS queries for selecting an S-GW.
context ctx_name
Optionally associates the specific context name where the DNS client service is configured for this MME
service. If this keyword is omitted, the DNS client service is configured to use the same context as this MME
service.
ctx_name is name of the configured context of the DNS client service expressed as an alphanumeric string of
1 through 79 characters that is case sensitive.
Usage Guidelines Use this command to specify a pre-configured context where a DNS client service is configured.
The DNS Client service configured in the specified context provides the DNS query support to locate MSCs,
peer MMEs, peer-SGSNs, P-GWs, or S-GWs for this MME service. For more information on DNS Client
service and support, refer to the DNS Client Service Configuration Mode Commands chapter.
A maximum of one context can be specified for each keyword.
Examples The following command associates a pre-configured context dns_ctx1 where a DNS client service is configured
for DNS query to MSC for this MME service:
dns msc context dns_ctx1
The following command associates a pre-configured context dns_ctx1 where a DNS client service is configured
for DNS query to P-GW for this MME service:
dns pgw context dns_ctx1
The following command associates a pre-configured context dns_ctx2 where a DNS client service is configured
for DNS query to S-GW:
dns sgw context dns_ctx2
emm
Defines the Evolved Mobility Management timer parameters, such as timeout durations for timers and
retransmission counts, for Non-Access Stratum (NAS) message retransmission in MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Resets the specified timer timeout to the system default value.
implicit-detach-timeout detach_dur
Sets the timer timeout duration (in seconds) after which subscriber will implicitly detached from the network
if there is no activity. Generally this timer value is 240 seconds (4 minutes) more than the timeout value of
the T3423 timer.
This timer starts when mobile reachable timer expires while the network is in EMM-IDLE mode and ISR is
activated and stops when a NAS signalling connection established.
detach_dur is an integer from 1 through 12000. Default: 3480
mobile-reachable-timeout mob_reach_dur
Sets the timeout timer duration (in seconds) after which reachability procedure will be discarded and reattempt
starts.
mob_reach_dur is an integer from 1 through 12000. Default: 3480
t3346-timeout t3346_dur
Sets the EMM backoff timer duration (in seconds). If an EMM request is rejected by MME because of
congestion, it shall have EMM cause as congestion (#22) and shall include back-off timer (T3346) IE. The
back-off timer shall be chosen randomly and shall be 10% below or above the configured T3346 timer value.
t3346_dur is an integer from 0 through 11160 (0-186 minutes). Default: 1500 seconds (25 minutes).
While storing this back-off timer expiry time, the MME shall adjust the mobile reachability timer and/or
implicit detach timer. This is to make sure that the sum of the mobile reachability timer + implicit detach
timer is greater than the back-off timer duration.
The MME will store the DB for at least the EMM back-off timer duration even if the attach is rejected because
of congestion. The MME will not start any timer for EMM back-off. Instead, back-off timer expiry time will
be stored in the DB as the DB is stored for at least back-off timer duration.
If an EMM call is rejected due to congestion control for EMM, the DB created during ULA will not be cleared
and the purge timer will be started for a time period 10% greater than the back-off timer duration. This is done
to make sure that DB is available during back-off timer duration to reject any requests during this period and
also to avoid the HSS signaling again if the UE comes back immediately after the back-off timer duration.
The MME will not reject any requests related to handovers as part of this feature even if EMM back-off timer
is running.
The MME will drop attach requests received during congestion while EMM back-off timer is running based
on configuration in congestion-action-profile. For example, if configuration is enabled to reject new call only
when low priority indication is set and the UE comes without low priority indication while back off timer is
running, the MME will accept the new call attempt from the UE.
The MME will not reject/drop attach requests received even if EMM back-off timer is running if the congestion
gets cleared.
The MME will forward SGS paging requests received from MSC for a UE attached in MME even if back-off
timer is running.
t3412-extended-timeout t3412_ext_dur
Sets the extended periodic TAU timer duration (in seconds), enabling the Operator to configure longer values
for the periodic TAU timer and Mobile Reachable timer. This helps the MME to reduce network load from
periodic TAU signaling and to increase the time until the UE detects a potential need for changing the RAT
or PLMN.
t3412_ext_dur is an integer from 0 through 1116000 (0-186 minutes). Default: 3600 seconds (60 minutes).
The UE must include the "MS network feature support" IE in the Attach Request/TAU Request. This IE
indicates to the MME that the UE supports the extended periodic timer T3412, in which case the MME sends
the extended-3412 IE in the attach/TAU response. The MME will not forward the extended-T3412 timer
value to any UE which has not indicated that it supports this extended-t3412 timer.
The MME supports storing the Subscribed-Periodic-RAU-TAU-Timer value if received as part of subscription
data, and deleting this stored value if the corresponding withdrawal flag is received in the DSR command.
For homers, the MME will send the extended-3412 IE value as received in
Subscribed-Periodic-RAU-TAU-Timer IE in subscription data.
For roamers, the MME takes the presence of Subscribed-Periodic-RAU-TAU-Timer IE in subscription data
as an indication and shall send the extended-3412 IE with the value from the local configuration.
The MME adjusts the configured mobile reachability timer value if the subscribed extended-3412 timer value
received from HSS is greater than the sum of the mobile reachability timer + implicit detach timer such that
the extended-3412 timer value becomes 10% less than the mobile reachability timer + implicit detach timer.
Refer to 3GPP TS 23.401 Section 4.3.17.3 (Version 10.4.0) & 29.272 for more details.
t3412-timeout t3412_dur
Sets the timeout duration (in seconds) for the T3412 timer. This timer is used for periodic tracking area update
(P-TAU). When this timer expires, the periodic tracking area updating procedure starts and the timer is set to
its initial value for the next start.
This timer starts when the UE goes from EMM-CONNECTED to EMM-IDLE mode and stops when the UE
enters EMM-CONNECTED mode.
t3412_dur is an integer from 1 through 11160. Default: 3240
t3413-timeout t3413_dur
Sets the timeout duration (in seconds) for the T3413 timer. The timer starts when MME initiates the EPS
paging procedure to the EMM entity in the network and requests the lower layer to start paging. This timer
stops for the paging procedure when a response received from the UE.
t3413_dur is an integer from 1 through 20. Default: 6
t3422-timeout t3422_dur
Sets the timeout duration (in seconds) for the T3422 timer. This timer starts when MME initiates the detach
procedure by sending a DETACH REQUEST message to the UE and stops upon receipt of the DETACH
ACCEPT message.
t3422_dur is tan integer from 1 through 20. Default: 6
t3423-timeout t3423_dur
Sets the timeout duration (in seconds) for the T3423 timer. This timer starts when UE enters the
EMM-DEREGISTERED state or when entering EMM-CONNECTED mode. It stops while the UE is in
EMM-REGISTERED.NO-CELL-AVAILABLE state and Idle mode Signalling Reduction (ISR) is activated.
t3423_dur is an integer from 1 through 11160. Default: 3240
t3450-timeout t3450_dur
Sets the timeout duration (in seconds) for the T3450 timer. This timer starts when MME initiates the Globally
Unique Temporary Identifier (GUTI) reallocation procedure by sending a GUTI REALLOCATION
COMMAND message to the UE and stops upon receipt of the GUTI REALLOCATION COMPLETE message.
This timer is also used for the Tracking Area update procedure.
t3450_dur is an integer from 1 through 20. Default: 6
t3460-timeout t3460_dur
Sets the timeout duration (in seconds) for the T3460 timer. This timer starts when the network initiates the
authentication procedure by sending an AUTHENTICATION REQUEST message to the UE and stops upon
receipt of the AUTHENTICATION RESPONSE message.
t3460_dur is an integer from 1 through 20. Default: 6
t3470-timeout t3470_dur
Sets the timeout duration (in seconds) for the T3470 timer. The MME starts this timer when the network
initiates the identification procedure by sending an IDENTITY REQUEST message to the UE and stops upon
receipt of the IDENTITY RESPONSE message.
t3470_dur is an integer from 1 through 20. Default: 6
T3411 Retransmission of
• EMM-DEREGISTERED. • At attach failure due to • ATTACH the ATTACH
ATTEMPTING-TO-ATTACH lower layer failure, REQUEST REQUEST or
T3410 timeout or attach sent
• EMM-REGISTERED. TRACKING
rejected with other EMM
ATTEMPTING-TO-UPDATE • TRACKING AREA UPDATE
cause values than those
AREA REQUEST
treated in subclause
UPDATE
5.5.1.2.5.
REQUEST
• At tracking area updating sent
failure due to lower layer
failure, T3430 timeout or
TAU rejected with other
EMM cause values than
those treated in subclause
5.5.3.2.5.
T3430 EMMT
-RACKN
IGA
-REAU
-PDATN
IGN
I-T
IA
ITED TRACKING AREA UPDATE Start T3411 or
REQUEST sent • TRACKING T3402 as
AREA described in
UPDATE subclause 5.5.3.2.6
ACCEPT
received
• TRACKING
AREA
UPDATE
REJECT
received
NOTE 2: The value of this timer is provided by the network operator during the attach and tracking area
updating procedures.
NOTE 3: The value of this timer may be provided by the network in the ATTACH ACCEPT message and
TRACKING AREA UPDATE ACCEPT message. The default value of this timer is identical to the value
of T3412.
NOTE 4: The value of this timer is provided by the network operator when a service request for CS fallback
is rejected by the network with EMM cause #39 "CS domain temporarily not available".
• GUTI
REALLOCATION
COMPLETE
received
Implicit All except The mobile reachable timer NAS signalling Implicitly detach
detach EMM-DEREGISTERED expires while the network is in connection the UE on 1st
timer EMM-IDLE mode and ISR is established expiry
activated
NOTE 1: Typically, the procedures are aborted on the fifth expiry of the relevant timer. Exceptions are
described in the corresponding procedure description.
Examples The following command sets the timeout value for EPS paging procedure timer T3413 for 10 seconds.
emm t3413-timeout 10
enb-cache-timeout
Configures the amount of time that eNodeB information is stored in cache after the eNodeB terminates the
connection.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default value of 10.
min
Specifies the amount of time (in minutes) that the MME stores eNodeB information after the eNodeB terminates
the connection. min is an integer value from 1 through 1440. Default: 10
Usage Guidelines Use this command to set the amount of time the MME stores eNodeB information in cache after the eNodeB
terminates the connection.
Examples The following command sets the amount of time the MME stores eNodeB information to 15 minutes:
enb-cache-timeout 15
encryption-algorithm-lte
Configures the precedence for LTE encryption algorithms to use for security procedures through this MME
service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description encryption-algorithm-lte priority1 { 128-eea0 | 128-eea1 | 128-eea2 } [ priority2 { 128-eea0 | 128-eea1 |
128-eea2 } ] [ priority3 { 128-eea0 | 128-eea1 | 128-eea2 } ]
default encryption-algorithm-lte
default
Sets the default LTE encryption algorithm for security procedures with configured priority value. Lowest
value has highest preference. Default configuration of LTE encryption algorithm is:
• priority1 with 128-eea0 encryption algorithm
• priority2 with 128-eea1 encryption algorithm
• priority3 with 128-eea2 encryption algorithm
priority1
Specifies the preference of integrity algorithm for security procedures on this MME service as priority 1.
priority2
Specifies the preference of integrity algorithm for security procedures on this MME service as priority 2.
priority3
Specifies the preference of integrity algorithm for security procedures on this MME service as priority 3.
128-eea0
Sets the Null ciphering algorithm (128-EEA0) for LTE encryption as the encryption algorithm for security
procedures. Default: Enabled
128-eea1
This keyword sets the SNOW 3G synchronous stream ciphering algorithm (128-EEA1) for LTE encryption
as the encryption algorithm for security procedures. SNOW 3G is a stream cipher that forms the base of the
3GPP confidentiality algorithm UEA2 and the 3GPP integrity algorithm UIA2. Default: priority2
128-eea2
Sets the Advance Encryption Standard (AES) ciphering algorithm (128-EEA2) for LTE encryption as the
encryption algorithm for security procedures. Default: priority3
Usage Guidelines Use this command to set the LTE encryption algorithms for security procedures to use with this MME service.
Caution When this command is executed, all the existing priority-to-algorithm mappings will be removed and the
newly configured ones will be applicable for security procedures.
Examples The following command sets the 128-EEA1 as the LTE encryption algorithm with priority 2 for security
procedures with an MME service:
encryption-algorithm-lte priority2 128-eea1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
esm
Defines the Evolved Session Management timer parameters like timeout durations for timers and retransmission
counts for the retransmission of Non-Access Stratum (NAS) messages in MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description esm { t3396-timeout t3396_dur | t3485-timeout t3485_dur | t3486-timeout t3486_dur | t3489-timeout
t3489_dur | t3495-timeout t3495_dur }
default esm { t3396-timeout | t3485-timeout | t3486-timeout | t3489-timeout | t3495-timeout }
default
Resets the specified Evolved Session Management timer timeout to the system default value.
t3396-timeout t3396_dur
Sets the ESM backoff timer duration (in seconds). If an ESM request is rejected because of congestion, the
reject will have ESM cause "Insufficient resources" and will include a back-off timer IE (T3396). This back-off
timer is chosen randomly and will be 10% below or above the configured T3396 timer value.
t3396_dur is an integer from 0 through 11160 (0-186 minutes). Default: 1500 seconds (25 minutes).
The MME will not start any timer for SM back-off, nor store the SM back-off timer expiry time. If an SM
request is received and if congestion exists, the request would be rejected based and a new random value will
be sent as the ESM back-off timer value.
The MME will reject any subsequent requests from the UE targeting to the same APN based on the presence
of congestion at that time and not based on the SM back-off time previously sent to the UE.
If the ESM cause value is #26 "insufficient resources" or #27 "missing or unknown APN", the MME will
include a value for timer T3396 in the reject message. If the ESM cause value is #26 "insufficient resources"
and the request message was sent by a UE accessing the network with access class 11 - 15 or if the request
type in the PDN CONNECTIVITY REQUEST message was set to "emergency", the MME will not include
a value for timer T3396.
t3485-timeout t3485_dur
Sets the timeout duration (in seconds) for the T3485 timer. This timer is used by the default EPS bearer context
activation procedure.
This timer starts when the MME sends an ACTIVATE DEFAULT EPS BEARER CONTEXT REQUEST
message to UE and stops when receives ACTIVATE DEFAULT EPS BEARER CONTEXT ACCEPT or
ACTIVATE DEFAULT EPS BEARER CONTEXT REJECT message from UE.
t3485_dur is an integer from 1 through 60. Default: 8
t3486-timeout t3486_dur
Sets the timeout duration (in seconds) for theT3486 timer. This timer is used by the default EPS bearer context
modification procedure.
This timer starts when the MME sends a MODIFY EPS BEARER CONTEXT REQUEST message to the
UE and stops when it receives a MODIFY EPS BEARER CONTEXT ACCEPT received or a MODIFY EPS
BEARER CONTEXT REJECT message from UE.
t3485_dur is an integer from 1 through 60. Default: 8
t3489-timeout t3489_dur
Sets the timeout duration (in seconds) for the T3489 timer. This timer is used for the default EPS bearer
context deactivation procedure.
This timer starts when the MME sends an ESM INFORMATION REQUEST message to the UE and stops
when receives a ESM INFORMATION RESPONSE message from the UE.
t3495_dur is an integer from 1 through 60. Default: 4
t3495-timeout t3495_dur
Sets the timeout duration (in seconds) for the T3495 timer. This timer is used for default EPS bearer context
deactivation procedure.
This timer starts when the MME sends a DEACTIVATE EPS BEARER CONTEXT REQUEST message to
UE and stops when receives DEACTIVATE EPS BEARER CONTEXT ACCEPT or DEACTIVATE EPS
BEARER CONTEXT REJECT message from UE.
t3495_dur is tan integer from 1 through 60. Default: 8
Usage Guidelines Use this command to set Evolved Session Management timers.
The following tables describe the triggers and states for timers:
NOTE 1: Typically, the procedures are aborted on the fifth expiry of the relevant timer. Exceptions are
described in the corresponding procedure description.
Examples The following command sets the timeout value for the default EPS bearer context activation procedure timer
(T3485) for 10 seconds.
esm t3485-timeout 10
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
gtpv2
Configures GTPv2 piggybacking support from the MME to the P-GW. A piggybacking flag is sent by the
MME to a P-GW in the S11 "Create Session Request" message and determines whether dedicated bearer
creation (Create Bearer Request) is piggybacked onto the "Create Session Response" message or not.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of enabled.
no
Disables the feature.
piggybacking
Specifies that piggybacking is to be performed by the P-GW.
Usage Guidelines Use this command to enable the sending of a piggybacking flag to the P-GW over the S11 interface requesting
that the Create Bearer Request message is piggybacked on the Create Session Response message (sent from
the P-GW to the MME.
heuristic-paging
Enables or disables the heuristic or optimized paging feature for the service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of disabled.
no
Disables the feature.
paging-map paging_map_name
Specifies the paging-map to be associated with this MME service. This keyword is only supported in Release
14.0 and higher.
Usage Guidelines
Caution The paging profiles need to be configured prior to configuring TAI management objects (tai-mgmt-db
and tai-mgmt-obj). Otherwise, the configuration would lead to high paging load in the MME node, at peak
traffic time, causing service outage
Use this command to enable or disable the heuristic paging feature for the service. Also known as idle-mode
paging, enabling this feature prompts the MME service to keep track of the eNodeBs to which the access
terminal (AT) most commonly attaches, thus reducing the signalling otherwise associated with continuous
paging.
If no paging-map is associated when this command is issued, the default heuristic paging behavior is used (as
opposed to intelligent paging behavior).
Refer to the Heuristic and Intelligent Paging chapter in the MME Administration Guide for more information
about this command.
Important Heuristic (optimized) Paging is a licensed feature and will not appear as a command option unless the
proper licensed is installed.
ho-resource-release-timeout
Configures the timer that is started when the source MME initiates a handover.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to the default setting of 5000 milliseconds.
timeout
Specifies the time in milliseconds that the MME will hold on to bearers and E-RABs after an S1-based
handover has been initiated.
timeout must be an integer from 500 through 15000.
Default: 5000.
Usage Guidelines Use this command to configure the amount of time in milliseconds that the MME will hold on to bearers and
E-RABs after an S1-based handover has been initiated. When this timer expires, the source MME will send
a UE Context Release to the source eNodeB. Refer to 3GPP TS 23.401 Section 5.5.1.2.2 for additional
information about the use of this timer.
Examples The following command configures the timer for 10000 milliseconds (10 seconds).
ho-resource-release-timeout 10000
integrity-algorithm-lte
Configures the precedence of LTE integrity algorithms to use for security procedures through this MME
service. By default the integrity algorithm is enabled on MME service and cannot be disabled.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Removes the preconfigured integrity algorithm and sets the default LTE integrity algorithm for security
procedures. Default configuration of LTE integrity algorithm is:
• priority1 with 128-eia1 integrity algorithm
• priority2 with 128-eia2 integrity algorithm
priority1
Specifies the preference of integrity algorithm for security procedures on this MME service as priority 1.
This is the mandatory and default priority keyword.
priority2
Specifies the preference of integrity algorithm for security procedures on this MME service as priority 2.
128-eia1
This keyword sets the SNOW 3G synchronous stream ciphering algorithm (128-EIA1) for LTE integrity as
the integrity algorithm for security procedures. SNOW 3G is a stream cipher that forms the base of the 3GPP
confidentiality algorithm UEA2 and the 3GPP integrity algorithm UIA2. Default: priority1
128-eia2
Sets the Advance Encryption Standard (AES) ciphering algorithm (128-EIA2) for LTE integrity as the integrity
algorithm for security procedures. Default: Enabled
Usage Guidelines Use this command to set the LTE integrity algorithms for security procedures to use with this MME service.
Caution Integrity algorithm is a mandatory aspect and can not be disabled in MME service.
Caution When this command is executed, all the existing priority to algorithm mappings will be removed and the
newly configured ones will be applicable for security procedures.
Examples The following command sets the AES ciphering algorithms (128-EIA2) as the LTE integrity algorithm with
priority as 1 for security procedures with an MME service:
integrity-algorithm-lte priority1 128-eia2
inter-rat-nnsf
Configures an NNSF (NAS Node Selection Functionality) entry to define a list of Served MMECs (MME
codes) that is indicated to the eNodeB in the S1 Setup Response. This optional configuration is used to aid
the eNodeB when selecting the MME for inter-rat handovers when the MME is co-located with an SGSN.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description inter-rat-nnsf collocated-mme plmn-id mcc mcc_value mnc mnc_value group-id mme_group_id {
mme-codes mmec | mme-code-range first_mme_code to last_mme_code }
no inter-rat-nnsf collocated-mme plmn-id mcc mcc_value mnc mnc_value group-id mme_group_id
no
Removes the specified NNSF entry.
collocated-mme
Specifies that the MME is co-located with an SGSN.
group-id mme_group_id
Configures the group id for this MME service.
mme_group_id must be an integer value from 0 through 65536.
mme-codes mmec
Configures a list of MMEC (MME codes) to be used.
mmec: must be entered as a series of codes, each separated by a space, such as: 10 25 102 103 105. Each code
must be an integer from 0 through 255.
A maximum of 16 MME Codes are allowed to be configured per inter-rat-nnsf entry.
Usage Guidelines Use this command to indicate a list of served MMECs, in addition to the one assigned to the MME service.
The complete list shall be notified to the eNodeB as Served MMECs in the S1 Setup Response. This would
aid the eNodeB in selecting a co-located MME during 2G/3G to 4G handovers.
When a UE moves from 2G/3G to 4G, selecting a co-located MME is not possible without some explicit
configuration. In this scenario, the entire second Most-Significant-Byte of P-TMSI is copied into the MME-Code
(MMEC) field. Depending on the NRI length, this could result in 'n' different MMEC values for the same
NRI value. For example:
• NRI length = 6 bits
• NRI value = 5 (Binary 00 0101)
• Possible MMECs: Binary 00 0101 xx -> {20, 21, 22, 23}
Selecting a co-located MME is only possible if the eNodeB knows that any UE meant for the above set of
MMECs should be directed to a given MME. This command enables the operator to specify MMECs that can
possibly be mapped from a given NRI value.
A maximum of 16 MME Codes are allowed to be configured per inter-rat-nnsf entry. This allows 4 SGSNs
with NRI length of 6, or 2 SGSNs with NRI length of 5. If more than 16 MMECs are required, an alternative
is to pick a dummy MME-Group-ID value and create a new nnsf-entry. The Serving MME-Group-ID could
also be used for this purpose as MME-Group-Id has no significance during MME node selection.
A Maximum of 32 inter-rat-nnsf entries are allowed. Regardless of the maximum entries configured, the
maximum limits placed by S1AP stack take precedence. For example, if the number of plmns configured
under 'network-sharing' and 'inter-rat-nnsf' exceeds the maxnoofPLMNsPerMME(32) limit set by
S1AP-S1-Setup-Response, then inter-rat-nnsf entries that exceed the limit(32) do not get included in the S1
Setup Response message.
Examples For NRI length = 6; NRI Value = 10 (Binary: 00 1010), when a UE moves from 2G/3G to 4G and maps MME
Code (8 bits) from P-TMSI, the MME Code value could be:
• Binary: 00 1010 xx, where xx can be binary 10 or 01 or 00 or 11
• Decimal: 40 or 41 or 42 or 43
So, all of the above values should be configured as MMECs as part of inter-rat-nnsf, as follows:
inter-rat-nnsf collocated-mme plmn-id mcc 121 mnc 102 mme-id group-id 32000 mme-codes 40 41 42
43
When updating an existing NNSF entry, any new MMECs must be included with the existing MMECs. For
example, to add additional MMECs (48 49 50 51) to the above command, enter the entire command again as
follows:
inter-rat-nnsf collocated-mme plmn-id mcc 121 mnc 102 mme-id group-id 32000 mme-codes 40 41 42
43 48 49 50 51
isda-guard-timeout
Sets the number of seconds for the Insert Subscription Data Answer (ISDA) guard timer. The time the MME
waits for current location information for the UE. If the current location is not learned before expiry, because
there is no paging response or location reporting control from the eNB, then the MME sends the ISDA with
the last-known location upon expiry of this timer.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Disables any configuration for this timer and resets the wait time to the default of 25 seconds.
seconds
Enter an integer from 1 to 100.
Usage Guidelines With this command, the operator can configure the ISDA guard timer to any value from 1 to 100 seconds.
Upon expiry of this wait timer, the MME sends the ISDA with the last-known location of the UE if the MME
receives the Insert Subscriber Data Response (ISDR) with both the location flags set (current and last-known
locations). Only when the ISDR is received, with both flags set, is the ISDA guard timer started. In situations
where the MME receives the ISDR with only the last-known location flag set, then the MME immediately
sends the ISDA with location information - no delay and this timer is not started even if configured.
When the ISDA guard timer expires, the paging procedure does not stop until the page timer expires but the
MME ignores the paging timer and sends the ISDA with the last-known location if the ISDR was received
with both location flags set and the UE is in EMM-idle mode.
While the MME is serving the ISDR (where both location flags are set) from the HSS, if the HSS tries to send
another similar request then the MME responds to the HSS with DIAMETER_UNABLE_TO_COMPLY.
This timer is separate from the paging timer and configuration of the ISDA guard timer can reduce the overall
delay before sending the ISDA.
Examples Instruct the MME to wait 10 seconds before sending the ISDA with the last-known location of the UE:
isda-guard-timeout 10
isr-capability
Enables or disables the Idle-mode Signaling Reduction (ISR) feature on the MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the ISR feature to the default setting (disabled) on MME service.
no
Disables the ISR feature on MME service.
Usage Guidelines Use this command to enable or disable the ISR feature on the MME service. When enabled, the MME can
perform ISR functions with a peer SGSN which also supports ISR.
Refer to the Idle-mode Signaling Reduction chapter in the MME Administration Guide for more information
about this command
Important This functionality is a license-controlled feature. A valid feature license must be installed to enable
Idle-mode Signaling Reduction.
legacy-tai-list-encoding
Using this command instructs the MME to override the default behavior (described in Usage section below)
and enables the MME to use "010" encoding value for the Tracking Area Identity (TAI) list IE for TAIs
belonging to different PLMNs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Disables the use of "010" encoding value for the TAI list IE for TAIs belonging to different PLMNs and
returns the MME to using the TAI list value encoding based on PLMN and TAC values of TAI entries, the
default behavior.
Usage Guidelines The operator can use this command to configure the encoding of TAI list values to "010" irrespective of
PLMN and TAI values, which overrides the default behavior (for releases 17.4 and forward). This commnd
ensures backward compatibility with previous releases.
If this command is not used, or the no command prefix is used, then the MME uses the default function and
encodes the TAI list IE value per the 3GPP TS 24.301. The default behavior has the MME automatically
encode "000", "001", or "010" depending upon the TAC values and PLMN configuration so that the TAI list
value for the IE is based on the list of Tics belonging to one PLMN, with consecutive or non-consecutive
TAC values configured in the TAI entries.
Examples Use the following command to override the MME's default behavior and to encode TAI list values to "010":
legacy-tai-list-encoding
local-cause-code-mapping apn-mismatch
Configures the reject cause code to send to a UE when an APN mismatch occurs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when an APN mismatch occurs, such as when
an APN is present in the HSS subscription but the HSS subscription for this IMSI has other APNs present in
the subscription. By default, the MME sends the UE the #23 - ESM Failure cause code for this condition.
If a condition is specified in both the call-control-profile associated with a call, and also the mme-service, the
cause configured for the call-control-profile will be signalled to the UE. See also the local-cause-code-mapping
command in the call-control-profile configuration mode. This command is described in the Call Control
Profile Configuration Mode Commands chapter.
Examples The following command maps the "PLMN not allowed" cause code to the APN mismatch condition:
local-cause-code-mapping apn-mismatch emm-cause-code plmn-not-allowed
local-cause-code-mapping apn-not-subscribed
Gives the operator the option to specify the local cause-code mapping when the UE-requested APN is not
subscribed.
Product MME
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the local cause code mapping to the default of #27 (Unknown or Missing APN).
Usage Guidelines The operator can specify "Requested-Option-Not-Subscribed" cause code value #33 will be sent in the Reject
message when the PDN Connectivity Request is rejected because no subscription is found. If the command
option is not configured, then by default the MME uses the cause code value #27 (Unknown or Missing APN)
in standalone PDN Connectivity Reject message when the UE-requested APN is not subscribed.
Examples The following instructs the MME to use cause code #33 ("Requested-Option-Not-Subscribed") in place of
the default #27 (Unknown or Missing APN):.
local-cause-code-mapping apn-not-subscribed esm-cause-code requested-service-option-not-subscribed
local-cause-code-mapping apn-not-supported-in-plmn-rat
This command maps the operator-preferred ESM/EMM cause code to be sent in Activation Reject messages
in place of the standard 3GPP Release 11 rejection cause #66 when activation of the requested APN is not
supported in current RAT and PLMN combination.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the cause code mapping to its default values. The default cause code values for Attach procedures
are emm-cause-code 19 and esm-cause-code 66. The default cause code values for TAU procedures are
emm-cause-code 15 and esm-cause-code 66 respectively. The default cause code for ESM procedure is 66.
apn-not-supported-in-plmn-rat
The keyword apn-not-supported-in-plmn-rat specifies that the cause codes to be used for a rejection due
to the requested APN not being supported in the current RAT and PLMN combination are those that are
mapped in the configuration.
• esm-cause-code configures the operator-preferred ESM cause code to be used if a NAS Request is
rejected due to this configuration.
• esm_cause_number specifies the ESM code replacement integer. The system accepts a value in the
range 0 through 255, however, the standards-compliant valid values are in the range 8 through 112.
• The attach keyword filter instructs the MME to use the mapped replacement cause code if an Attach
procedure is rejected due to the noted APN not supported error condition.
• The tau keyword filter instructs the MME to use the mapped replacement cause code if an TAU procedure
is rejected due to the noted APN not supported error condition.
Usage Guidelines This command is used to remap the ESM and EMM cause codes sent in activate rejections (due to APN not
supported) to operator desired ESM or EMM cause codes. The default cause code values for Attach procedures
are emm-cause-code 19 and esm-cause-code 66. The default cause code values for TAU procedures are
emm-cause-code 15 and esm-cause-code 66. The default cause code for esm-proc is 66.
Examples The following command is used to remap cause code #66 to cause code #20, this cause code will be sent if a
bearer management request is rejected.
local-cause-code-mapping apn-not-supported-in-plmn-rat esm-cause-code 20 esm-proc
local-cause-code-mapping auth-failure
Configures the reject cause code to send to a UE when an authentication failure occurs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when an authentication failure occurs. By
default, the MME sends the UE the #3 - Illegal MS cause code when encountering a context transfer failure
from an MME.
This condition occurs for TAU and ATTACH procedures in the following cases:
• The Authentication response from the UE does not match the expected value in the MME.
• Security Mode Reject is send by the UE.
• The UE responds to any identity request with a different type of identity (ie, the MME could query for
IMSI and the UE responds with IMEI).
The following are not considered for the authentication failure condition:
• HSS returning a result code other than SUCCESS.
• HSS not available.
• EIR failures.
• UE not responding to requests.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "network-failure" cause code to the authentication failure condition:
local-cause-code-mapping auth-failure emm-cause-code network-failure
local-cause-code-mapping congestion
Configures the reject cause code to send to a UE when a procedure fails due to a congestion condition.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a UE procedure fails due to a congestion
condition on the MME. By default, the MME sends the UE the #22 - CongestionEMM cause code and ESM
cause code when encountering congestion.
If a condition is specified in both the call-control-profile associated with a call, and also the mme-service, the
cause configured for the call-control-profile will be signalled to the UE. See also the local-cause-code-mapping
command in the call-control-profile configuration mode. This command is described in the Call Control
Profile Configuration Mode Commands chapter.
Examples The following command maps the "network failure" cause code to the congestion event:
local-cause-code-mapping congestion emm-cause-code network-failure
local-cause-code-mapping ctxt-xfer-fail-mme
Configures the reject cause code to send to a UE when a UE context transfer failure from a peer MME occurs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a UE context transfer failure from a
peer MME occurs. By default, the MME sends the UE the #9 - MS identity cannot be derived by the network
cause code for this condition.
After the peer node has been identified, the MME sends a Context Request to the peer node. If the peer node
is an MME, and if the context transfer procedure fails, this condition is detected.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "network-failure" cause code to the context transfer failure from MME
condition:
local-cause-code-mapping ctxt-xfer-fail-mme emm-cause-code network-failure
local-cause-code-mapping ctxt-xfer-fail-sgsn
Configures the reject cause code to send to a UE when a UE context transfer failure from a peer SGSN occurs.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a UE context transfer failure from a
peer SGSN occurs. By default, the MME sends the UE the #9 - MS identity cannot be derived by the
network cause code when encountering this condition.
After the peer node has been identified, the MME sends a Context Request to the peer node. If the peer node
is an SGSN, and if the context transfer procedure fails, this condition is detected.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "network-failure" cause code to the context transfer failure from SGSN
condition:
local-cause-code-mapping ctxt-xfer-fail-sgsn emm-cause-code network-failure
local-cause-code-mapping gw-unreachable
Configures the reject cause code to send to a UE when a gateway (S-GW or P-GW) does not respond during
an EMM procedure.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a gateway does not respond. By default,
the MME sends the UE the #19 - ESM Failure cause code when encountering this condition.
Defaults:
Prior to StarOS 15.0 MR5, the MME sends the UE the #19 - ESM Failure cause code when encountering
this condition.
In StarOS 15.0 MR5 and higher releases, the MME sends the UE the #19 - ESM Failure cause code for
Attach procedures, and #40 - NO-EPS-BEARER-CONTEXT-ACTIVATED for TAU procedures.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "network-failure" cause code to the gateway unreachable condition:
local-cause-code-mapping gw-unreachable emm-cause-code network-failure
local-cause-code-mapping hss-unavailable
Configures the reject cause code to send to a UE when the HSS does not respond.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when the HSS does not respond. By default,
the MME sends the UE the #17 - Network failure cause code when encountering this condition.
The cause code configured for this condition will be signaled in TAU and ATTACH REJECT messages.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "tracking-area-not-allowed" cause code to the HSS unavailable condition:
local-cause-code-mapping hss-unavailable emm-cause-code tracking-area-not-allowed
local-cause-code-mapping newcall-policy-restrict
Configures the EPS Mobility Management (EMM) reject cause code to send to a UE when a UE requests
access but the call control profile has set the call disposition to reject.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Usage Guidelines Use this command to configure the cause code returned to a UE when a UE procedure fails, such as when the
UE requests access to a restricted zone. By default, the MME sends the UE the #22 - Congestion cause code
when encountering this condtion.
Examples The following command sets the "network-failure" cause code for newcall-policy-restrict calls:
local-cause-code-mapping newcall-policy-restrict emm-cause-code network-failure
local-cause-code-mapping no-active-bearers
Configures the reject cause code to send to a UE when the context received from a peer SGSN (during a TAU
procedure) does not contain any active PDP contexts.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when the context received from a peer SGSN
(during a TAU procedure) does not contain any active PDP contexts. By default, the MME sends the UE the
#40 - No PDP context activated cause code when encountering this condition.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "plmn-not-allowed" cause code to the no active bearer condition:
local-cause-code-mapping no-active-bearers emm-cause-code plmn-not-allowed
local-cause-code-mapping peer-node-unknown
Configures the reject cause code to send to a UE when peer node resolution is not successful.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when peer node resolution is not successful.
By default, the MME sends the UE the #9 - MS identity cannot be derived by the network cause code when
encountering this condition.
During processing of a TAU Request, the resolution of a peer MME that had allocated the temporary identity
that is signaled to the UE takes several steps in the MME. This resolution can be done based on DNS or based
on local configuration. This condition occurs when all mechanisms for peer node resolution are done with no
success.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "plmn-not-allowed" cause code to the peer node unknown condition:
local-cause-code-mapping peer-node-unknown emm-cause-code plmn-not-allowed
local-cause-code-mapping pgw-selection-failure
Configures the reject cause code to send to a UE when a failure occurs during P-GW selection.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a failure occurs during P-GW selection.
By default, the MME sends the UE the #17 - Network failure cause code when encountering this condition.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "plmn-not-allowed" cause code to the P-GW selection failure condition:
local-cause-code-mapping pgw-selection-failure emm-cause-code plmn-not-allowed
local-cause-code-mapping restricted-zone-code
Configures the reject cause code to send to a UE when a procedure fails.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a UE procedure fails, such as when the
UE requests access to a restricted zone. By default, the MME sends the UE the #15 - No Suitable Cells in
Tracking Area cause code when encountering this condition.
If a condition is specified in both the call-control-profile associated with a call, and also the mme-service, the
cause configured for the call-control-profile will be signalled to the UE. See also the local-cause-code-mapping
command in the call-control-profile configuration mode. This command is described in the Call Control
Profile Configuration Mode Commands chapter.
Examples The following command maps the "PLMN not allowed" cause code to the restricted zone code event:
local-cause-code-mapping restricted-zone-code emm-cause-code plmn-not-allowed
local-cause-code-mapping sgw-selection-failure
Configures the reject cause code to send to a UE when a failure occurs during S-GW selection.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a failure occurs during S-GW selection.
By default, the MME sends the UE the #17 - Network failure cause code when encountering this condition.
If a cause code mapping is specified in both the call-control-profile associated with a call, and also the
mme-service, the cause configured for the call-control-profile will be signalled to the UE. See also the
local-cause-code-mapping command in the call-control-profile configuration mode. This command is
described in the Call Control Profile Configuration Mode Commands chapter.
Examples The following command maps the "plmn-not-allowed" cause code to the S-GW selection failure condition:
local-cause-code-mapping sgw-selection-failure emm-cause-code plmn-not-allowed
local-cause-code-mapping vlr-down
Configures the cause code to send in a ATTACH ACCEPT or TAU ACCEPT to a UE that attachment to the
VLR has failed because a VLR down condition is present.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a VLR down condition is present. By
default, the MME sends the UE the #16: "MSC temporarily not reachable cause code when encountering
this condition.
If a condition is specified in both the call-control-profile associated with a call, and also the mme-service, the
cause configured for the call-control-profile will be signalled to the UE. See also the local-cause-code-mapping
command in the call-control-profile configuration mode. This command is described in the Call Control
Profile Configuration Mode Commands chapter.
Examples The following command maps the "network failure" EMM cause code to the VLR down condition:
local-cause-code-mapping vlr-down emm-cause-code network-failure
local-cause-code-mapping vlr-unreachable
Configures the cause code to send in a ATTACH ACCEPT or TAU ACCEPT to a UE that attachment to the
VLR has failed because a VLR unreachable condition is present.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Use this command to configure the cause code returned to a UE when a VLR unreachable condition is present.
By default, the MME sends the UE the #16: "MSC temporarily not reachable cause code when encountering
this condition.
If a condition is specified in both the call-control-profile associated with a call, and also the mme-service, the
cause configured for the call-control-profile will be signalled to the UE. See also the local-cause-code-mapping
command in the call-control-profile configuration mode. This command is described in the Call Control
Profile Configuration Mode Commands chapter.
Examples The following command maps the "network failure" EMM cause code to the VLR unreachable condition:
local-cause-code-mapping vlr-unreachable emm-cause-code network-failure
location-reporting
Enables or disables the UE location reporting function on the MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Disables the location reporting feature on MME service.
no
Disables the location reporting feature on MME service.
Usage Guidelines Use this command to enable or disable the UE location reporting feature on the MME service. When enabled
the MME forwards a location report request for a specific UE from the P-GW to the eNodeB.
Important Location reporting, also known as User Location Information (ULI) Reporting, is a licensed feature and
requires the purchase of the ULI Reporting feature license.
Examples The following command sets the MME service to allow for location reporting for UEs:
location-reporting
mapping
Configures how the MME maps the Target RNC-ID fields to the Target eNodeB-ID and TAC fields for
Inter-RAT Gn/Gp handovers.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Sets the command to use the default value of maptype-default-includes-only-enb.
maptype-default-includes-only-enb
Default mapping logic
Maps the Target RNC-ID fields to Target eNodeB-ID fields as follows:
• PLMN of LAI => PLMN of MME
• LAC of LAI => MME Group ID
• RAC => Not used.
• RNC-ID (12 or 16bits) => Lowest 12 or 16 bits of eNB ID.
• TAC is picked from the list of TAIs supported by the target eNB.
maptype1-includes-enb-tai
Maps the Target RNC-ID fields to Target eNodeB-ID fields as follows:
• PLMN of LAI => PLMN of TAI and eNB
• LAC of LAI => TAC of TAI
Usage Guidelines Use this command to configure how the MME maps the Target RNC-ID fields to the Target eNodeB-ID and
TAC fields for Inter-RAT Gn/Gp handovers.
max-bearers per-subscriber
Specifies the maximum number of EPS bearers that a subscriber may simultaneously use to access this MME
service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Configures the maximum EPS bearers for a subscriber to use simultaneously to the default value of 11.
max_bearer
Specifies the maximum number of EPS bearers for a subscriber may simultaneously use to access this MME
service.
max_bearer is an integer from 1 through 11. Default: 11
Usage Guidelines Use this command to set the maximum number of EPS bearers that a subscriber may simultaneously use to
access this MME service.
Examples The following command specifies that a maximum of 6 simultaneous EPS bearers can be facilitated for a
subscriber at any given time:
max-bearers per-subscriber 6
max-paging-attempts
This command configures the maximum number of paging attempts allowed for network requested service
creation to a subscriber.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Configures the maximum number of paging attempts to the default value of 3.
max_paging_attempts
Specifies the maximum number of paging attempts allowed for network requested service creation to a
subscriber.
max_paging_attempts is an integer from 1 through 10. Default: 3
Usage Guidelines Use this command to set the maximum number of paging attempts allowed for network requested service
creation to a subscriber.
When Heuristic Paging is enabled, this setting applies only to messages sent to all eNodeBs in all TAIs present
in the TAI list. Paging to the last known eNodeB and paging the TAI from which the UE was last heard is
attempted only once. As a result, when max-paging-attempts is set to 3, a maximum of 5 paging retries are
attempted with Heuristic Paging enabled.
Refer to the Heuristic and Intelligent Paging chapter in the MME Administration Guide for more information
about Heuristic Paging.
Examples The following command specifies that a maximum of 6 paging attempt retransmissions allowed for network
requested service creation to a subscriber:
max-paging-attempts 6
max-pdns per-subscriber
Specifies the maximum number of Packet Data Networks (PDNs) that a subscriber may simultaneously access
through this MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Configures the maximum PDNs that a subscriber can simultaneously access through this MME service to the
default value of 3.
max_pdn
Specifies the maximum number of PDNs that a subscriber may simultaneously access through this MME
service.
max_pdn is an integer from 1 through 11. Default: 3
Usage Guidelines Use this command to set the maximum number of PDNs that a subscriber may simultaneously access through
this MME service.
Examples The following command specifies that a maximum of 2 simultaneous PDNs can be accessed by a subscriber
at any given time through this MME service:
max-pdns per-subscriber 2
mme-id
Configures the MME identifier within an MME service. The MME identifier is constructed form the MME
group ID and MME Code.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Removes the configured MME identifier for this MME service.
Caution Removing the MME identifier is a disruptive operation; the MME service is removed from the system.
group-id grp_id
Specifies the group identifier for the group of which this MME belongs as an integer from 0 through 65535.
mme-code mme_code
Specifies the unique code for this MME service as an integer from 0 through 255.
Usage Guidelines Use this command to set the MME identifier for this MME service. This MME identifier will be the identity
of this MME in network.
Caution Changing or removing the MME identifier is a disruptive operation; the MME service will be re-started
or removed from service.
Examples The following command configures the MME identifier with group id as 41025 and MME code as 101 for
this MME service:
mme-id group-id 41025 mme-code 101
mmemgr-recovery
Configures the recovery action for the MME manager.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Resets the function configuration to the MME's default value of reset S1 peers.
no-reset
Specifies that the recovery action is not to reset S1 peers.
reset-s1-peers
Specifies that the recovery action is to reset S1 peers. This is the default action.
Usage Guidelines Use this command to set a recovery action for the MME Manager.
Examples The following command configures the MME Manager recovery action to reset all S1 peers:
mmemgr-recovery reset-s1-peers
msc
Creates and manages an Mobile Switching Center (MSC) server configuration, for the MME service, for an
MSC enhanced with Single Radio Voice Call Continuity (SRVCC). The MSC server acts as an endpoint for
the Sv interface.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description msc { msc_name | [ ipv4_address | ipv6_address ] } [ ip-address [ ipv4_address | ipv6_address ] [ offline |
online ]
no msc { msc_name | [ ipv4_address | ipv6_address ] }
no msc_name
Removes the MSC configuration from the MME service.
msc_name
Specifies a name for this peer MSC server.
msc_name must be an alphanumeric string from 1 to 63 characters.
offline
Mark this MSC server offline for maintenance. Once this command is issued, the MME will no longer send
future handover requests to this MSC server. No GTPv2 messages are generated when offline/online mode
is changed.
Once the MSC server is set for offline, the online keyword must be used to return the server to online mode.
online
Mark this MSC server for online mode. Once this command is issued, the MSC server is added back into the
MSC selection algorithm and normal operation is returned. By default, an MSC server is online unless the
offline keyword is specified.
Usage Guidelines Use this command to configure a peer MSC server used during SRVCC handovers. For details on the
configuration of the MSC and the MME's usage of SRVCC, refer to the Single Radio Voice Call Continuity
feature chapter in the MME Administration Guide.
Also, this command can set an MSC server offline for maintenance.
Examples For Release 16.0 and higher, the following command defines an MSC server msc1 that will be selected by
DNS. Any MSCs configured for DNS-based selection must be defined without an IP address:
msc msc1
The following command defines a default MSC server with an IPv4 address of 10.2.3.20. The MME will
select the default when no other MSC selection logic (DNS selection or MSC pool areas) are configured, or
when these fail to return an MSC address.
msc default ip-address 10.2.3.20
For Release 15.0 and higher:
The following command defines an MSC server mscwest with an IPv4 address of 10.2.3.4:
msc mscwest ip-address 10.2.3.4
The following command marks the above MSC server offline:
msc mscwest ip-address 10.2.3.20 offline
The following command defines a default MSC server with an IPv4 address of 10.2.3.20. The MME will
select the default when MSC pool areas are not configured, or when an MSC address fails to be returned.
msc default ip-address 10.2.3.20
For Release 14.0 and earlier:
The following command specifies an IPv4 address for the peer MSC server as 10.2.3.4:
msc 10.2.3.4
msc-mapping
This command creates a mapping between the MSC ISDN number and the MSC's IP-address (either IPv4 or
IPv6) to ensure location continuity for SRVCC handover. This mapping is required to include the MSV ID
in the target service node IE for the Emergency_Call_Handover event.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Removes a specific MSC IP address mapping definition from the MME Service configuration.
ip-address
ipv4_address | ipv6_address Specifies the IP address for the MSC as an IPv4 dotted-decimal or as an IPv6
colon-separated-hexadecimal notation.
isdn
isdn_number: Enter a numeric string upto 15 digits long.
Usage Guidelines The MME Service supports a maximum of 24 MSC IP address to ISDN mapping definitions.
Use the show mme-service command to see the MSC IP address to ISDN mappings created with this command.
Examples Map the IPv4 192.168.61.2 address of the MSC to ISDN 123456789012345
msc-mapping ip-address 192.168.61.2 isdn 123456789012345
nas gmm-qos-ie-mapping
Configures which QOS the MME uses in NAS GMM QoS IE and GTPv1 Context response message when
the subscriber comes to MME via a handover from a GN/GP SGSN.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
gngp-imported-qos
Configures the MME to send the QoS received from GN/GP SGSN (whenever applicable).
native-eps-qos
Configures the MME to send the EPS (4G) QoS received from HSS.
This is the default setting.
Usage Guidelines When a subscriber comes to MME via a handover from Gn/Gp SGSN, this command controls whether the
MME is to use the QoS received from the SGSN, or whether to use the updated EPS QoS received from the
HSS. This value is then mapped to gmm-qos-ie in subsequent NAS messages and in GTPv1 Context response
messages.
Examples The following command configures the MME to use the QoS values from the Gn/Gp SGSN in gmm-qos-ie
NAS messages and GTPv1 Context response messages.
nas gmm-qos-ie-mapping gngp-imported-qos
nas-max-retransmission
Sets the retransmission counter for all type of Non-Access Stratum (NAS) messages in an MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Resets the retransmission counter to the default value of 4.
nas_retrans_count
Sets the maximum number of retransmission of NAS messages permitted during any procedure after which
the activation procedure will be discarded.
nas_retrans_count is an integer from 1 through 10. Default: 4
Usage Guidelines Use this command to set maximum number of retries allowed for any type of NAS messages.
NAS Messages sent by the MME which require a response from the UE for procedure completion are
retransmitted. Retransmission happens based on timer expiry. The timers are configured through the emm
and esm commands. NAS messages are retransmitted per configuration, and if no response is received from
the UE, the pending transaction is abandoned. If the transaction is a DETACH or PDN DISCONNECT
REQUEST, the transaction is completed without further UE signaling.
The timeout duration configured through the emm and esm commands will be applicable between two retries.
Examples The following command sets the maximum number of retries allowed as 4 for all type of NAS messages in
an MME service.
default nas-max-retransmissions
network-sharing
Configures additional PLMN IDs for this MME service. Refer to the plmn-id command to create the base
PLMN identifier for an MME service. Each PLMN ID consists of the Mobile Country Code (MCC) and
Mobile Network Code (MNC). A maximum of four network sharing entries can be configured per MME
service. These PLMN IDs will be communicated to the eNodeBs in the S1 SETUP response and MME CFG
Update messages.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description network-sharing plmn-id mcc number mnc number mme-id group-id id mme-code code
no network-sharing plmn-id mcc number mnc number
no
Disables the network sharing mode on this MME service.
Caution Removing the PLMN identifier is a disruptive operation; the MME service will be restarted.
mme-id group-id id
Specifies the group identifier for the group to which this MME belongs as an integer from 0 through 65535.
mme-code code
Specifies the unique code for an MME service as an integer from 0 through 255.
Usage Guidelines Use this command to configure additional PLMN IDs for an MME service. In a given MME service, each
PLMN ID (MCC and MNC) must be unique.
Caution Changing or removing the PLMN identifier is a disruptive operation; the MME service will be restarted.
Examples The following command configures the network sharing parameters to an MCC of 123, an MNC of 12, a
MME-ID/Group ID of 100, and a MME code of 50:
network-sharing plmnid mcc 123 mnc 12 mme-id group-id 100 mme-code 50
nri
Configures the network resource identifier (NRI) length used for source SGSN discovery via NRI-FQDN
(Fully Qualified Domain Name) based DNS resolution.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description [ no ] nri length length plmn-id mcc mcc_value mnc mnc_value
no
Removes a configured NRI length.
length length
Specifies the number of bits to be used in the P-TMSI (bits 23 to 18) to define the NRI as an integer from 1
through 8.
Usage Guidelines Use this command to retrieve the NRI (identity of an SGSN) stored in bits 23 to 18 of the packet-temporary
mobile subscriber identity (P-TMSI). Up to eight NRI length values can be configured.
Important In the absence of this configuration, the MME treats the NRI as invalid. The MME will use a plain
RAI-based FQDN (and not an NRI-based FQDN) for DNS queries made to resolve the source SGSN.
Examples The following command creates an NRI length of 5 and associates it with an SGSN pool with the PLMN-ID
of 123:
nri length 5 plmnid mcc 123 mnc 23
peer-mme
Configures parameters that, when matched by another MME, specifies that MME as a peer for inter-MME
relocations.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description peer-mme { gummei mcc number mnc number group-id id mme-code code address ipv4_address | tai-match
priority value mcc number mnc number tac { area_code | any | start_area_code to end_area_code } address
ipv4_address }
no peer-mme { gummei mcc number mnc number group-id id mme-code code | tai-match priority value
}
no
Removes the configured peer Globally Unique MME Identifier (GUMMEI) or TAI match priority from this
service.
gummei mcc number mnc number group-id id mme-code code address ipv4_address
Specifies that an MME with values matching those configured in this GUMMEI is to be considered a peer
MME. This variable supports the lookup of an IP address for a peer MME based on the exact match of the
supporting keyword below (which make up the GUMMEI).
mcc number: Sets the mobile country code (MCC) for peer match as an integer from 100 through 999.
mnc number: Sets the mobile network code (MNC) for this peer match as a 2- or 3-digit integer from 00
through 999.
group-id id: Specifies the group identifier for the group to which this MME belongs as an integer from 0
through 65535.
mme-code code: Specifies the unique code for an MME service as an integer from 0 through 255.
address ipv4_address: Specifies the IP address of the peer MME in IPv4 dotted-decimal notation.
tai-match priority value mcc number mnc number tac { area_code | any | start_area_code to end_area_code
} address ipv4_address
Specifies that an MME with values matching those configured in this Tracking Area Identifier (TAI) match,
is to be considered a peer MME. This keyword provides a priority-ordered list of TAI descriptions where the
Tracking Area Code (TAC) field may be either an exact value, a range of values, or a "wildcard" value. It
also provides an IP address of the peer MME corresponding to the TAI description.
priority value:
mcc number: Sets the mobile country code (MCC) for peer match as an integer from 100 through 999.
mnc number: Sets the mobile network code (MNC) for this peer match as an integer from 00 through 999.
tac area_code: Sets a specific Tracking Area Code (TAC) for the peer MME match as an integer from 1
through 65535.
tac any: Specifies that any TAC value can be considered for a peer MME.
tac start_area_code to end_area_code: Specifies a range of TACs. MMEs within this range and matching
the rest of the criteria in this command are to be considered peer MMEs. start_area_code and end_area_code
are integers from 1 through 268435455.
address ipv4_address: Sets a specific IP address for this TAI peer MME match in IPv4 dotted-decimal
notation.
Usage Guidelines Use this command to configure parameters that, when matched by another MME, specifies that MME as a
peer for inter-MME relocations.
This command allows configuration for two relocation scenarios:
• gummei: an MME receives either an Attach or a TAU request with a Globally Unique Temporary
Identity (GUTI) that originated from another MME.
• tai-match: an MME receives an S1 Handover Required message and must select a new MME based on
the TAI.
Examples The following command identifies a peer MME with GUMMEI parameters:
peer-mme gummei mcc 123 mnc 12 group-id 40000 mme-code 100 address 10.2.3.4
peer-sgsn rai
Statically configures Routing Area Identity (RAI) parameters of the peer SGSN environment to facilitate
MME-SGSN relocations over S3 or Gn/Gp interfaces.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description peer-sgsn rai mcc mcc_value mnc mnc_value [ nri value ] rac value lac value address ip_address capability
[ gn ] [ gp ] [ s16 ] [ s3 ]
no peer-sgsn rai mcc mcc_value mnc mnc_value [ nri value ] rac value lac value
no
Deletes the specified peer-SGSN RAI parameter configuration from the MME Service configuration.
mcc mcc_value
Specifies the mobile country code (MCC) portion of the PLMN identifier as an integer from 100 through 999.
mnc mnc_value
Specifies the mobile network code (MNC) portion of the PLMN identifier as a 2- or 3-digit integer from 00
through 999.
nri value
Specifies the Network Resource Identifier (NRI) value, used as an additional identity, as an integer from 0
through 65535.
rac value
Specifies the Routing Area Code (RAC) used to facilitate a lookup for a specific peer SGSN as an integer
from 0 through 255.
lac value
Specifies the Location Area Code (LAC) value, used to facilitate a lookup for a specific peer SGSN, as an
integer from 0 through 65535.
address ip_address
Specifies an existing IP address of the peer SGSN in IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal
notation.
capability [ gn ] [ gp ] [ s16 ] [ s3 ]
Configures the GTP interface capability of the peer SGSN.
gn: Specifies that the peer SGSN is capable of communication over the Gn interface.
gp: Specifies that the peer SGSN is capable of communication over the Gp interface.
s16: Specifies that the peer SGSN is capable of communication over the S16 interface.
s3: Specifies that the peer SGSN is capable of communication over the S3 interface.
Usage Guidelines Use this command to configure parameters to facilitate a lookup for a specific peer SGSN. These parameters,
when matched by an SGSN, specifies that SGSN as a peer for inter-RAT relocations.
The peer-sgsn command allows configuration for two relocation scenarios:
• Routing Area Identity (RAI) configuration is used for the lookup of an IP address for a peer MME based
on the exact match of the RAI (and optionally NRI).
• Radio Network Controller Identification (RNC-ID) configuration is used for the lookup of an IP address
for a peer MME based on the exact match of the RNC-ID.
Up to 32 (combined total) peer-SGSN RAI and RNC-ID entries may be configured per MME service.
Examples The following command configures an SGSN lookup using RAI parameters with Gp interface capability:
peer-sgsn rnc-id mcc 123 mnc 12 nri 1557 rac 33 lac 3542 address 10.4.3.2 capability gp
peer-sgsn rnc-id
Statically configures Radio Network Controller Identification (RNC-ID) parameters of the peer SGSN
environment to facilitate MME-SGSN relocations over S3 or Gn/Gp interfaces.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description peer-sgsn rnc-id mcc mcc_value mnc mnc_value rnc value address ip_address capability [ gn ] [ gp ] [
s16 ] [ s3 ]
no peer-sgsn rnc-id mcc mcc_value mnc mnc_value rnc value
no
Deletes the specified peer-SGSN RAI parameter configuration from the MME Service configuration.
mcc mcc_value
Specifies the mobile country code (MCC) portion of the PLMN identifier as an integer from 100 through 999.
mnc mnc_value
Specifies the mobile network code (MNC) portion of the PLMN identifier as a 2- or 3-digit integer from 00
through 999.
rnc value
Specifies the Radio Network Controller (RNC) identification number used to facilitate a lookup for a specific
peer SGSN as an integer from 0 through 65535.
address ip_address
Specifies an existing IP address of the peer SGSN in IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal
notation.
capability [ gn ] [ gp ] [ s16 ] [ s3 ]
Configures the GTP interface capability of the peer SGSN.
gn: Specifies that the peer SGSN is capable of communication over the Gn interface.
gp: Specifies that the peer SGSN is capable of communication over the Gp interface.
s16: Specifies that the peer SGSN is capable of communication over the S16 interface.
s3: Specifies that the peer SGSN is capable of communication over the S3 interface.
Usage Guidelines Use this command to configure parameters to facilitate a lookup for a specific peer SGSN. These parameters,
when matched by an SGSN, specifies that SGSN as a peer for inter-RAT relocations.
The peer-sgsn command allows configuration for two relocation scenarios:
• Radio Network Controller Identification (RNC-ID) configuration is used for the lookup of an IP address
for a peer MME based on the exact match of the RNC-ID.
• Routing Area Identity (RAI) configuration is used for the lookup of an IP address for a peer MME based
on the exact match of the RAI (and optionally NRI).
Multiple peer-sgsn RNC-ID records can be configured for the same MCC/MNC/RNC, each with different IP
addresses. During a handover, if the initial peer SGSN rejects the forward relocation request, the MME will
step through any alternate peer SGSNs to attempt the handover.
Up to 32 (combined total) peer-SGSN RAI and RNC-ID entries may be configured per MME service.
Examples The following command configures an SGSN lookup using RNC-ID parameters with Gn interface capability:
peer-sgsn rnc-id mcc 123 mnc 12 rnc 2000 address 10.2.3.4 capability gn
pgw-address
Configures the IPv4 or IPv6 address of the PDN Gateway (P-GW), specifies the protocol for S5 and S8
interfaces, and sets other parameters within the MME service. By default S5 and S8 use GTP protocol for
this.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Removes a previously configured IP address for a P-GW along with the S5 and S8 interface of P-MIP protocol
type, and other parameters from this MME service.
ipv4_address | ipv6_address
Specifies the IP address of the P-GW in IPv4 dotted-decimal or IPv6 colon-separated-hexadecimal notation.
s5-s8-protocol pmip
Specifies that Proxy-MIP is to be used for S5 and S8 interfaces with the P-GW. By default S5 and S8 interface
uses GTP protocol.
pmip Sets the protocol to Proxy-MIP for S5 and S8 interface.
weight weight
Specifies the weight (preference) assigned to the address P-GW for load balancing. weight is an integer from
1 through 100 where 1 is the least preferred and 100 is the most preferred. If no weight is specified, the P-GW
address is assigned a default weight of 1.
If a weight is assigned to an address, the weights of the P-GW(s) (that are operational) are totaled, and then
a weighted round-robin selection is used to distribute new primary PDP contexts among the P-GW(s) according
to their weights. As with all weighted round-robin algorithms, the distribution does not look at the current
distribution, but simply uses the weights to distribute new requests. For example, two P-GWs assigned weights
of 70 and 30 would distribute 70% of calls to one, and 30% to the other. The sum of all weights do not need
to total 100.
Usage Guidelines Use this command to configure the PDN Gateway (P-GW) addresses to use with MME service. This command
also changes the default protocol from GTP to P-MIP for the S5 and S8 interface, and assigns a weight to use
when sharing the load between associated P-GWs. A maximum of 16 P-GW addresses can be configured
with this command.
This command only changes the use of protocol for the S5 and S8 interface. By default a P-GW uses GTP
protocol for S5 and S8 interfaces. This command allows an operator to change the protocol to P-MIP for S5
and S8 interface.
When weight is used, the weights of the operational P-GW(s) are totaled and then weighted round-robin
selection is used to distribute new default bearer contexts among P-GW(s).
Examples The following command associates the P-GW IP address of 192.168.3.1 to the MME service with S5 and S8
protocol as P-MIP and weight as 90:
pgw-address 192.168.3.1 s5-s8-protocol pmip weight 90
The following command removes the above configured P-GW IP address and other parameters:
no pgw-address 192.168.3.1 s5-s8-protocol pmip
plmn-id
Configures the Public Land Mobile Network (PLMN) identifier for this MME service. The PLMN identifier
consists of the Mobile Country Code (MCC) and Mobile Network Code (MNC). A single PLMN ID can be
configured per MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Removes the configured PLMN identifier for this MME service.
Caution Removing the PLMN identifier is a disruptive operation; the MME service will be restarted.
mcc mcc_value
Specifies the mobile country code (MCC) portion of the PLMN identifier as an integer from 100 through 999.
mnc mnc_value
Specifies the mobile network code (MNC) portion of the PLMN identifier as a 2- or 3-digit integer from 00
through 999.
Usage Guidelines Use this command to set the PLMN identifier for this MME service.
Caution Changing or removing the PLMN identifier is a disruptive operation; the MME service will be restarted.
Important To configure additional PLMN IDs for this MME service, refer to the network-sharing command described
in this chapter.
Examples The following command configures the PLMN identifier with MCC value as 102 and MNC value as 20 for
this MME service:
plmn-id mmc 102 mnc 20
policy attach
Configures parameters for the UE Attach procedure.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description policy attach { imei-query-type { imei | imei-sv | none } [ verify-equipment-identity [ allow-on-eca-timeout
| deny-greylisted | deny-unknown | verify-emergency ] ] | set-ue-time { disable | enable [ prefer-mme |
prefer-msc ] }
default policy attach { imei-query-type | set-ue-time }
default
Returns the command to its default setting of none for imei-query-type and disabled for set-ue-time.
• deny-greylisted: Configures the MME to deny grey-listed equipment during the attach procedure.
• deny-unknown: Configures the MME to deny unknown equipment during the attach procedure.
• verify-emergency: Configures the MME to ignore the IMEI validation of the equipment during the
attach procedure in emergency cases. This keyword is only supported in release 12.2 and higher.
Usage Guidelines Use this command to configure various MME settings used during the UE Attach procedure.
Examples The following command configures the MME to query the UE for its IMEI and to verify the UEs equipment
identity over the S13 interface with an EIR:
policy attach imei-query-type imei verify-equipment-identity
policy idle-mode
Configures the user-defined behavioral policies of session management for an LTE subscriber in an MME
service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the policy configuration to the default behavior for subscriber IDLE mode Detach. The default behavior
is Detach implicit.
idle-mode detach
Configures the IDLE mode Detach behavior of a UE.
detach
Defines the Detach procedure while the UE is in IDLE mode.
explicit
Enables the Explicit Detach while a UE is in IDLE mode. The system will page the UE before Detach procedure
is started, and then perform the Explicit Detach procedure.
implicit
Enables the Implicit Detach while a UE is in IDLE mode. The system never sends any message to the UE
before Detach, and executes the Implicit Detach procedure immediately. This is the default behavior.
Usage Guidelines Use this command to set the user-defined policies for session management in this MME service.
Examples The following command sets the Idle Mode Detach policy to Implicit for a user in this MME service:
policy idle-mode detach implicit
policy inter-rat
Configures inter-RAT policy settings.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Disables the function.
ignore-sgsn-context-id
Configures the MME to ignore any Context-ID mismatch between HSS and HLR and to use the Context-ID
from the HSS to override the Context-ID from the source SGSN. If this option is disabled (default), the MME
will drop the PDN when there is a Context-ID mismatch.
indirect-forwarding-tunnels always
Enables establishment of Indirect Data Forwarding Tunnels (IDFT) for Gn/Gp-based Serving Radio Network
Subsystem (SRNS) relocations. By default, the MME is configured to never establish IDFT.
select-topologic-sgw interface gn
Configures the MME to select the S-GW based on topological closeness to the P-GW for Gn/Gp handoff
scenarios. Weighted distribution will occur across node pairs in the same degree and same order. By default
this functionality is disabled.
During inter-RAT Gn/Gp based handoffs, the MME does not learn the P-GW host name from the old Gn/Gp
SGSN as part of UE context. Without the P-GW host name, selection of the topologically closest S-GW is
not possible per 3GPP standards. This functionality enables the MME to use a proprietary mechanism for
learning the P-GW host name. For S3 & S10 cases, there is no need to enable this command, as GTPv2 allows
the P-GW host name to be communicated to/from S4-SGSN/MME.
This functionality requires the gw-selection co-location or gw-selection topology commands to be enabled
in the call-control-profile mode.
Note: The P-GW is anchored in the inter-RAT handoff scenarios, so regardless of the preferred weight specified
in gw-selection, the MME always considers the S-GW's weight for weighted distribution purposes.
Usage Guidelines Use this command to enable or disable establishment of indirect data forwarding tunnels for Gn/Gp-based
SRNS relocations, and to enable or disable Context-Identifier overriding, and to enable or disable learning
the P-GW host name during Gn/Gp handoffs for purposes of topologically-close S-GW distribution.
Examples The following command enables establishment of indirect data forwarding tunnels for Gn/Gp-based SRNS
relocations:
policy inter-rat indirect-forwarding-tunnels always
policy network
Configures the MME to indicate to the P-GW that all peer SGSNs support dual-addressing for bearers and,
subsequently, dual-addressing must be supported for all IPv4 and IPv6 PDNs. Dual-addressing on SGSNs is
based on the UE's capability to support inter-RAT roaming.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of disabled.
no
Removes the ability to send dual-addressing support messaging from the MME to the P-GW.
dual-addressing-supported
Specifies that the MME shall indicate to the P-GW that dual-addressing is supported.
Usage Guidelines Use this command to configure the MME to send messaging to the P-GW that indicate that all peer SGSNs
support dual-addressing for bearers and, subsequently, dual-addressing must be supported for all IPv4 and
IPv6 PDNs.
Important This command can be used for Pre-release 8 and Release 8 SGSNs.
policy overcharge-protection
Enables overcharge protection where the MME can detect and signal a Loss of Signal Contact to the S-GW
which in turn informs the P-GW to stop charging.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of disabled. This provides the same behavior as the no keyword
option.
no
Disables overcharge protection. This provides the same behavior as the default keyword option.
s1ap-cause-code-group group_name
group_name: Specify the name of a preconfigured S1-AP Cause Code Group.
When the received cause code from the eNodeB matches any the cause codes defined in this Cause Code
Group, the MME sets the ARRL (Abnormal Release of Radio Link) bit in the Indication IE of the Release
Access Bearer Request to the S-GW.
For more information about creating an S1-AP Cause Code Group, refer to the cause-code-group command
in the LTE Policy Configuration Mode Commands chapter, and the class command in the S1AP Cause Code
Configuration Mode Commands chapter.
Usage Guidelines
Important Overcharge protection is a license-controlled feature. A valid feature license must be installed prior to
configuring this feature. Contact your Cisco account representative for more information.
Use this command to enable or disable overcharging protection for this MME service. When enabled, the
MME can detect and signal a Loss of Signal Contact to the S-GW which in turn informs the P-GW to stop
charging for the UE.
Refer to the Overcharging Protection chapter of the MME Administration Guide for more information about
this feature.
Examples The following command enables overcharging protection for the S1-AP cause code defined in the S1AP Cause
Code Group group1:
policy overcharge-protection s1ap-cause-code-group group1
policy overload
Configures the traffic overload policy to control congestion in this service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the traffic overload policy action to the fault behavior of Reject.
drop
Specifies that the system is to drop the incoming packets with new session requests to avoid overload on
MME node. Default: Disabled
reject
Configures the system to reject the new session/call request and responds with a reject message when the
threshold for allowed call sessions is crossed on the MME node. Default: Enabled
Usage Guidelines Use this command to set the user-defined policies for new call connection attempts when an MME service is
overloaded.
Congestion policies at the service-level can be configured for an individual service. When congestion control
functionality is enabled, these policies dictate how services respond when the system detects that a congestion
condition threshold has been crossed.
Examples The following command sets the nw call connect policy to reject the new session/call request in an MME
service:
policy overload reject
policy pdn-deactivate
Configures the MME to deactivate a PDN connection if the charging characteristics (CC) AVP changes in
the standalone Insert Subscriber Data Request (ISDR) or the Update Location Answer (ULA).
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
This command filter instructs the MME to disable the PDN deactivation configuration defined with this policy
command.
pdn-deactivate
This keyword configures the MME to deactivate the PDN connection based on the AVP included to filter the
keyword.
cc-change
This filter represents the charging characteristics AVP. If it is included in the command, then the MME
deactivates the PDN connection when the charging characteristics (CC) AVP changes in the standalone Insert
Subscriber Data Request (ISDR) or the Update Location Answer (ULA).
Usage Guidelines With policy pdn-deactivate cc-change configured, the MME updates the subscriber DB with the CC
information so that the MME would be able to create a PDN connection with the new CC values.
If the deactivated PDN is the last PDN, then the UE is detached from the network and during the UE's next
Attach procedure the updated CC information is taken from the subscriber DB and included in a Create Session
Request.
If the information is absent from the DB, and if CC IE is not present in transferred PDNs of Context Response
message during 3G to 4G TAU, then the MME does not send local default CC IE in CSReq and the PDN is
activated
'Disabled' is the default behavior. If deactivation for CC changes is not enabled, then the MME updates the
APN's CC information in the subscriber DB and keeps the PDN active if the CC information changes in or
is absent from the ISDR.
To confirm the MME's current configuration regarding PDN deactivation, use the following command. The
illustration below is a partial display to indicate the current configuration, which will be either 'enabled' or
'disabled':
show mme-service name service_name
…
…
Policy S1-Reset : Idle-Mode-Entry
Policy PDN-Deact CC-Change : Enabled
Policy Nas-Non-Del : Disabled
…
Examples The following command configures the MME to deactivate the PDN connection when the CC information
changes in or is absent from received ISDR:
policy pdn-deactivate cc-change
policy pdn-reconnection
Configures the action by the MME when a PDN connection request to an already connected APN is being
processed by the MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the policy for PDN reconnection to its default behavior of Reject.
multiple
Allows multiple connections to a PDN with the same APN and PDN Type. In this case, the existing connection
is left unchanged, and the MME attempts to establish an additional connection to the PDN. Default: Disabled
reject
Configures the MME to deny or reject the request, by sending a PDN Connection Reject command. This is
the default behavior. Default: Enabled
restart
Deletes the existing connection and initiates an attempt to establish a new connection. Default: Disabled
Usage Guidelines Use this command to set the user-defined policies for PDN reconnection attempt procedures initiated by a
UE in an MME service.
While attached the UE can request connections to PDNs. The PDNs are identified by APN (Access Point
Name) and PDN Type (ipv4, ipv6 or ipv4v6).
If the UE requests connection to a PDN for which a connection with the same APN name and PDN type
already exists, the MME can: 1) deny or reject the request, by sending a PDN connection reject command; 2)
allow multiple connections to a PDN with same APN and PDN Type; or 3) delete the existing connection,
and attempt to establish a new connection.
Examples The following command sets the PDN reconnect policy to delete the existing PDN and start the attempt to
establish a new connection in an MME service:
policy pdn-reconnection restart
policy s1-reset
Configures how the MME responds to an S1 interface reset.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of idle-mode-entry.
detach-ue
detach-ue: Specifies that UEs are to be implicitly detached from the service upon S1 interface reset.
idle-mode-entry
idle-mode-entry: Specifies that UEs are to be placed into an idle mode condition during S1 interface reset.
Usage Guidelines Use this command to configure how the MME reacts to an S1 interface reset condition.
Examples The following command configures the MME to place UEs into an idle state while the S1 interface is being
reset:
policy s1-reset idle-mode-entry
policy sctp-down
Configures how the MME responds to a failure of the Stream Control Transmission Protocol (SCTP) connection
from the eNodeB.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of idle-mode-entry.
detach-ue
detach-ue: Specifies that UEs are to be detached from the service when the SCTP connection from the eNodeB
fails.
idle-mode-entry
idle-mode-entry: Specifies that UEs are to be placed into an idle mode condition when the SCTP connection
from the eNodeB fails.
Usage Guidelines Use this command to configure how the MME reacts to an SCTP connection failure condition.
Examples The following command configures the MME to place UEs into an idle state while the SCTP connection from
the eNodeB fails:
policy sctp-down idle-mode-entry
policy service-request
Configure the behavior of the MME when an initial context setup failure is received during a service request
or extended service request procedure.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default behavior, where it detaches the UE when an initial context setup failure
is received during a service request or extended service request procdure.
Usage Guidelines Use this command to configure the behavior of the MME when an initial context setup failure is received
during a service request or extended service request procedure.
Examples The following command configures the MME to detach the UE when an initial context failure occurs and the
eNodeB returns a cause code which matches any of the cause codes configured in the idle S1-AP cause code
group:
policy service-request initial-context-setup-failure s1ap-cause-code-group idle action idle-mode-entry
policy srvcc
Configures the MME to initiate an HSS Purge after the SRVCC HO where the UE supports DTM. It also
allows configuration of a purge timeout value in seconds.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default behavior, where the MME does not initiate a HSS Purge after the SRVCC
HO.
no
Returns the command to its default behavior, where the MME does not initiate a HSS Purge after the SRVCC
HO. This provides the same function as the default keyword.
purge-timer seconds
Defines how long in seconds the Purge Timer will run. This is applicable only for SRVCC Handoff without
PS Handoff support scenarios.
For example, if purge-timer is set to 20 seconds :
If the Context Transfer happens 10 seconds after SRVCC HO, the MME intiates an HSS Purge.
If the Context Transfer happens 30 seconds after SRVCC HO, the MME will NOT initiate an HSS Purge
because the Purge Timer has expired.
seconds must be entered as an integer from 1 through 24000.
Usage Guidelines Use this command to configure the MME to perform the Purge UE procedure to the HSS for UEs which
support Dual Transfer Mode (DTM). When configured, the MME initiates an HSS Purge after the following
two SRVCC HO scenarios:
For SRVCC Handoff with PS Handoff support, the Purge S6a message is sent immediately after successful
completion of the Handoff. For this scenario, the configurable purge timer is not used.
For SRVCC Handoff without PS Handoff support, the configurable timer is initiated and the Purge S6a
message is sent if a SGSN Context Request is received prior to timer expiry. If a Context Failure occurs, no
HSS Purge S6a message is sent.
Examples The following command configures the MME to perform the Purge UE procedure and sets the purge timer
to 20 seconds.
policy srvcc purge-timer 20
policy tau
Configures parameters for the tracking area update (TAU) procedure.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description policy tau { imei-query-type { imei | imei-sv | none } [ verify-equipment-identity [ allow-on-eca-timeout
| deny-greylisted | deny-unknown | verify-emergency ] ] | initial-context-setup-failure
s1ap-cause-code-group group_name action detach-ue | set-ue-time { disable | enable [ prefer-mme |
prefer-msc ] }
default policy tau { imei-query-type | initial-context-setup-failure | set-ue-time }
default
Returns the command to its default settings:
imei-query-type: none
initial-context-setup-failure: Returns the MME to the default behavior, where it moves the UE to IDLE
MODE when an initial context setup failure is received during a TAU procedure.
set-ue-time: disabled
Usage Guidelines Use this command to configure various MME settings used during the tracking area update (TAU) procedure.
Examples The following command configures the MME to query the UE for its IMEI and to verify the UEs equipment
identity over the S13 interface with an EIR:
policy tau imei-query-type imei verify-equipment-identity
The following command configures the MME to detach the UE when an initial context failure occurs and the
eNodeB returns a cause code which matches any of the cause codes configured in the "detach" S1-AP cause
code group:
policy tau initial-context-setup-failure s1ap-cause-code-group detach action detach-ue
pool-area
Creates an MSC server pool area for the Sv interface or specifies an existing pool area, and enters MME MSC
Server Pool Area Configuration Mode.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Removes the configured pool-area for this MME service.
pool_area_name
Specifies the name of the pool-area as an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to create an MSC server pool area for the Sv interface or specify an existing pool area
configuration and enter the MME Pool Area Configuration Mode.
The command also defines the MSC server selection method for the pool area, using either the IMSI hash
value, or round-robin.
This command is also used to remove an existing pool area.
A maximum of 24 pool areas can be configured per MME service.
When configured, the MME attempts to select an MSC using the following selection order:
1. Pool area that matches the PLMN and of type hash.
2. Pool area that matches the PLMN and of type round-robin.
3. Pool area that does not have PLMN associated and of type hash.
4. Pool area that does not have PLMN associated and of type round-robin .
Entering this command results in one of the following prompts, based on the pool selection method specified:
[context_name]host_name(config-mme-pool-area-hash-value)#
[context_name]host_name(config-mme-pool-area-round-robin)#
Additional commands are defined in the MME MSC Server Pool Area Configuration Mode Commands chapter.
Examples The following command defines a pool area named msc_pool_east and configures it for the round robin
selection mode.
pool-area msc_pool_east type round-robin
ps-lte
Configures the Public Safety LTE (PS-LTE) mode of operation for this MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
Disables PS-LTE mode of operation.
Usage Guidelines This feature requires that a valid license key be installed. Contact your Cisco Account or Support representative
for information on how to obtain a license.
Use this command to enable the MME service for use in a Public Safety LTE (PS-LTE) network. In this mode,
the MME is co-located with an S-GW and at least one P-GW, and the MME must always use the co-located
S-GW and a co-located P-GW for all calls that it handles. This requires configuring the IP addresses of the
S11 interface of the S-GW as part of the MME service configuration.
Configuration of the S5/S8 interface to the P-GW must be configured separately as part of an APN profile
configuration (refer to the pgw-address command within the APN Profile Configuration Mode chapter in the
Command Line Interface Reference).
When enabled, all other S-GW selection mechanisms are overridden. The MME will only use the S-GW
configured for PS-LTE operation and the P-GW configured in the matching APN profile, regardless of any
other configuration present.
Examples The following command enables PS-LTE mode for this MME service and configures the IP address of the
S11 interface for the S-GW as 192.60.60.7.
ps-lte sgw 192.60.60.7
relative-capacity
Configures a relative capacity variable that is sent to the eNodeB for use in selecting an MME in order to load
balance the pool.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of 255.
number
Specifies the relative capacity or weight of an MME compared to others in an MME pool as an integer from
0 through 255.
Default: 255
Usage Guidelines Use this command to configure the relative capacity or weight of this MME in comparison to other MMEs
in a pool. This value is sent to the eNodeB in the S1AP S1 SETUP RESPONSE message.
If this value is changed after the S1 interface is initialized, the MME CONFIGURATION UPDATE message
is used to update the eNodeB with the change.
Examples The following command sets this MME with a relative capacity or weight of 100:
relative-capacity 100
s13
Enables the MME to send additional Mobile Identity check Requests (MICR) towards the EIR over the S13
interface.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
no
This command filter instructs the MME to remove and disable the specified feature configuration from the
MME Service configuration.
Usage Guidelines By default, this additional imei checking functionality is disabled. Use this command to configure the MME
to send additional Mobile Identity check Requests (MICR) towards the EIR over the S13 interface. You must
choose at least one triggering UE procedure. You may repeat the command as needed to configure multiple
triggering UE procedures.
Examples The following commands must be issued separately. They instruct the MME to send additional IMEI check
Requests to the EIR during UE Attach procedures and UE Handovers :
s13 additional-id-check attach
s13 additional-id-check handover
s1-mme ip
Configures the quality of service (QoS) differentiated service code point (DSCP) used when sending packets
of a particular 3GPP QoS class over the S1-MME interface.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description s1-mme ip qos-dscp { af11 | af12 | af13 | af21 | af22 | af23 | af31 | af32 | af33 | af41 | af42 | af43 | be | cs0
| cs1 | cs2 | cs3 | cs4 | cs5 | cs6 | cs7 | ef }
default s1-mme ip qos-dscp
qos-dscp { af11 | af12 | af13 | af21 | af22 | af23 | af31 | af32 | af33 | af41 | af42 | af43 | be | cs0 | cs1 | cs2 |
cs3 | cs4 | cs5 | cs6 | cs7 | ef }
Default: af11
Specifies the DSCP for the specified QoS traffic pattern. qos-dscp can be configured to any one of the
following:
af11: Assured Forwarding 11 per-hop-behavior (PHB)
af12: Assured Forwarding 12 PHB
af13: Assured Forwarding 13 PHB
af21: Assured Forwarding 21 PHB
af22: Assured Forwarding 22 PHB
af23: Assured Forwarding 23 PHB
af31: Assured Forwarding 31 PHB
af32: Assured Forwarding 32 PHB
af33: Assured Forwarding 33 PHB
af41: Assured Forwarding 41 PHB
af42: Assured Forwarding 42 PHB
Usage Guidelines DSCP levels can be assigned to specific traffic patterns to ensure that packets are delivered according to the
precedence with which they are tagged. The diffserv markings are applied to the IP header of every subscriber
packet transmitted over the S1-MME interface(s).
Examples The following command sets the DSCP-level for traffic sent over the S1-MME interface to af12:
s1-mme ip qos-dscp af12
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the SCTP port to the default value of 36412 to communicate with the eNodeBs using S1-MME interface.
port_num
Specifies the SCTP port number to communicate with the eNodeBs using S1-MME interface as an integer
from 1 through 65535. Default: 36412
Usage Guidelines Use this command to assign the SCTP port with SCTP socket to communicate with the eNodeB using S1AP.
Only one SCTP port can be associated with one MME service.
Examples The following command sets the default SCTP port number 699 for to interact with eNodeB using S1AP on
S1-MME interface:
default s1-mme sctp port
s1-ue-context-release
Specifies the cause code to be sent in a UE-CONTEXT-RELEASE message initiated by the MME upon the
reception of any unexpected procedure over Initial-UE from the eNodeB, such as TAU, Service Request,
Extended Service Request, Attach Request..
Product MME
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Syntax Description s1-ue-context-release reason init-ue-from-enodeb cause type { nas value nas_value | radio value
radio_value }
default s1-ue-context-release reason init-ue-from-enodeb cause
default
Resets the MME Service configuration to the system defaults.
• 2 - Successful Handover
• 3 - Release due to E-UTRAN Generated Reason
• 4 - Handover Cancelled
• 5 - Partial Handover
• 6 - Handover Failure In Target EPC/eNB Or Target System
• 7 - Handover Target not allowed
• 8 - TS1RELOCoverall Expiry
• 9 - TS1RELOCprep Expiry
• 10 - Cell not available
• 11 - Unknown Target ID
• 12 - No Radio Resources Available in Target Cell
• 13 - Unknown or already allocated MME UE S1AP ID
• 14 - Unknown or already allocated eNB UE S1AP ID
• 15 - Unknown or inconsistent pair of UE S1AP ID
• 16 - Handover desirable for radio reasons
• 17 - Time critical handover
• 18 - Resource optimisation handover
• 19 - Reduce load in serving cell
• 20 - User inactivity
• 21 - Radio Connection With UE Lost
• 22 - Load Balancing TAU Required
• 23 - CS Fallback Triggered
• 24 - UE Not Available For PS Service
• 25 - Radio resources not available
• 26 - Failure in the Radio Interface Procedure
• 27 - Invalid QoS combination
• 28 - Inter-RAT redirection
• 29 - Interaction with other procedure
• 30 - Unknown E-RAB ID
• 31 - Multiple E-RAB ID instances
• 32 - Encryption and/or integrity protection algorithms not supported
• 33 - S1 intra-system Handover triggered
• 34 - S1 inter system Handover triggered
Usage Guidelines By default, an MME initiates the UE-CONTEXT-RELEASE with cause NAS-Normal-Release whenever the
MME receives any procedure Request over Initial-UE if the UE is in the connected state. This command
makes it possible for the operator to configure a preferred cause code for the reason of the disconnect.
Important In earlier releases, the keyword was init-ue-from-enodeb-for-tau. In release 19.2, the name and behavior
associated with this keyword changed. the keyword name is init-ue-from-enodeb. In support of backward
compatibility, the MME will accept configurations with either form of the keyword. When the operator
explicitly saves the configuration, the configuration will save using the new form of the keyword.
Beginning with release 19.2, the init-ue-from-enodeb reason instructs the MME to initiate the
UE-CONTEXT-RELEASE with cause NAS-Normal-Release whenever the MME receives a request over
Initial-UE not just for TAU but for all TAU and non-TAU scenarios (such as Service Request, Attach, and
Extended-Service-Request) if the UE is in the connected state.
setup-timeout
Configures the timeout duration for setting up MME calls in this MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Sets the call setup timeout duration to the default value of 60 seconds.
dur
Specifies the call setup timeout duration (in seconds) for MME calls after which the attempt will be discarded.
dur is an integer from 1 through 10000. Default: 60
Usage Guidelines Use this command to configured the timeout duration for setting up an MME call with an MME service. One
this timer expires, the call setup procedure will be discarded within this MME service.
Examples The following command sets the default setup timeout duration of 60 seconds for MME calls:
default setup-timeout
sgw-retry-max
Sets the maximum number of SGW selection retries to be attempted during Attach/HO/TAU. By default, this
functionality is not enabled.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
Command Modes Exec > Global Configuration > MME Service Configuration
configure > mme-service profile_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-mme-serviceprofile_name)#
no
Disables the configuration for the maximum number of retries.
max_number
Sets the maximum number of retries possible. Enter an integer from 0 to 5. If 0 (zero) is configured, then the
MME sends Create-Session-Request to the 1st SGW and if that SGW does not reply, the MME does not select
any further SGW to retry. The MME then rejects the ongoing procedure (Attach/HO/TAU) and sends a Reject
message.
Usage Guidelines Using the this command sets a limit to the maximum number of SGW selection retries to be attempted during
Attach/HO/TAU. This means, the total number of tries would be 1 (the initial try) + the sgw-retry-max value
(the maximum number of retries).
Entering a value with this command overrides the default behavior. If no value is configured, then the MME
uses or falls back to the default behavior which is in compliance with 3GPP TS 29.274, Section 7.6. The MME
sends Create-Session-Request message to one SGW in the pool. If the SGW node is not available, the MME
picks the next SGW from the pool and again sends a Create-Session-Request message. The MME repeats this
process. For an Attach procedure, the MME tries up to five (1 + 4 retries) different SGWs from the pool. In
the case of a HO procedure, the MME will try every SGW in the entire pool of SGWs sent by the DNS. If
there are no further SGW nodes available in the DNS pool or if the guard timer expires, then MME stops
trying and sends a Reject with cause "Network-Failure" towards the UE and the UE must restart the
Attach/Handover procedure.
Benefits of this configuration -- The amount of signaling at Attach or Handover can be reduced and the amount
of time to find an available SGW can be reduced.
If the sgw-retry-max command is configured under both the MME service and the Call-Control Profile, then
the configuration under Call-Control Profile takes precedence.
Examples Use this command to enable the functionality for limiting the number of SGWs tried during Attach/HO/TAU
to 2 retries:
sgw-retry-max 2
snmp trap
Enables or disables the SNMP trap for S1 interface connection establishment.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Returns the command to its default setting of disabled.
no
Disables the SNMP trap.
s1-initial-establishment
Specifies that the SNMP trap for the initial S1 interface connection establishment is to be enabled or disabled.
s1-path-establishment
Specifies that the SNMP trap for the S1 path establishment is to be enabled or disabled.
Usage Guidelines Use this command to enable or disabled the SNMP trap for S1 interface connection establishment.
statistics
Configures the statistics collection mode for the MME service.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Configures the command to its default setting, where statistics are collected per eNodeB.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to collect statistics for this MME service at the eNodeB level (default), or at the TAI level.
Caution Changing this collection mode will restart the MME service and will clear all statistics at the MME
service and eNodeB level.
When configured to collect statistics per TAI, the MME will collect statistics only for the TAIs that are
configured in the LTE TAI Management Database that is associated with the MME service.
If a specific TAI is configured within multiple TAI Management Databases, the records collected for that TAI
will be a sum of all counters for all TAI Management Databases to which it belongs.
Refer to the TAI Schema chapter in the Statistics and Counters Reference for a listing of all bulk statistics
impacted by this command.
Refer also to the show mme-service statistics command to display TAI statistics.
Examples The following command configures this MME service to collect statistics per TAI, instead of per eNodeB.
statistics collection-mode tai -noconfirm
ue-db
Configures the UE database that is maintained by the MME as a cache of EPS contexts per UE keyed by
IMSI/GUTI to allow the UE to attach by a Globally Unique Temporary Identity (GUTI) and reuse previously
established security parameters. This cache will be maintained in each session manager where the first attach
occurred for the UE.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > MME Service
configure > context context_name > mme-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mme-service)#
default
Resets the UE database purge timer timeout to the default value of 10080 minutes.
purge-timeout dur_mins
Sets the timeout duration (in minutes) for MME to store the UE database in cache memory. This timer starts
when the UE goes dormant.
dur_mins is an integer from 1 through 20160. Default: 10080
Usage Guidelines Use this command to set timeout duration for MME to hold UE database information in cache memory.
The MME DB acts as a cache for storing subscriber related information. This subscriber related information
helps reduce signaling traffic. The MME DB is a part of the Session Manager and interfaces between the
Session Manager Application and Evolved Mobility Management Manager to provide access to the cached
data.
Examples The following command configures the MME database cache timer to hold the UE information up to 7 days
(10080 minutes) in the MME Database:
default ue-db purge-timeout
The MPLS-IP Configuration Mode is used for configuration of Multiprotocol Label Switching (MPLS) IP
forwarding specific parameters. This mode is entered from the Context Configuration Mode.
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration
configure > context context_name > mpls-ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mpls)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
protocol ldp
Creates or removes the MPLS label distribution protocol (LDP) configuration, or configures an existing
protocol and enters the MPLS-LDP Configuration Mode in the current context. This command configures
the protocol parameters for MPLS LDP.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > MPLS-IP Configuration
configure > context context_name > mpls-ip
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mpls)#
no
Removes the configured MPLS-LDP mode from MPLS-IP Configuration mode.
Usage Guidelines Use this command to create/ or remove the MPLS LDP configuration, or configure an existing protocol. If
required mode already exists it enters the MPLS-LDP Configuration Mode in the current context.
Entering this command results in the following prompt:
[context_name]hostname(config-ldp)#
The commands configured in this mode are defined in the MPLS-LDP Configuration Mode Commands chapter.
Examples The following command creates and enters the MPLS-LDP Protocol mode:
protocol ldp
The MRME Service Configuration Mode provides commands to enable a trusted WLAN network to provide
access to the Evolved Packet Core (EPC) using a AAA peer functionality.
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
Important Available commands or keywords/variables vary based on platform type, product version, and installed
license(s).
aaa
This command allows you to control the range of EAP-payload size, or restrict the Framed-MTU AVP from
being forwarded in the Auth-Request message to the AAA server.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
no
Disables SaMOG from forwarding Framed-MTU AVP in the Auth-Request message to the AAA server.
eap_payload_size
Specifies the EAP payload limit for the AAA server to use during the Auth-Response on the link between the
NAS and the peer.
twan_profile_name must be an integer from 64 through 1500.
Usage Guidelines This command enables SaMOG to support EAP TLS and EAP TTLS-based authentication. Use this command
to control the range of EAP-payload size, or restrict the Framed-MTU AVP from being forwarded in the
Auth-Request to the AAA server.
Examples The following command sets the EAP payload size to 1000:
aaa send framed-mtu 1000
associate
This command associates one or more TWAN profile with this MRME service.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
no
Removes the association of the TWAN profile with the MRME service.
twan_profile_name
Specifies the twan profile to associate with the MRME service.
twan_profile_name must be an integer from 1 through 64.
Usage Guidelines Use this command to associate one or more TWAN profile with the MRME service. Once a TWAN profile
is associated with the MRME service, SaMOG uses the Radius clients and access type for the clients configured
under the TWAN Profile while processing the Radius messages from WLC.
For more information on configuring the Radius clients and access type, refer the TWAN Profile Configuration
Mode Commands section.
Examples The following command associates the TWAN profile twan1 with this MRME service.
associate twan-profile twan1
attribute
This command allows you to include SSID and Calling-Stationd-Id AVP values as part of DER messages
over STa Interfaces.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
Syntax Description attribute sta { anid { called-station-id | ssid | ssid-wlan-prefix | wlan-string } | calling-station-id { imsi |
ue-mac } }
default attribute sta { anid | calling-station-id }
default
Sets the configuration to its default value.
Default calling-station-id: imsi
Default anid: wlan-string
Usage Guidelines Use this command to include the received SSID and Calling-Station-Id values in the ANID/ Calling-Station-Id
AVP as part of DER messages over STa Interfaces.
Examples The following command includes ue-mac information from the calling-station-id in the DER message.
attribute sta calling-station-id ue-mac
bind
This command allows you to configure an IPv4 and/or IPv6 address to be used as the connection point for
establishing SaMOG sessions to handle authentication and accounting messages.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
no
Removes a previously configured binding.
address ipv4_address
Specifies the IP address of an interface to be used as the connection point for establishing SaMOG sessions.
ipv4_address must be an IPv4 address expressed in dotted-decimal notation.
Important To define more than one NAS IP address per context, in Global Configuration Mode, use the aaa
large-configuration command.
Important In this release, the configuration of the IPv6 bind address is supported as lab quality only.
Specifies the IPv4 or IPv6 address to be used as the connection point between the WLC and the SaMOG
gateway for the RADIUS interface. You can optionally bind a secondary IPv4 address (if the primary bind
address is an IPv6 address) or IPv6 address (if the primary bind address is an IPv4 address) to the MRME
service.
The second bind address can be bond in the same command or separate commands. When the second bind
address is provided, the MRME service restarts and existing sessions are lost for the other bind address.
ipv4_address must be an IPv4 address expressed in dotted-decimal notation.
ipv6_address must be an IPv6 address expressed in colon (or double-colon) notation.
auth-port auth_port_number
Specifies the authentication port number of the interface where authentication requests are received. The
system binds the default authentication port to 1812.
In addition to the authentication port, the accounting port and maximum subscriber limit can also be configured
optionally.
auth_port_number must be an integer from 1 through 65535.
acct-port acct_port_number
Specified the accounting port number of the interface where accounting requests are received. The system
binds the default accounting port to 1813.
In addition to the accounting port, the maximum subscriber limit can also be configured optionally.
acct_port_number must be an integer from 1 through 65535.
max-subscribers max_subscriber_number
Specifies the maximum number of subscriber sessions allowed.
max_subscriber_number must be an integer from 0 through 4,000,000.
Usage Guidelines Use this command to configure the IPv4 address to be used as the connection point for establishing SAMOG
sessions for handling authentication and accounting messages.
Examples Release 19 and later: The following command binds the MRME service with the IPv6 address of 192.168.1.254
and a secondary IPv6 address of 7777::101:1 with an accounting port number of 58 and maximum subscriber
limit of 1000.
bind ipv4-address 192.168.1.254 ipv6-address 7777::101:1 acct-port 58 max-subscribers 1000
Release 18 and earlier: The following command binds the service with an IP address of 196.10.2.3 with an
accounting port number of 58 and maximum subscriber limit of 1000.
bind address 196.10.2.3 acct-port 58 max-subscribers 1000
disconnect
This command allows you to specify the delay duration before which the call is disconnected.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
default
Configures this command to its default setting.
delay-time default: 30 seconds
wait-time default: 10 seconds
delay-time seconds
Specifies to configure the timer to retain the session on receiving an Accounting Stop, and for roaming
scenarios, session continuity on receiving an Accounting Start.
seconds must be an integer from 1 through 60.
wait-time seconds
Specifies to configure the timer to wait for accounting start message from the new WLC after processing the
accounting stop message from the old WLC.
seconds must be an integer of 10 through 300.
Usage Guidelines Specifies to configure the timer to wait for accounting stop message after triggering a Disconnect Req Message
to WLC for an SAMOG session.
Examples The following command sets the disconnect wait time to 60 seconds.
disconnect wait-time 60
dns-P-GW
This command allows you to configure the source context in which the DNS client is configured, or
enable/disable P-GW selection based on topology and load-balancing of P-GWs, based on weights from DNS.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
default
Returns the command to its default value.
default dns-pgw context: MRME will fetch the dns-client configuration from the current context.
default dns-pgw selection topology: MRME will perform P-GW selection based on the topology.
no
If previously configured, removes the dns-pgw configuration.
context context_name
Specifies to configure the source context in which the DNS client is configured.
context_name must be an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to configure the source context in which the DNS client is configured, or enable/disable
P-GW selection based on topology and load-balancing of P-GWs, based on weights from DNS.
In case of topology-based selection, when the DNS procedure outputs a list of P-GW host names for the APN
FQDN, MRME performs the longest suffix match and selects the P-GW which is topologically closest to the
MRME/subscriber. In case of weight-based selection, if there are multiple entries with the same priority in
the list of P-GW host names for the APN FQDN in the output from the DNS procedure, calls are distributed
to the P-GWs according to the weight field in RRs. The weight field specifies a relative weight for entries
with the same priority.
Examples This command will configure the source context in which the DNS client is configured to "mrmectx".
dns-P-GW context mrmectx
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
fqdn
This command allows you to configure the MRME fully qualified domain name (FQDN) to match the longest
suffix during dynamic allocation.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
default
Returns the command to the default setting of "null".
no
Removes the configured FQDN from the MRME service configuration.
fqdn_name
Specifies the MRME FQDN name that will be used for the longest suffix match during dynamic allocation.
fqdn_name must be an alphanumeric string of 1 to 255 characters.
Usage Guidelines Use this command to configure the MRME FQDN under MRME service to match the longest suffix during
dynamic allocation.
pgw-selection
This command provides P-GW selection related parameters for this MRME service.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
no
Removes the configuration.
local-configuration-preferred
Specifies the SaMOG Gateway to perform P-GW selection based on local configuration.
When this keyword is enabled,
• In the case of dynamic P-GW selection from the AAA server (APN FQDN based selection), the SaMOG
Gateway first tries to establish session with the locally configured P-GWs. If the locally configured
P-GWs are not reachable, APN FQDN resolution is performed, and SaMOG Gateway tries to establish
session with the resolved IP addresses.
• In the case of static P-GW selection from the AAA server (IP address or P-GW FQDN ), SaMOG tries
to establish session with the AAA server provided P-GW address (IP address or resolved P-GW FQDN).
If the AAA server provided P-GW addresses are not reachable, session setup fails.
fallback pgw-id
Specifies the SaMOG Gateway to trigger fall back to locally configured P-GW addresses (or DNS resolved
P-GW addresses using APN FQDN) when session establishment with AAA provided P-GW address or DNS
provided P-GW address for P-GW FQDN fails.
Usage Guidelines Use this command to enable SaMOG Gateway to perform P-GW selection based on local configuration.
When the local-configuration-preferred keyword is enabled, SaMOG first uses the locally configured P-GW
addresses to fall-back to. When the locally configured P-GW addresses are not reachable, SaMOG then uses
APN FQDN based P-GW address resolution.
When the local-configuration-preferred keyword is not enabled, SaMOG first uses APN FQDN based
P-GW address resolution to fall-back to. When the P-GW address resolved using APN FQDN is not reachable,
SaMOG then uses the locally configured P-GW addresses.
When session establishment with AAA provided P-GW address or DNS provided P-GW address for P-GW
FQDN fails, fall-back is triggered when the fallback pgw-id keyword is enabled.
Examples The following command enables the SaMOG Gateway to use locally configured P-GW addresses first for
P-GW resolution:
pgw-selection local-configuration-preferred
radius
This command allows you to specify the IP address and shared secret of the RADIUS accounting and
authentication client from which RADIUS accounting and authentication requests are received.
Important From release 16.0 onwards, this command has been deprecated. Instead, use the radius command described
under the TWAN Profile Configuration Mode Commands section.
Product SaMOG
Command Modes Exec > Global Configuration > Context Configuration > MRME Service Configuration
configure > context context_name > mrme-service mrme_service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-mrme-service)#
Syntax Description radius client ipv4_address[/mask] { [ encrypted ] key key [ [ disconnect-message [ dest-port port_number
] ] [ acct-onoff { [ aaa-context aaa_context_name ] [ aaa-group aaa_group_name ] [ clear-sessions ] } ] }
no
Removes a previously configured RADIUS client.
ipv4_address[/mask]
Specifies the IP address, and optional subnet mask of the RADIUS client from which RADIUS accounting
and authentication requests are received.
ipv4_address[/mask] must be an IPv4 address expressed in dotted-decimal notation.
Usage Guidelines Use this command to specify the IP address and shared secret of the RADIUS accounting and authentication
client from which RADIUS accounting and authentication requests are received.
Examples The following command configures the service to communicate with a RADIUS client with an IP address of
190.21.33.40 and an encrypted shared secret of key1234Ax3Z, and clear the session when accounting on/off
messages are received:
radius client 190.21.33.40 encrypted key 123 4Ax3Z acct-onoff clear-sessions
setup-timeout
This command is currently not supported in this release.
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
all-nsvc-failure-action
Configure how the SGSN handles the NSE when all NSVCs go down.
Product All
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
default
By default, the NSE is not cleared if all NSVCs go down.
clear-nse
Instructs the SGSN to SGSN to clear NSEs if all NSVCs to the BSC are down. This CLI clears the info only
in cases where all the NSVC of NSE go down due to ALIVE time out.
Usage Guidelines Enable the SGSN to clear NSE information when all NSVCs go down.
Examples Use the following command to configure the SGSN to clear NSEs when all NSVCs go down.
all-nsvc-failure-action clear-nse
associate
This command supports the association of DSCP template at network-service-entity ip local level.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
no
Removes the associate services needed for all IP nses.
dscp-template
Configures DCSP for all IP nses.
downlink [<template-name>]
Specifies the DSCP template used for the downlink packets.
template-name: String of size 1 up to 64.
Usage Guidelines Configuring this command allows the SGSN to send the configured DSCP value to:
• All the GPRS nodal messages.
• All the subscriber specific messages, when dscp template association at gprs-service level and nsei level
are absent.
• Atleast one nsvl should be configured, before configuring the DSCP marking at "network-service-entity
ip-local" level.
• After the removal of last nsvl, DSCP template association is removed from "network-service-entity
ip-local".
Examples The following example associates a DSCP template at network-service-entity ip local level.
associate dscp-template downlink DSCP
bssgp-timer
This command has been deprecated.
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the global configuration mode.
Product SGSN
max-ns-retransmissions
This command configures the maximum number of transmission retries counter.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
default
Resets the specified counter configuration to the default value.
alive count
Sets the maximum number of alive retries.
count: Must be an integer between 0 and 10. Default is 3.
sns-proc count
Sets the maximum number of retries for the SNS procedure
count: Must be an integer between 0 and 5. Default is 3.
ns-timer
This command sets the network service (NS) counters for the SNS procedure and testing.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
default
Resets the selected timer configuration to its default value.
sns-guard timeout_val
Sets the SNS-guard timer which is used in the auto-learn procedure to clean-up learnt BSC/NSE informtation.
Timeout value is in seconds.
timeout_val: Enter an integer from 1 to 300. Default is 60.
sns-prov timeout_val
Sets the SNS procedure timeout value in seconds.
timeout_val: Enter an integer from 1 to 10. Default is 5.
test timeout_val
Sets the test procedure timeout value in seconds.
timeout_val: Enter an integer from 1 to 60. Default is 30 seconds.
nsvc-failure-action
This command enables and disables the sending of an NS-STATUS message with cause 'ip-test fail' when
NSVC goes down.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
default
Resets the command configuration to its default value. The default action is not to send an NS-STATUS
message. This is applicable only to NSVCs that are auto-learned and not configured.
send-ns-status
Enables the sending of the NS-STATUS message.
Usage Guidelines Use this command to enable or disable sending an NS-STATUS messages when an NSVC goes down.
nsvl
This command creates and instance of a network service virtual link (NSVL) and enters the NSVL configuration
mode.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration
configure > network-service-entity ip-local
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local)#
no
Removes the identified NSVL definition from the configuration.
instance nsvl_id
Identifies a specific NSVL configuration instance.
nsvl_id: Must be an integer from 0 to 3.
Examples Enter the NSVL configuration sub-mode to modify the configuration for NSVL instance 2:
nsvl instance 2
peer-network-service-entity
This command has been replaced by the Network Service Entity - Peer NSEI Frame Relay configuration
mode.
retry-count
This command has been replaced by the max-ns-retransmissions command.
timer
This command has been replaced by the ns-timer command.
The Network Service Entity (NSE) - Peer NSEI configuration mode configures the Frame Relay parameters
for the peer NSE. This mode is a sub-mode of the Global Configuration mode. This sub-mode provides the
commands and parameters to define the management functionality for the Gb interface between a BSS and
an SGSN over a 2.5G GPRS Frame Relay network connection.
Command Modes Exec > Global Configuration > Network Service Entity - Frame Relay Peer NSEI Configuration
configure > network-service-entity peer-nsei nsei_number frame-relay
Entering the above command sequence results in the following prompt:
[local]host_name(nse-fr-peer-nsei-nse_id)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
bssgp-timer
This command has been deprecated.
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the previous configuration mode.
Product SGSN
ns-reset-mode
The command configures automatic NS-Reset for a specific Frame Relay peer NSE (network service entity).
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - Frame Relay Peer NSEI Configuration
configure > network-service-entity peer-nsei nsei_number frame-relay
Entering the above command sequence results in the following prompt:
[local]host_name(nse-fr-peer-nsei-nse_id)#
default
Resets the configuration to the passive mode.
active
Configures active mode so that the SGSN is enabled to initiate NS-Reset without manual intervention.
passive
Configures passive mode which means the SGSN continues not to initiate NS-Reset.
This is the default mode.
Usage Guidelines Use this command to configure the SGSN for active mode regarding the peer NSE, so that the SGSN will
initiate:
• NS-Reset when NSVC-DLCI binding is done.
• NS-Reset when the link goes down and then comes back.
• NS-Unblock upon receipt of NS-Reset-Ack message.
• if the NS layer can go down and the SGSN will mark the link as 'Blocked-Dead'. If the link comes up
later, the NS layer state for that link will remain in the Blocked state.
Examples Configure active mode to perform NS-Reset when the link goes down and comes back up:
ns-reset-mode active
ns-vc
This command creates a network service virtual circuit (NSVC) for this frame relay NSE and enters the
configuration sub-mode to define the NSVC parameters. These parameters are described in the NSVC
Configuration Mode chapter elsewhere in this CLI Reference Guide.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - Frame Relay Peer NSEI Configuration
configure > network-service-entity peer-nsei nsei_number frame-relay
Entering the above command sequence results in the following prompt:
[local]host_name(nse-fr-peer-nsei-nse_id)#
no
Removes the specified NSVC configuration.
id ns-vc_id
This keyword defines the NSVC configuration identifier.
ns-vc_id: Must be an integer from 0 to 65535
Examples Gain access to the NSVC configuration mode to change the 4th instance.
ns-vc id 4
The Network Service Virtual Connection (NSVC) configuration mode is a sub-mode of the Network Service
Entity (NSE) - Peer NSEI (for Frame Relay) configuration mode. The NSVC sub-mode creates a configuration
instance for a specific NSVC, within the Gb interface, between a BSS and an SGSN in a 2.5G GPRSFrame
Relay network connection.
Command Modes Exec > Global Configuration > Network Service Entity - Frame Relay Peer NSEI Configuration > NSVC
Configuration
configure > network-service-entity peer-nsei peer_nsei framerelay > ns-vc nsvc_id
Entering the above command sequence results in the following prompt:
[local]host_name(nse-fr-peer-nsei-nse_id-nsvci-nsvc_instance)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the previous configuration mode.
Product SGSN
Usage Guidelines Returns to the network service entity - frame relay configuration mode.
The Network Service Virtual Link configuration mode is a sub-mode of the Network Service Entity - IP
configuration mode. This sub-mode provides the commands and parameters to define the NSVL of the Gb
interface between a BSS and an SGSN in a 2.5G GPRS IP network connection.
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration > NSVL Configuration
configure > network-service-entity ip > nsvl instance nsvl_id
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local-nsvl-nsvl_instance)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the current configuration mode and returns to the previous configuration mode.
Product SGSN
Usage Guidelines Return to the network service entity - IP configuration mode. mode.
nsvl-address
This command configures the IP address of the NSVL. end-point.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration > NSVL Configuration
configure > network-service-entity ip > nsvl instance nsvl_id
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local-nsvl-nsvl_instance)#
ip-address ip_address
Identifies the address of the NSVL.
ip_address: Must be specified using the standard IPv4 dotted decimal notation or colon notation for IPv6.
context ctxt_name
Identifies the specific context associated with this NSVL address.
ctxt_name: Enter up to 79 alphanumeric characters.
port port_num
Specifies the UDP port to associate with the NSVL end-point.
port_num: Must be an integer from 1 to 65535.
Usage Guidelines Use this command to configure the IP address, context name and port number for the NSVL end-point.
weight
This command configures the signaling or data weight for NSVL.
Product SGSN
Command Modes Exec > Global Configuration > Network Service Entity - IP Configuration > NSVL Configuration
configure > network-service-entity ip > nsvl instance nsvl_id
Entering the above command sequence results in the following prompt:
[local]host_name(nse-ip-local-nsvl-nsvl_instance)#
data data_weight
Defines the data weight for the NSVL.
data_weight: Must be an integer from 0 to 255. Default is 1.
signaling signaling_weight
Defines the signaling weight for the NSVL.
signaling_weight: Must be an integer from 0 to 255. Default is 1.
Usage Guidelines Configure the weight of the signaling or data for the NSVL.
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
enable
Enables the use of the Network Time Protocol (NTP) for synchronizing the system clock. By default, NTP
is not enabled externally and should be configured when the system is initially installed. When enabled, the
active ASR 5000 SMC or ASR 5500 MIO will synchronize with external sources. If not enabled, the active
SMC or MIO will use its local clock as a time source. In the event of an NTP server or network outage, an
already running SMC or MIO will continue to use NTP to maintain time accuracy, but in a holdover mode.
Product All
context
Default: local
Specifies the context for which NTP is to be enabled as an alphanumeric string of 1 through 79 characters.
Important NTP must be configured for use in the local context only. Use of other contexts will cause issues.
Usage Guidelines Sets the NTP server to be used for the system. Only one NTP server may be active at any given time.
The system uses NTP to synchronize internal clocks on the chassis to external time sources (typically GPS
NTP sources, or other Stratum 2 or 3 servers, switches or routers).
All cards with CPUs synchronize to the active SMC or MIO internally. This occurs even if an external NTP
server is not configured. In the event of a SMC or MIO switchover, all other cards will start synchronizing
with the newly active SMC or MIO automatically.
If any NTP server is enabled, the chassis system clock will be synchronized to the active NTP server which
covers all contexts for timing synchronization.
Refer to the System Administration Guide for additional information on configuring NTP.
Examples The following command enables use of NTP for the local context.
enable
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
server
Configures a Network TIme Protocol (NTP) server for use by the local NTP client in synchronizing the system
clock.
Product All
Syntax Description server ip_address [ prefer ] [ version number ] [ minpoll poll_period ] [ maxpoll poll_period ]
no server ip_address
no
Indicates the server specified is to be removed from the list of NTP servers for clock synchronization.
ip_address
Specifies the IP address of the NTP server to be used for clock synchronization in IPv4 dotted-decimal or
IPv6 colon-separated-hexadecimal notation.
NTP should be configured for at least three external NTP servers. With three or more servers, outlyers and
broken or misconfigured servers can be detected and excluded. Generally, the more servers the better (within
reason).
prefer
Indicates the NTP server specified to be the preferred server. Only one server can be set to preferred. The
preferred server is the first one contacted for clock synchronization.
Important Do not change the version, minpoll or maxpoll keyword settings unless instructed to do so by Cisco
TAC.
version number
Specifies the network timing protocol version to use for server communications as an integer from 1 to 4.
Default: 4 (RFC 5905)
minpoll poll_period
Specifies the minimum polling interval (in seconds) for NTP messages as a power of 2. poll_period is the
exponent (power of) expressed as an integer from 6 through 17. For example, if you specify the number 6,
the value is 2^6 and the resultant poll period is 64 seconds. Default: 6
maxpoll poll_period
Specifies the maximum polling interval (in seconds) for NTP messages as a power of 2. poll_period is the
exponent (power of) expressed as an integer from 6 through 17. For example, if you specify the number 10,
the value is 2^10 and the resultant poll period is 1024 seconds. Default: 10
Important Adding, removing, or modifying an NTP server configuration entry causes the NTP client to restart itself
and resynchronize with all configured NTP servers.
Examples The following command adds the NTP server with address 10.2.3.4 to the list of NTP servers.
server 10.2.3.4
This mode is used to configure a pool of IP addresses associated with a pool ID and pool type (either MME
or S4-SGSN) for Network Triggered Service Restoration (NTSR).
Command Modes Exec > Global Configuration > NTSR Pool Configuration
configure > ntsr pool pool-id id > pool-type type
Entering the above command sequence results in the following prompt:
[local]host_name(config-ntsr-pool)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
peer-ip-address
Configures a pool of IP addresses associated with a pool ID and pool type (either MME or S4-SGSN) for
Network Triggered Service Restoration (NTSR).
Product S-GW
Command Modes Exec > Global Configuration > NTSR Pool Configuration
configure > ntsr pool pool-id id > pool-type type
Entering the above command sequence results in the following prompt:
[local]host_name(config-ntsr-pool)#
no
Removes the specified IP address configuration.
ipv4 ipv4_address
Specifies an IPv4 address to use with an associated NTSR pool ID and pool type (either MME or S4-SGSN).
ipv6 ipv6_address
Specifies an IPv6 address to use with an associated NTSR pool ID and pool type (either MME or S4-SGSN).
Usage Guidelines Use this command to configure a pool of IP addresses associated with a pool ID and pool type (either MME
or S4-SGSN) for Network Triggered Service Restoration (NTSR).
Before using this command, operators must configure an NTSR pool ID and pool type by executing the ntsr
pool command in Global Configuration Mode
Examples To configure a an IPv4 address associated with a pool ID and pool type (either MME or S4-SGSN) for Network
Triggered Service Restoration (NTSR).
peer-ip-address ipv4-address 1.1.1.1
Using the Operator Policy feature allows the operator to fine-tune any desired restrictions or limitations
needed to control call handling per subscriber or for a group of callers across IMSI ranges.
Operator Policy configuration mode associates APNs, APN profiles, IMEI ranges, IMEI profiles, an APN
remap table and a call control profile to an operator policy. These profiles and tables are created and defined
within their own configuration modes to generate sets of rules and instructions that can be reused and assigned
to multiple policies.
Command Modes Exec > Global Configuration > Operator Policy Configuration
configure > operator-policy policy_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-opr-policy-policy_name)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
apn
This command identifies an APN (access point name) and associates it with an APN profile (created separately
in the APN Profile Configuration mode).
Product MME
SAEGW
SaMOG
S-GW
SGSN
Command Modes Exec > Global Configuration > Operator Policy Configuration
configure > operator-policy policy_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-opr-policy-policy_name)#
no
Disables the specified APN to APN Profile correspondence.
default-apn-profile apn_profile_name
Enables the use of a default APN profile comprised of default values for all parameters. this profile will be
used when none of the configured APNs match the APN in the incoming Request.
apn_profile_name must be an alphanumeric string of 1 through 64 characters.
apn-profile apn_profile_name
apn_profile_name must be an alphanumeric string of 1 through 64 characters.
operator-identifier apn_op_id
Links the specified APN operator ID with the specified APN profile.
apn_op_id: must be an alphanumeric string of 1 through 18 characters, including dots (.). The string must be
in the following format, where # represents a digit: MNC###.MCC###.GPRS.
webauth-apn-profile apn_webauth_name
Specify the APN profile to be used for SaMOG web authorization.
apn_webauth_name must be an alphanumeric string of 1 through 64 characters.
Important The SaMOG Web Authorization feature is license dependent. Contact your Cisco account representative
for more information on license requirements.
Usage Guidelines Use this command, to associate APNs with APN profiles. This command can be repeated to associate multiple
APNs with profiles.
Examples Associate the APN profile named apnprof1 to APN network ID starflash.com:
apn apnprof1 network-identifier starflash.com
Associate congestion control with APN network ID starflash.com:
apn network-identifier starflash.com congestion-control
associate
Associate an APN remap table and a call control profile with the operator policy.
Product MME
SAEGW
S-GW
SGSN
Command Modes Exec > Global Configuration > Operator Policy Configuration
configure > operator-policy policy_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-opr-policy-policy_name)#
no
Removes the association definition from the policy configuration.
apn-remap-table table_id
Identifies the APN remap table to be associated with the operator policy.
table_id must be an alphanumeric string of 1 through 65 characters.
call-control-profile profile_id
Identifies a call control profile to be associated with the operator policy.
profile_id must be an alphanumeric string of 1 through 64 characters.
Usage Guidelines Use this command to associate an APN remap table and/or a call control profile with this Operator Policy.
The APN remap table and the call control profile contain the definitions that instruct the SGSN or MME how
to handle calls. Only one of each of these can be associated with an operator policy.
Examples Associate the stardust.net_APNremap1 APN remap table with this operator policy:
associate apn-remap-table stardust.net_APNremap1
description
Associates a description with or names an operator policy.
Product MME
SAEGW
S-GW
SGSN
Command Modes Exec > Global Configuration > Operator Policy Configuration
configure > operator-policy policy_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-opr-policy-policy_name)#
description
Enter an alphanumeric string of 1 through 100 characters. If the string includes spaces, punctuation, and
case-sensitive letters, it must be bracketed with double quotation marks (" ").
no
Removes the existing description from this operator policy.
Usage Guidelines Identity this particularly operator policy using descriptive text.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
imei
Defines a range of IMEI (International Mobile Equipment Identity) numbers and associates an IMEI profile
with the range definition.
Product SGSN
Command Modes Exec > Global Configuration > Operator Policy Configuration
configure > operator-policy policy_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-opr-policy-policy_name)#
Syntax Description imei range IMEI_number to IMEI_number { imei-profile profile_name | sv ## imei-profile profile_name }
no imei range IMEI_number to IMEI_number sv ##
no
Removes the IMEI definition from the policy configuration.
sv ##
Identifies the software version to fine-tune the IMEI definition. This keyword should only be included if the
IMEISV is retrievable.
## must be a 2-digit integer.
imei-profile profile_name
Identify the IMEI profile that defines the actions appropriate to the devices identified within the specified
range.
profile_name must be an alphanumeric string of 1 through 64 characters.
Usage Guidelines This command defines the IMEI ranges that will be used by the operator policy to determine if the device is
appropriately selected for actions defined in the specified IMEI profile.
Examples All devices with an IMEI of 123123* requesting Attach shall be subject to actions in the blacklist_profile1
imei range 1231230 to 1231239 imei-profile name blacklist_profile1
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
activate client id
Activates/deactivates a Common Object Request Broker Architecture (CORBA) client for the ORBEM
interface.
Product All
no
Deactivates the specified client
id name
Specifies the client to be activated. name must refer to a previously configured CORBA client expressed as
an alphanumeric string of 1 through 10 characters.
Usage Guidelines Activates CORBA clients after they have been configured or deactivated by the system or by configuration.
client id
Configures or removes a CORBA client from the ORBEM interface.
Product All
no
Removes the specified client from the configuration.
id name
Specifies the client to be configured. name must be an alphanumeric string of 1 through 10 characters.
encrypted password
Specifies the use of an encrypted password for use by the chassis while saving configuration scripts. Signifies
that ORBEM messages are transported using SSL encryption techniques. StarOS displays the encrypted
keyword in the configuration file as a flag that the variable following the password keyword is the encrypted
version of the plain text password. Only the encrypted password is saved as part of the configuration file.
password
Specifies the plain text password for the CORBA client. pwd must be an alphanumeric string of 1 through 35
characters.
pwd
Specifies the password for the CORBA client.
For an encrypted password, pwd must be an alphanumeric string of 1 through 212 characters.
For an unencrypted password, pwd must be an alphanumeric string of 1 through 35 characters.
Usage Guidelines Use this command to configure or remove a CORBA client from the ORBEM interface.
CORBA clients must be configured prior to being activated.
Examples The following command sets a plain text password for CORBA client ems:
client id ems password ems1001
end
Exits the current configuration mode and returns to the Exec mode.
Product All
event-notif-iiop-port
Configures the port number for Internet inter-ORB event notifications.
Product All
default
Restores the port number for the inter-ORB event notifications to the system default: 7778.
number
Specifies the port number to use as an integer from 1 through 65535. Default: 7778
Usage Guidelines Explicitly set the port number when the default port number is not the desired port value for integrating
multiple products together for standardized inter-ORB communications.
Event notification port configured is only used if the Internet inter-ORB transport is enabled via the
iiop-transport command with the event notification service being enabled as well.
Examples The following command sets the IIOP port number to 5466:
event-notif-iiop-port 5466
event-notif-service
Enables or disables the ORB Notification Service and allows the configuration of filters dictating which event
notifications are sent.
Product All
Syntax Description [ no ] event-notif-service [ filter { event-id event_id [ to final_event_id ] | facility event_facility level
event_level } ]
default event-notif-service filter
default
Restores the ORB Notification Service filter to its default behavior of sending all "error" level and higher
events, and "info" level events for the orbs facility, CLI command logs, and license change logs.
no
Disables the event notification service.
filter
Specifies a filter that determines for which events the system sends notifications.
Important In Release 20.0, HNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HeNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
Important In Release 20.0, HNBGW is not supported. This keyword must not be used in Release
20.0. For more information, contact your Cisco account representative.
event_level
specifies the severity level of the event notification to filter and can be configured to one of the following:
• critical: display critical events
• error: display error events and all events with a higher severity level
• warning: display warning events and all events with a higher severity level
• unusual: display unusual events and all events with a higher severity level
• info: display info events and all events with a higher severity level
• trace: display trace events and all events with a higher severity level
• debug: display all events
Usage Guidelines This command is used to enable or disable the ORB Notification Service. Additionally, it can be used to
configure filters dictating which events are sent. This service is disabled by default.
Filters can be configured for a specific event identification number (event ID), a range of event IDs, or specific
severity levels for events for particular facilities.
When no filters are configured and the service is enabled, the ORB Notification Service sends all "error" level
and higher events, and "info" level events for the orbs facility, CLI command logs, and license change logs.
Multiple instance of this command can be executed to configure multiple filters.
event-notif-siop-port
Configures the port to use for secure socket layer (SSL) inter-ORB event communication.
Product All
default
Restores the port to use for secure socket layer inter-ORB event communication to the system default: 7777.
number
Specifies the port number to use as an integer from 1 through 65535. Default: 7777
Usage Guidelines Explicitly set the port number when the default port number is not the desired port value for integrating
multiple products together for inter-ORB communications using SSL.
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
iiop-port
Configures the port number for Internet Inter-ORB Protocol (IIOP) communications.
Product All
default
Restores the port number for inter-ORB communications to the system default: 14132.
no
Disables the IIOP port.
number
Specifies the port number to use as an integer from 1 through 65535. Default: 14132
Usage Guidelines Explicitly set the port number when the default port number is not the desired port value for integrating
multiple products together for standardized inter-ORB communications.
Internet inter-ORB port is only used if IIOP transport is enabled via the iiop-transport command.
Examples The following commands sets the IIOP port number to 2546:
iiop-port 2546
iiop-transport
Enables/disables use of the Internet Inter-ORB Protocol (IIOP) for management across the network.
Product All
no
Disables internet inter-ORB protocol communication across the network.
Usage Guidelines Enables the transport of IIOP messages to support remote management across the network.
The default is IIOP transport disabled.
Examples The following command enables ORB-based management across the network:
iiop-transport
iop-address
Sets the IP address used by the ORBEM Server to advertise service.
Product All
default
Restores the IP address for inter-ORB communications to the system default: IP address of the current context.
ip_address
Specifies the IP address to use for inter-ORB communications using IPv4 dotted-decimal notation.
Usage Guidelines Change the inter-ORB IP address when the IP address of the current context should not be used. The IP address
of the local context may not be appropriate when the ORB configuration across nodes would cause conflicts
with the IP addresses.
The default inter-ORB IP address is the IP address of the current context.
Examples The following command sets the inter-ORB IPv4 address to 10.2.3.4:
iop-address 10.2.3.4
max-attempt
Configures the maximum number of failed login attempts after which the client is deactivated.
Product All
default
Restores the maximum number of failed login attempts before which the client is deactivated to the system
default: 3 attempts.
count
Specifies the number of failed login attempts prior to deactivating a client. The value must be an integer from
1 through 10. Default: 3 attempts
Usage Guidelines Adjust the maximum number of attempts to a smaller value to increase the security level of the system.
session-timeout
Configures the amount of idle time (no activity) before a client session is terminated.
Product All
default
Restores the amount of idle time (no activity) before a session is terminated to the system default: 300 seconds.
seconds
Specifies the number of seconds of idle time before a client session is terminated. The value must be must be
an integer from 1 through 86400. Default: 300 seconds
Usage Guidelines Reduce the session timeout when the maximum number of sessions allowed is frequently being reached.
Setting this to a lower value will help release idle sessions faster to allow use by other clients.
Examples The following sets the session timeout value to 150 seconds:
session-timeout 150
siop-port
Configures the SSL I/O port for inter-ORB events.
Product All
default
Restores the secure socket layer I/O port for inter-ORB events to the system default: 14131.
default
Restores the secure socket layer I/O port for inter-ORB events to the system default: 14131.
number
Specifies the port number to use as an integer from 1 through 65535. Default: 14131
Usage Guidelines Explicitly set the port number when the default port number is not the desired port value for integrating
multiple products together for inter-ORB communications.
Examples The following command sets the SIOP port number to 2466:
siop-port 2466
ssl-auth-policy
Configures the SSL peer authentication policy used by the ORBEM server.
Product All
auth-none
Specifies that the ORBEM server does not authenticate the peer. This is the default setting.
auth-once
Specifies that the ORBEM server authenticates the peer once (no fail).
auth-once-fail
Specifies that the ORBEM server authenticates the peer once (fail if no certificate).
auth-peer
Specifies that the ORBEM server authenticates the peer every time (no fail).
auth-peer-fail
Specifies that the ORBEM server authenticates the peer every time (fail if no certificate).
Usage Guidelines Use to configure the peer authentication policy used by the SSL transport of ORBEM.
Examples The following command sets the policy to authenticate the peer once without failure.
ssl-auth-policy auth-once
ssl-certificate
Defines the certificate to be used by the SSL transport of ORBEM.
Product All
string certificate
Specifies an ORBEM SSL certificate. certificate is an alphanumeric string of up to 4096 characters.
file url
Default: /usr/ssl/certs/orbscert.pem
Specifies an ORBEM SSL certificate file and location. url is an alphanumeric string of up to 1024 characters.
Usage Guidelines Use to configure the certificate to be used by the SSL transport of ORBEM. Note that if the file option is used,
the certificate content is read from the url and converted into a quoted string.
Examples The following command defines the certificate cert3.pem file as being located in the /usr/ssl/certs directory:
ssl-certificate file /usr/ssl/certs/cert3.pem
The following command defines the certificate string (the string shown is abbreviated):
ssl-certificate string
"-----BEGIN CERTIFICATE-----\n\
MIIELDCCA5WgAwIBAgIBATANBgkqhkiG9w0BAQQFADCBsTELMAkGA1UEBhMCVVMx\n\
FjAUBgNVBAgTDU1hc3NhY2h1c2V0dHMxEjAQBgNVBAcTCVRld2tzYnVyeTEeMBwG\n\
A1UEChMVU3RhcmVudCBOZXR3b3JrcyBJbmMuMSIwIAYDVQQLExlFbGVtZW50IE1h\n\
bmFnZW1lbnQgU3lzdGVtMQ4wDAYDVQQDEwVPUkJFTTEiMCAGCSqGSIb3DQEJARYT\n\
b3JiZW1AbnVsaW5raW5jLmNvbTAeFw0wMjA5MDYxMjE5MTNaFw0yMjA5MDExMjE5\n\
MTNaMIGxMQswCQYDVQQGEwJVUzEWMBQGA1UECBMNTWFzc2FjaHVzZXR0czESMBAG\n\
A1UdDgQWBBSpuGGMTwgaq8H+e70ZPIFHVZjiWDCB3gYDVR0jBIHWMIHTgBRkVBzy\n\
4zW5Gv0pXcwT07PtzCm53qGBt6SBtDCBsTELMAkGA1UEBhMCVVMxFjAUBgNVBAgT\n\
DU1hc3NhY2h1c2V0dHMxEjAQBgNVBAcTCVRld2tzYnVyeTEeMBwGA1UEChMVU3Rh\n\
cmVudCBOZXR3b3JrcyBJbmMuMSIwIAYDVQQLExlFbGVtZW50IE1hbmFnZW1lbnQg\n\
U3lzdGVtMQ4wDAYDVQQDEwVPUkJFTTEiMCAGCSqGSIb3DQEJARYTb3JiZW1AbnVs\n\
aW5raW5jLmNvbYIBADANBgkqhkiG9w0BAQQFAAOBgQATOdeDWikcoUIU8Gth9wr4\n\
Z5Fi8akXHhKhN7UMKyiW/Nn5NyfqPIA+9JwYMqwVOG8ybtfBQIGRCQodbXUm6Z9Z\n\
cM3XxWKVKHVolGS83f/JfpSLnuGkBIW8m3p/snHBH2BtgNT8OLItlTdBHedTKL72\n\
ZIxGF9/ok9hUqU4ikzQcEQ==\n\
-----END CERTIFICATE-----\n"
ssl-private-key
Configures the SSL private key used by the ORBEM server.
Product All
string key
Specifies an ORBEM SSL private key. key is an alphanumeric string of up to 4096 characters.
file url
Default: /usr/ssl/certs/orbscert.pem
Specifies the ORBEM SSL private key file location. url is an alphanumeric string of up to 1024 characters.
Usage Guidelines Use to configure the private key for the SSL transport of ORBEM. Note that if file option is used, the private
key is read from the url and converted into a quoted string.
Examples The following command defines the private-key cert3.pem file as being located in the /usr/ssl/certs directory:
ssl-private-key file /usr/ssl/certs/cert3.pem
The following command defines the private-key string (the string shown is abbreviated):
ssl-private-key string
"-----BEGIN RSA PRIVATE KEY-----\n\
MIICXQIBAAKBgQC6Dh79iaK/zZG/Kwme2XS6G8/n3/+sac6huxI1WNyammyYZKZp\n\
XTjHUlS92fvn0UUM4tFjN4XoqveSiqy3IqUhnVKS3+0L7s9beanQUJuR9MdLy9Ho\n\
7qh720wpN4isqN7YfGLoqGslLQjhS8z6ZT0ZUhyusY0rE6yHTV23nHKNtQIDAQAB\n\
9br1iVWvy/N23WXwZIiH+e1tBfHqlSd/0wJBANEEOgH/vJse/YdHeYjlT76IcGRp\n\
Tq6ldBXdoLRDGUF2AqdboJ7wWCOJQO34XbBtmWFfTkqz48Mi6uh3/5kDfH8CQGAl\n\
XObwPFRztvkXprZfh7IekxAIuoHiT1JsEKSIGPzEqDY2rmoWDghOvPETO+5zWEQk\n\
TXzLaRHgbIy9MKnXSt8CQQCcBfT7VndEfG9VWyPzeL4vx4ZhUMZQ6FIJdXo7Xq9x\n\
mzX8hgIcfdg3tahlNt35gL/DjUY7d14+MgLrRf3Udbk9\n\
-----END RSA PRIVATE KEY-----\n"
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
area authentication
Enables authentication for the specified OSPF area.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables authentication for the specified area.
ip_address
Specifies the IP address of the area where authentication will be enabled in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the area where authentication will be enabled. This must be an integer
from 0 through 4294967295.
authentication
Sets the OSPF authentication type to use the simple authentication method.
message-digest
Sets the OSPF authentication type to use the message digest 5 (MD5) authentication method.
Examples The following command enables authentication for an OSPF area defined by the IP address 192.168.100.10
and the OSPF authentication type to MD5:
area 192.168.100.10 authentication message-digest
area default-cost
Configures the default cost for an area.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Deletes the default cost for the area.
ip_address
Specifies the IP address of the area in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the area as an integer from 0 through 4294967295.
cost_value
Sets the default cost to be configured for the specified area as an integer from 0 through 16777215.
Usage Guidelines Use this command to configure the default cost for an OSPF area.
Examples The following command sets the default cost for an OSPF area defined by the IP address 192.168.100.10 to
300:
area 192.168.100.10 default-cost 300
area nssa
Defines an area as an NSSA (Not So Stubby Area) and configures OSPF parameters for it.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
ip_address
Specifies the IP address of the NSSA in IPv4 dotted-decimal notation.
decimal-value
Specifies the identification number of the NSSA area as an integer from 0 through 4294967295.
default-information-originate
Originates default information to the NSSA area.
no-redistribution
Does not redistribute external routes to the NSSA area.
no-summary
Does not inject inter-area routes into NSSA.
translate-always
Configures the NSSA-ABR (Area Border Router) to always translate
translate-candidate
Configure NSSA-ABR for translate election. (This is enabled by default.)
translate-never
Configure NSSA-ABR to never translate.
Examples The following command defines the area designated by the IP address 192.168.100.10 as an NSSA area:
area 192.168.100.10 nssa
area stub
Defines an area as an OSPF stub area.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
ip_address
Specifies the IP address of the stub area in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the stub area as an integer from 0 through 4294967295.
stub
Specifies this is a stub area.
no-summary
Disables (stops) the ABR (Area Border Router) from sending summary link state advertisements (LSAs) into
the stub area.
Usage Guidelines Use this command to define an OPSF area as a stub area.
Examples The following command defines the OSPF area defined by the IP address 192.168.100.10 as a stub area:
area 192.168.100.10 stub
area virtual-link
Configures a virtual link between an area that cannot be physically connected to the network backbone and
an area that is physically connected to the network backbone.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables area virtual-link.
ip_address
Specifies the IP address of the transit area in IPv4 dotted-decimal notation.
decimal_value
Specifies The identification number of the transit area as an integer from 0 through 4294967295.
router_id_address
Specifies the router id of the ABR to be linked to in IPv4 dotted-decimal notation.
Usage Guidelines Use this command to create a virtual link between an area that is connected to the network backbone and an
area that cannot be connected to the network backbone.
Examples The following command creates a virtual link between the OSPF areas defined by the IP address 192.168.100.10
and the IP address 192.168.200.20:
area 192.168.100.10 virtual-link 192.168.200.20
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables area virtual link authentication.
ip_address
Specifies the IP address of the transit ares in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the transit area as an integer from 0 through 4294967295.
router_id_address
Specifies the router id of the ABR to be linked to in IPv4 dotted-decimal notation.
authentication
Sets the OSPF authentication type to use the simple authentication method.
message-digest
Sets the OSPF authentication type to use the message digest (MD) authentication method.
null
Set the OSPF authentication type to use no authentication, thus disabling either MD or clear text methods.
text
Set the OSPF authentication type to use the clear text authentication method.
Usage Guidelines Use this command to set the authentication method for a virtual link between an area that is connected to the
network backbone and an area that cannot be connected to the network backbone.
Examples The following command sets the authentication method for a virtual link between the OSPF areas defined by
the IP address 192.168.100.10 and the IP address 192.168.200.20 to use no authentication:
area 192.168.100.10 virtual-link 192.168.200.2 null
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the area virtual link authentication key.
ipaddress
Specifies the IP address of the transit area in IPv4 dotted-decimal notation.
decimal-value
Specifies the identification number of the transit area as an integer from 0 through 4294967295.
router_id_address
Specifies the router id of the ABR to be linked to in IPv4 dotted-decimal notation.
encrypted password
encrypted_key is an alphanumeric string of 1 through 523 characters.
Use this if you are pasting a previously encrypted authentication key into the CLI command.
password password_key
The password to use for authentication. password_key is an alphanumeric string of 1 through 16 characters
that denotes the authentication password. This variable is entered in clear text format.
Usage Guidelines Use this command to specify the authentication password for a virtual link between an area that is connected
to the network backbone and an area that cannot be connected to the network backbone.
Examples The following command creates an authentication password of 123456 for a virtual link between the OSPF
areas defined by the IP address 192.168.100.10 and the IP address 192.168.200.20:
area 192.168.100.10 virtual-link 192.168.200.20 authentication-key password 123456
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the area virtual link intervals.
ipaddress
Specifies the IP address of the area in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the transit area as an integer from 0 through 4294967295.
router_id_address
Specifies the router id of the ABR to be linked to in IPv4 dotted-decimal notation.
dead-interval dead_value
Specifies The interval (in seconds) that the router should wait, during which time no packets are received and
after the router considers a neighboring router to be off-line. dead_value must be an integer from 1 through
65535.
hello-interval hello_value
Specifies the interval (in seconds) before sending a hello packet. hello_value must be an integer from 1 through
65535.
retransmit-interval rt_value
Specifies the interval (in seconds) that router should wait before retransmitting a packet. rt_value must be an
integer from 1 through 3600.
transmit-delay td_value
Specifies the interval (in seconds) that the router should wait before transmitting a packet. td_value must be
an integer from 1 through 3600.
Usage Guidelines Use this command to set the intervals or delay types for a virtual link between an area that is connected to the
network backbone and an area that cannot be connected to the network backbone.
Examples The following command sets the retransmit interval for a virtual link between the OSPF areas defined by the
IP address 192.168.100.10 and the IP address 192.168.200.20 to 60 seconds:
area 192.168.100.10 virtual-link 192.168.200.20 retransmit-interval 60
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
Syntax Description area { ip_address | decimal_value } virtual-link router_id_address message-digest-key key_id md5
{ encrypted password encrypted__key | password password_key }
no area { ipaddress | decimal_value } virtual-link router_id_address message-digest-key key_id
no
Disables the area virtual link message digest key.
ip_address
Specifies the IP address of the transit area in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the transit area as an integer from 0 through 4294967295.
router_id_address
Specifies the router id of the ABR to be linked to in IPV4 dotted-decimal notation.
message-digest-key key_id
Specifies the key identifier number. key_id must be an integer from 1 through 255.
password password_key
Specifies the password to use for authentication. password_key is an alphanumeric string from 1 through 16
characters that is entered in clear text format.
Usage Guidelines Use this command to enable the use of MD5-based OSPF authentication for a virtual link between an area
that is connected to the network backbone and an area that cannot be connected to the network backbone.
Examples The following command enables the use of MD5-based OSPF authentication for a virtual link between the
OSPF areas defined by the IP address 192.168.100.10 and the IP address 192.168.200.20, sets the MD5 Key
ID to 25, and the password to 123456:
area 192.168.100.10 virtual-link 192.168.200.20 message-digest-key 25 md5 password 123456
bfd-all-interfaces
Enables or disables Bidirectional Forwarding Detection (BFD) on all OSPF interfaces.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables BFD capability on all interfaces.
Usage Guidelines Use this command to configure BFD on all OSPF interfaces. See the System Administration Guide for additional
information on how to configure BFD.
capability graceful-restart
Configures graceful-restart. By default, this capability is set to enabled.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the graceful-restart capability.
default
Enables the graceful-restart capability if it has been disabled.
default-information originate
Creates a default external route into an OSPF routing domain.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the default external route.
always
Always advertise the route regardless of whether or not the software has a default route.
metric metric_value
Sets the OSPF metric used in creating the default rout as an integer from 1 through 16777214.
metric-type { 1 | 2 }
Sets the default route metric type.
1: Sets the OSPF external link type for default routes to Type 1.
2: Sets the OSPF external link type for default routes to Type 2.
route-map route_map_name
Specifies the name of the default route-map to be use as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to set the default external route into an OSPF routing domain.
Examples The following command sets the default external route to originate from the route map named rmap1:
default-information originate route-map rmap1
default-metric
Configures the default metric value for the OSPF routing protocol. All OSPF interfaces have a cost, which is
a routing metric that is used in the link-state calculation. Routes with lower total path metrics are preferred
over those with higher path metrics. When several equal-cost routes to a destination exist, traffic is distributed
equally among them. The default metric is a global parameter that specifies the cost applied to all OSPF routes
by default.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
metric-value
Sets the metric value expressed as an integer from 1 through 16777214. Default: 26385.
no
Enables or disables the default metric value for OSPF.
Usage Guidelines Use this command to set the default metric for routes.
distance
Configures the OSPF route administrative distances for all OSPF route types or based on specific route type.
Administrative distance is the measure used by Cisco routers to select the best path when there are two or
more different routes to the same destination from two different routing protocols. Administrative distance
defines the reliability of a routing protocol. Each routing protocol is prioritized in order of most to least reliable
(believable) using an administrative distance value. A lower numerical value is preferred.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the OSPF route administrative distances for all OSPF route types.
distance_value
Specifies the OSPF route administrative distances as an integer from 1 to 255. The default distance value is
110.
intra-area distance_value: sets the OSPF route administrative distance for all routes within an area. This
must be an integer from 1 through 255. The default is 110.
no
Enables or disables the specified option.
Usage Guidelines Use this command to set the administrative distance for OSPF routes.
Examples The following command sets the administrative distance for all OSPF route types to 30:
distance 30
distribute-list
Enables or disables the filtering of networks in outgoing routing updates.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the filtering of networks in outgoing routing updates.
route_access_list
Specifies the name of the OSPF route access list to use for filtering as an alphanumeric string of 1 through
63 characters.
connected
Filters connected routes.
rip
Filters RIP routes. (RIP is not supported at this time.)
static
Filters static routes.
no
Disables the specified option.
Usage Guidelines Use this command to enable the filtering of outgoing route updates by using the specified route access list.
Examples The following command uses the route access list named ral1 to filter outgoing routing updates for all connected
routes:
distribute-list ral1 out connected
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip vrf
Configures the Virtual Routing and Forwarding (VRF) instances for OSPF routing protocol.
Product All
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the VRF instances and removes the configured VRF context association for OSPF routing.
vrfvrf_name
Configures Virtual Routing & Forwarding (VRF) parameters.
vrf_name is name of a preconfigured VRF context configured in Context Configuration Mode via the ip vrf
command. It is an alphanumeric string of 1 through 63 characters.
Usage Guidelines Use this command to configure the IP VRF forwarding also to associate the preconfigured VRF context with
the specific tunnel interface.
This command creates and enters the OSPF VRF Configuration Mode if required to configure the VRF context
instances for OSPF routing.
Examples The following command enables preconfigured VRF context instance ospf_vrf1 for OSPF routing and enters
the OSPF VRF Configuration mode:
ip vrf ospf_vrf1
neighbor
Configures OSPF routers that interconnect to non-broadcast networks.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables OSPF routers that interconnect to non-broadcast networks.
ip_address
Specifies the interface IP address of the OSPF neighbor expressed using IPv4 dotted-decimal notation.
poll-interval poll_interval_value
Default: 120
Sets the number of seconds in the dead neighbor polling interval as an integer from 1 through 65535
priority priority_value
Default: 0
Sets the 8-bit number that represents the router priority value of the non-broadcast neighbor associated with
the specified IP address. This must be an integer from 0 through 255. This keyword does not apply to
point-to-multipoint interfaces.
Usage Guidelines Use this command to configure OSPF routers that connect to non-broadcast networks.
Examples The following command specifies an OSPF router neighbor with the IP address of 192.168.100.10:
neighbor 192.168.100.10
network area
Enables OSPF on an interface and defines the OSPF area for that network.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables OSPF on an interface and defines the OSPF area for that network.
network_ip_address/network_mask
Specifies the network address and mask as well as the interface on which OSPF will be enabled.
network_ip_address in entered in IPv4 dotted-decimal notation, followed by the "/" and the mask (CIDR).
area_id
Specifies the OSPF area identification number for the specified network as an integer from 0 through
4294967295.
area_ip_address
Specifies the IP address of the OSPF area for this network. This must be entered in IPv4 dotted-decimal
notation.
Usage Guidelines Use this command to specify the IP address of the network interface that the OSPF router will use.
Examples The following command specified that the OSPF router will use the interface at IP address 192.168.1.0 with
a netmask of 24:
network 192.168.1.0/24
ospf graceful-restart
Configures OSPF graceful-restart settings.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
Syntax Description ospf graceful-restart { grace-period grace_period | helper { never | policy { only-reload |
only-upgrade } } }
grace-period grace-period
Specifies the OSPF graceful restart grace period (in seconds) as an integer from 1 through 1800. Default grace
period is 60 seconds.
Examples The following command sets the graceful restart grace period to 60 seconds:
ospf graceful-restart grace-period 60
ospf graceful-restart helper policy only-reloadL
ospf graceful-restart helper policy only-upgrade
ospf router-id
This command configures the router ID for the OSPF process.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the router ID for the OSPF process.
router-id ip_address
Specifies the router ID for the OSPF process. ip_address is entered using IPv4 dotted-decimal notation.
Usage Guidelines Use this command to set the router ID for the current OSPF router process.
passive-interface
Enables or disables the suppression of OSPF routing updates on the specified interface.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the name assigned to a logical interface within the specific context.
interface_name
Specifies the name assigned to a logical interface within the specific context as an alphanumeric string of 1
through 79 characters.
Usage Guidelines Use this command to suppress router updates on an interface in the current context.
Examples The following command suppresses OSPF routing updates on the interface named Intfc1:
passive-interface Intfc1
redistribute
Redistributes routes from other protocols to OSPF neighbors using the OSPF protocol.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the redistributed routes.
connected
Redistributes connected routes.
rip
Specifies that RIP routes will be redistributed. (RIP is not supported at this time.)
static
Redistributes static routes.
metric metric_value
Sets the OSPF metric used in the redistributed route. This must be an integer from 1 through 16777214.
metric-type { 1 | 2 }
Default: 2
Sets route metric type that is applied to redistributed routes.
1: Sets the OSPF external link type for routes to Type 1.
2: Sets the OSPF external link type for routes to Type 2.
route-map route_map_name
Filter routes through the specified route map before redistribution. route_map_name specifies the name of
the route-map to use as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to define what routing protocols should have their routes redistributed into OSPF.
Examples The following command defines that BGP routes should be redistributed:
redistribute connected
refresh timer
Adjusts settings for the OSPF refresh timer.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the refresh timer.
value
Default: 10
Specifies the minimum amount of time (in seconds) to wait before refreshing a Link-state Advertisement
(LSA). This must be an integer from 10 through 1800.
Usage Guidelines Use this command to define the amount of time to wait before refreshing an LSA.
router-id
Configures the router ID for the OSPF process.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables the router ID for the OSPF process.
ip_address 196
Specifies the router ID for the OSPF process in IPv4 dotted-decimal notation.
Usage Guidelines Use this command to set the router ID for the current OSPF router process.
timers spf
Sets the Shortest Path First (SPF) timers.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration
configure > context context_name > router ospf
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf)#
no
Disables SPF timers.
delay_value
Default: 5
Specifies the delay time (in seconds) between receiving changes to an SPF calculation. This must be an integer
from 0 through 4294967295.
hold_time_value
Default: 10
Specifies the hold time (in seconds) between consecutive SPF calculations. This must be an integer from 0
through 4294967295.
Usage Guidelines Use this command to set the SPF delay and hold timers for the current OSPF router process.
Examples The following command sets the delay timer to 15 and the hold timer to 15:
timers spf 15 15
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
area
Configures an Open Shortest Path First Version 3 (OSPFv3) area and enables authentication for that area.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
no
Disables authentication for the specified area.
decimal_value | ipv4address
decimal_value: Specifies the identification number of the area where authentication will be enabled as an
integer from 0 through 4294967295.
ipv4address: Specifies the IP address of the area where authentication will be enabled in IPv4 dotted-decimal
notation.
default-cost default_integer_value
Sets the OSPFV3 authentication area's default cost as an integer from 1 through 16777215.
stub [ no-summary ]
Sets the OSPFV3 stub area. Only Router-LSAs, Network-LSAs, Inter-area Prefix-LSAs, Intra-area Prefix-LSAs
and Link-LSAs are allowed in a Stub area.
no-summary Does not inject inter-area routes into stub area.
virtual-link virtuallink_neighbour_Ipv4_address
Configures a virtual link to the authentication area.
virtuallink_neighbour_Ipv4_address is the IPv4 address for the virtual link of the authenticated area in
dotted-decimal notation.
The following interval timers can be set for the virtual link:
• dead-interval virtuallink_dead_interval: Sets the virtual link dead-interval (in seconds) as an integer
from 1 through 65535.
• hello-interval virtuallink_hello_interval: Sets the virtual link hello interval (in seconds) as an integer
from 1 through 65535.
• retransmit-interval virtuallink_retransmit_interval: Sets the virtual link retransmit interval (in seconds)
as an integer from 1 through 3600.
• transmit-delay virtuallink_transmit_delay: Sets the virtual link transmit delay (in seconds) as n integer
from 1 through 3600.
Usage Guidelines Use this command to establish OPSFv3 areas and enable authentication.
Examples The following command enables authentication for an OSPFv3 area defined by the IP address 192.168.100.10
with default cost of 256
area 192.168.100.10 default-cost 256
default-metric
Configures the default metric value for routes redistributed from another protocol into Open Shortest Path
First Version 3 (OSPFv3).
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
no
Disables the default metric.
default_metric_integer_value
Specifies the default metric as an integer from 1 through 16777214.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
passive-interface
Configures an interface as being OSPFv3 passive. If a network interface is configured as passive, it will not
receive or send any OSPFv3 packets.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
no
Disables the passive interface.
interface_name
Specifies an OSPFv3 passive interface as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to configure an OPSFv3 passive interface in this context.
Examples The following command configures the OSPF-if1 interface to be OSPFv3 passive.
passive-interface OSPF-if1
redistribute
Redistributes routes from other protocols to OSPFv3 neighbors using the OSPFv3 protocol.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
Syntax Description [ no ] redistribute { connected | static } redistribute connected [ metric metric_value [ metric-type
external_metric_type ] [ route-map route_map_name ] ] [ metric-type external_metric_type [
route-map route_map_name ] ] [ route-map route_map_name ] static [ metric metric_value [
metric-type external_metric_type ] [ route-map route_map_name ] ] [ metric-type
external_metric_type [ route-map route_map_name ] ] [ route-map route_map_name ]
no
Disables the route redistribution.
connected
Redistributes connected routes.
static
Redistributes static routes.
metric metric_value
Specifies the OSPFv3 default metric value as an integer from 0 through 16777214.
metric-type external_metric_type
Specifies the OSPFv3 external metric type as the integer 1 or 2
route-map route_map_name
Specifies a route map as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to configure OPSFv3 redistribution of connected or static routes.
router-id
Sets the OSPFv3 router ID for the Open Shortest Path First Version 3 (OSPFv3) routing process.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
no
Disables the router-id.
router_id_ipaddress
Specifies the router-id an IPv4 address in dotted-decimal notation.
Usage Guidelines Use this command to configure OPSF v3 router id to the given IPv4 address.
Examples The following command configures OSPFv3 router id to the given IPv4 address.
router-id 11.22.22.21
timers spf
Sets OSPFv3 the delay in the time between the detection of a topology change and when the SPF algorithm
actually runs.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPFv3 Configuration
configure > context context_name > router ospfv3
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospfv3)#
no
Disables the SPF delay timer.
spf_delay_timer_value
Sets the Shortest Path First (SPF) delay timer (in milliseconds) as an integer from 0 through 4294967295.
Usage Guidelines Use this command to configure the OPSFv3 SPF delay timer.
The OSPF VRF Configuration sub-mode is used to configure the virtual routing and forwarding (VRF)
context instances for OSPF routing protocol. This mode includes commands that configure VRF instance
for OSPF routing parameters.
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
area
Configures various parameters, including authentication, area identification, virtual link ID, and delay/interval
values for the specified OSPF area using a specific VRF instance.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
Syntax Description [no] area { decimal_value | ip_address } {authentication [ message-digest ] | default-cost cost_value |
nssa [default-information-originate] [no-redistribution] [no-summary] [translate-always]
[translate-candidate] [translate-never] | stub [no-summary] | virtual-link router_ip_address [authentication
{message-digest | null | text}| authentication-key {encrypted password encrypted_string | password
password_string}| message-digest-key key_id md5 [encrypted password encrypted_string | password
password_string]} [dead-interval] [hello-interval] [retransmit-interval] [transmit-delay]
no
Disables or removes configured parameters for the specified OSPF area using a specific VRF instance.
ip_address
Specifies the IP address of the area where authentication will be enabled in IPv4 dotted-decimal notation.
decimal_value
Specifies the identification number of the area where parameters to be configured as an integer from 0 through
4294967295.
authentication
Sets the OSPF authentication type to use the simple authentication method.
message-digest
Sets the OSPF authentication type to use the message digest 5 (MD5) authentication method.
default-cost cost_value
Sets the default cost for an OSPF area. cost_value must be an integer from 0 through 16777215.
stub [no-summary]
Specifies an OSPF area as an stub area configures the NSSA-ABR never to translate. By default this is disabled.
no-summary: Disables (stops) the ABR from sending summary LSAs into the stub area.
virtual-link router_id
Specifies the router identifier which provides a virtual link between an area that cannot be physically connected
to the network backbone and an area that is physically connected to the network backbone.
router_id must be an IP address in IPv4 dotted-decimal notation of the ABR to be linked to.
authentication-key
Configures the authentication password for the virtual link between an area that cannot be physically connected
to the network backbone and an area that is physically connected to the network backbone.
message-digest-key key_id
Specifies the MD key identifier number for virtual link connection. key_id must be an integer from 1 through
255.
md5
Sets the message digest to MD5 for virtual link connection.
dead-interval value
Specifies the dead interval (in seconds) that the router should wait, during which time no packets are received
and after the router considers a neighboring router to be off-line. value must be an integer from 1 through
65535.
hello-interval value
Specifies the hello interval (in seconds) before sending a hello packet. value must be an integer from 1 through
65535.
retransmit-interval value
Specifies the delay between retransmission (in seconds) that router should wait before retransmitting a packet.
value must be an integer from 1 through 3600.
transmit-delay value
Specifies the interval (in seconds) that the router should wait before transmitting a packet. value must be an
integer from 1 through 3600.
Usage Guidelines Use this command to configure/set the various network/connection/authentication parameters of OPSF areas
using specific VRF instance.
Examples The following command enables authentication for an OSPF area defined by the IP address 192.168.100.10
and the OSPF authentication type to MD5:
area 192.168.100.10 authentication message-digest
The following command defines the area designated by the IP address 192.168.100.10 as an NSSA area where
translation of NSSA candidate is enabled by default:
area 192.168.100.10 nssa
The following command defines the OSPF area defined by the IP address 192.168.100.10 as a stub area:
area 192.168.100.10 stub
The following command creates a virtual link between the OSPF areas defined by the IP address 192.168.100.10
and the IP address 192.168.200.20:
area 192.168.100.10 virtual-link 192.168.200.20
The following command sets the authentication method for a virtual link between the OSPF areas defined by
the IP address 192.168.100.10 and the IP address 192.168.200.20 to use no authentication:
area 192.168.100.10 virtual-link 192.168.200.20 null
The following command creates an authentication password of 123456 for a virtual link between the OSPF
areas defined by the IP address 192.168.100.10 and the IP address 192.168.200.20:
area 192.168.100.10 virtual-link 192.168.200.20 authentication-key password 123456
The following command enables the use of MD5-based OSPF authentication for a virtual link between the
OSPF areas defined by the IP address 192.168.100.10 and the IP address 192.168.200.20, sets the MD5 Key
ID to 25, and the password to 123456:
area 192.168.100.10 virtual-link 192.168.200.20 message-digest-key 25 md5 password 123456
The following command sets the retransmit interval for a virtual link between the OSPF areas defined by the
IP address 192.168.100.10 and the IP address 192.168.200.20 to 60 seconds:
area 192.168.100.10 virtual-link 192.168.200.20 retransmit-interval 60
default-information originate
Creates a default external route into an OSPF routing domain.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the default external route.
always
Always advertise the route regardless of whether or not the software has a default route.
metric metric_value
Sets the OSPF metric used in creating the default rout as an integer from 1 through 16777214.
metric-type { 1 | 2 }
Sets the default route metric type.
1: Sets the OSPF external link type for default routes to Type 1.
2: Sets the OSPF external link type for default routes to Type 2.
route-map route_map_name
Specifies the name of the default route-map to be use as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to set the default external route into an OSPF routing domain.
Examples The following command sets the default external route to originate from the route map named rmap1:
default-information originate route-map rmap1
default-metric
Configures the default metric value for the OSPF routing protocol.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
metric_value
Sets the metric value expressed as an integer from 1 through 16777214. The default metric value setting is
26385.
no
Enables or disables the default metric value for OSPF.
Usage Guidelines Use this command to set the default metric for routes.
distance
Configures the OSPF route administrative distances for all OSPF route types or based on specific route type.
Administrative distance is the measure used by Cisco routers to select the best path when there are two or
more different routes to the same destination from two different routing protocols. Administrative distance
defines the reliability of a routing protocol. Each routing protocol is prioritized in order of most to least reliable
(believable) using an administrative distance value. A lower numerical value is preferred.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the soecified option.
distance_value
Specifies the OSPF route administrative distances as an integer from 1 to 255. The default distance value is
110.
external ext_distance_value
Sets the OSPF route administrative distance for routes from other routing domains, learned by redistribution.
ext_distance_value must be an integer from 1 through 255. The default is 110.
inter-area inter_distance_value
Sets the OSPF route administrative distance for routes from one routing area to another. inter_distance_value
must be an integer from 1 through 255. The default is 110.
intra-area intra_distance_value
Sets the OSPF route administrative distance for all routes within an area. intra_distance_value must be an
integer from 1 through 255. The default is 110.
Usage Guidelines Use this command to set the administrative distance for OSPF routes.
Examples The following command sets the administrative distance for all OSPF route types to 30:
distance 30
distribute-list
Enables or disables the filtering of networks in outgoing routing updates.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the filtering of networks in outgoing routing updates.
route_access_list
Specifies the name of the OSPF route access list to use for filtering as an alphanumeric string of 1 through
63 characters.
connected
Filters connected routes.
rip
Filters RIP routes. (RIP is not supported at this time.)
static
Filters static routes.
Usage Guidelines Use this command to enable the filtering of outgoing route updates by using the specified route access list.
Examples The following command uses the route access list named ral1 to filter outgoing routing updates for all connected
routes:
distribute-list ral1 out connected
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
neighbor
Configures OSPF routers that interconnect to non-broadcast networks.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables OSPF routers that interconnect to non-broadcast networks.
ip_address
Specifies the interface IP address of the OSPF neighbor expressed using IPv4 dotted-decimal notation.
poll-interval poll_interval_value
Default: 120
Sets the number of seconds in the dead neighbor polling interval as an integer from 1 through 65535
priority priority_value
Default: 0
Sets the 8-bit number that represents the router priority value of the non-broadcast neighbor associated with
the specified IP address. This must be an integer from 0 through 255. This keyword does not apply to
point-to-multipoint interfaces.
Usage Guidelines Use this command to configure OSPF routers that connect to non-broadcast networks.
Examples The following command specifies an OSPF router neighbor with the IP address of 192.168.100.10:
neighbor 192.168.100.10
network
Enables OSPF on an interface and defines the OSPF area for that network.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables OSPF on an interface and defines the OSPF area for that network.
network_ip_address/mask
Specifies the network address and mask as well as the interface on which OSPF will be enabled.
network_ip_address in entered in IPv4 dotted-decimal notation, followed by the "/" and the mask in CIDR
notation.
area_id
Specifies the OSPF area identification number for the specified network as an integer from 0 through
4294967295.
area_ip_address
Specifies the IP address of the OSPF area for this network. This must be entered in IPv4 dotted-decimal
notation.
Usage Guidelines Use this command to specify the IP address of the network interface that the OSPF router will use.
Examples The following command specified that the OSPF router will use the interface at IP address 192.168.1.0 /24
an area ID 2345 and IP address 192.168.1.5:
network 192.168.1.0/24 area 2345 192.168.1.5
ospf router-id
Configures the router ID for the OSPF process.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the specified OSPF router.
ip_address
Specifies the router ID for the OSPF process as an IP address entered using IPv4 dotted-decimal notation
Usage Guidelines Use this command to set the router ID for the current OSPF router process.
passive-interface
Enables or disables the suppression of OSPF routing updates on the specified interface.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the name assigned to a logical interface within the specific context.
interface_name
Specifies the name assigned to a logical interface within the context as an alphanumeric string of 1 through
79 characters.
Usage Guidelines Use this command to suppress router updates on an interface in the current context.
Examples The following command suppresses OSPF routing updates on the interface named Intfc1:
passive-interface Intfc1
redistribute
Redistributes routes from other protocols to OSPF neighbors using the OSPF protocol.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the redistributed routes.
connected
Redistributes connected routes.
rip
Specifies that RIP routes will be redistributed. (RIP is not supported at this time.)
static
Redistributes static routes.
metric metric_value
Sets the OSPF metric used in the redistributed route. This must be an integer from 1 through 16777214.
metric-type { 1 | 2 }
Default: 2
Sets route metric type that is applied to redistributed routes.
1: Sets the OSPF external link type for routes to Type 1.
2: Sets the OSPF external link type for routes to Type 2.
route-map route_map_name
Filter routes through the specified route map before redistribution. route_map_name specifies the name of
the route-map to use as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to define what routing protocols should have their routes redistributed into OSPF.
Examples The following command defines that BGP routes should be redistributed:
redistribute connected
refresh timer
Adjusts settings for the OSPF refresh timer.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the refresh timer.
value
Default: 10
Specifies the minimum amount of time (in seconds) to wait before refreshing a Link-state Advertisement
(LSA). This must be an integer from 10 through 1800.
Usage Guidelines Use this command to define the amount of time to wait before refreshing an LSA.
router-id
Configures the router ID for the OSPF process.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables the router ID for the OSPF process.
ip_address 92
Specifies the router ID for the OSPF process in IPv4 dotted-decimal notation.
Usage Guidelines Use this command to set the router ID for the current OSPF router process.
timers spf
Sets the Shortest Path First (SPF) timers.
Product PDSN
HA
GGSN
Command Modes Exec > Global Configuration > Context Configuration > OSPF Configuration > OSPF VRF Configuration
configure > context context_name > router ospf > ip vrf vrf_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-ospf-vrf)#
no
Disables SPF timers.
delay_value
Default: 5
Specifies the delay time (in seconds) between receiving changes to an SPF calculation. This must be an integer
from 0 through 4294967295.
hold_time_value
Default: 10
Specifies the hold time (in seconds) between consecutive SPF calculations. This must be an integer from 0
through 4294967295.
Usage Guidelines Use this command to set the SPF delay and hold timers for the current OSPF router process.
Examples The following command sets the delay timer to 15 and the hold timer to 15:
timers spf 15 15
The Out-Address configuration mode provides the commands to configure the outbound parameters for the
SCCP entities as part of the gtt-address-map configuration.
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the configuration mode and returns to the Global configuration mode.
Product SGSN
gt-address
Configures the SCCP short address.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
gt_address
A string of 1 to 15 digits to define the GT-address
gt-format
The GT-format provides four formats that can be used during GTT.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
format_num
1: Selects GT-format 1 options which include nature-of-address and odd/even. Once selected, the system
enters GT-Format1 configuration mode.
2: Selects GT-format2 options which include translation-type. Once selected, the system enters GT-Format2
configuration mode.
3: Selects GT-format3 options which include encoding-scheme, numbering-plan3 and translation-type.
Once selected, the system enters GT-Format1 configuration mode.
4: Selects GT-format4 options which include encoding-scheme, nature-of-address, numbering-plan, and
translation-type. Once selected, the system enters GT-Format4 configuration mode.
Usage Guidelines Select the a GT-format that include encoding-scheme as part of the GTT process.
Examples gt-format 3
ni-indicator
Configures the National and International indicator to use during the GTT process.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
ni_ind
Select one of the following as the appropriate type of national indicator for the address structure:
• national
• international
Usage Guidelines Select the international indicator to be used for out-going addresses.
point-code
Selects and configures the SS7-type point code for use with the out-going address.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
pt_code
Enter 1 to 11 digits in the point code format predefined during variant selection of GTT association.
Usage Guidelines Define an ITU point code to be used for out-going address processing.
routing-indicator
Selects the type of routing and the indicator to be included in the out-going message.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
routing_ind
Select one of the following options:
• gt: Inserts an indicator that identifies routing based on global title.
• ssn: Inserts an indicator that identifies routing based on the subsystem number.
Examples routing-indicator gt
ssf
Selects the subservice field as factor in the out-going address processing. ssf sets the network indicator in the
subservice field for SS7 Message Signal Units (MSUs). The indicator carried in the message's routing
information typically identifies the structure of the point code as a message from within a nation or as a
message coming from outside the national - international.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
sub_svc_fld
Select one of the following options:
• international: The network indicator identifies the message as international with a point code structure
that does not match the national point code structure,
• national: The network indicator identifies the messages as having a national point code structure.
• reserved: Provides an alternate network indicator for national messages.
• spare: Provides an alternate network indicator for international messages.
Usage Guidelines Select the international NI for inclusion in out-going address subservice fields.
ssn
Selects the subsystem number to be included in the out-going message.
Product SGSN
Command Modes Exec > Global Configuration > GTT Address-Map Configuration > Out-Address Configuration
configure > gtt address-map map_id > out-address address_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-gtt-addrmap-outaddr-out_address)#
sub_sys_num
Enter an integer from 1 to 255.
Examples ssn 44
Command Modes Exec > ACS Configuration > P2P Advertisement Server Group Configuration
active-charging service service_name > p2p-ads-group ads_group_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-acs-p2p-ads)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
ad-source
This command allows to configure the P2P advertisement source that can be a HTTP host or SSL server.
Product ADC
Command Modes Exec > ACS Configuration > P2P Advertisement Server Group Configuration
active-charging service service_name > p2p-ads-group ads_group_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-acs-p2p-ads)#
no
If previously configured, disables the configured ad-source.
operator
Specifies how to match.
operator must be one of the following:
• =: Equals
• contains: Contains
• ends-with: Ends with
• starts-with: Starts with
http_host_name/ssl_server_name
Specifies the name of the HTTP host or SSL server to match and must be an alphanumeric string of 1 through
127 characters. SSL supports the Server Name indication (SNI) field.
Usage Guidelines This command allows to configure the P2P advertisement source that can be a HTTP host or SSL server when
the user runs an active application session. The ad-source can be server name indication for HTTPS flows
and host name for HTTP-based ad flows.
Important The maximum number of ad-source lines that can be configured is 32.
Examples The following command matches the ad-source string ending with admob.com:
ad-source ends-with admob.com
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
map-to-application
This command allows to configure the P2P advertisement application that will map the advertisement group
to the corresponding application/protocol.
Product ADC
Command Modes Exec > ACS Configuration > P2P Advertisement Server Group Configuration
active-charging service service_name > p2p-ads-group ads_group_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-acs-p2p-ads)#
no
If previously configured, disables the application/protocol mapping.
p2p_list
Specifies the list of protocols/applications supported in the P2P plugin.
+
More than one protocol/application supported in the P2P plugin can be entered within a single command.
Usage Guidelines This command allows to configure the P2P advertisement application that will map the advertisement group
to the application protocol.
The maximum number of map-to-application rule lines that can be configured is equal to the number of the
applications present in p2p_list supported by P2P plugin.
Examples The following command maps the ads-group to the slacker-radio application:
map-to-application slacker-radio
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC- Action-Set Configuration Mode provides the parameters to indicate the policy and charging as
well as event generation related decisions that will get activated when the corresponding PCC-Condition-Group
is evaluated to TRUE within a profile. A maximum of 32 actions can be configured in an instance of
PCC-Action-Set.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
af-media-type
This command is used to set the action to be taken when specific media type is received from Application
Function (AF) over Rx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
Syntax Description [ no ] af-media-type {application | audio | control | data | message | other | text | video} {qos-profile
qos_profile_value[ gate-status { disabled | enabled | enabled-downlink | enabled-uplink ] } | {gate-status
{disabled | enabled | enabled-downlink | enabled-uplink [qos-profile qos_profile_value] }[ monitoring-key
mon_key_id
no
Removes the configured action set for different type of AF media in a PCC-Action-Set configuration instance
for PCC-Service configuration.
application
Sets the AF media type to 'application' data traffic for various action configuration under PCC-Action-Set
configuration instance for PCC-Service configuration.
audio
Sets the AF media type to 'audio' data traffic for various action configuration under PCC-Action-Set
configuration instance for PCC-Service configuration.
control
Sets the AF media type to 'control' signal for various action configuration under PCC-Action-Set configuration
instance for PCC-Service configuration.
data
Sets the AF media type to 'data' for various action configuration under PCC-Action-Set configuration instance
for PCC-Service configuration.
message
Sets the AF media type to 'message' data for various action configuration under PCC-Action-Set configuration
instance for PCC-Service configuration.
other
Sets the AF media type to 'other', out of AF media type configured here, for various action configuration under
PCC-Action-Set configuration instance for PCC-Service configuration.
text
Sets the AF media type to 'text' for various action configured under PCC-Action-Set configuration instance
for PCC-Service configuration.
video
Sets the AF media type to 'video' for various action configured under PCC-Action-Set configuration instance
for PCC-Service configuration.
qos-profile qos_profile_value
Associate the pre-defined PCC-QoS-Profile for specific AF media type for various action configured for
PCC-Service configuration.
qos_profile_value is name of the pre-defined PCC-QoS-Profile in PCC-Service Configuration instance.
monitoring-key mon_key_id
Specifies the Monitoring Key identifier to be associated with AF-Media-type under PCC-Action-Set
configuration instance for PCC-Service configuration.
mon_key_id must be a preconfigured monitoring key having integer between 1 through 65535.
Usage Guidelines Use this command to set the action to be taken when specific media type is received from Application Function
(AF) over Rx interface.
It also associates the pre-defined PCC-QoS-Profile and Gate function and monitoring key with specific media
type.
Examples The following command sets the AF media type to 'video' with PCC-QoS-Profile named video_qos1 with
gate status enabled in downlink and uplink traffic for various action configured under PCC-Action-Set
configuration instance for PCC-Service configuration:
af-media-type video qos-profile video_qos1 gate-status enabled
associate monitoring-key
This command associates a Monitoring Key id with a PCC-Usage-Monitor in PCC-Action-Set configuration
for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the associated Monitoring Key id configuration along with the association to PCC-Usage-Monitor
from a PCC-Action-Set configuration instance for PCC-Service configuration.
mon_key_id
Specifies the Monitoring Key identifier to be associated with PCC-Usage-Monitor under PCC-Action-Set
configuration instance for PCC-Service configuration.
mon_key_id must be an integer between 1 through 65535.
usage-monitor usage_mon_name
Specifies the PCC-Usage-Monitor associated with Monitoring Key mon_key_id under PCC-Action-Set
configuration instance for PCC-Service configuration.
usage_mon_name is a pre-configured PCC-Usage-Monitor instance in PCC-Service Profile Configuration
for PCC-Service configuration.
Usage Guidelines Use this command to associate a Monitoring Key with a PCC-Usage-Monitor in PCC-Action-Set configuration
for PCC-Service instance.
There is a Many-To-Many relationship between Usage-Monitor and Monitoring-Key. Operator can change
this relationship using required command.
When usage is reported to IPCF for a particular Monitoring Key, the usage is added to all the Usage Monitoring
to which the Monitoring Key is associated.
Operator can break the relationship between Monitoring Key and Usage Monitor by dissociate monitoring-key
command.
Examples The following command associates Monitoring Key id 102 with pre-defined PCC-Usage-Monitor named
usage_mon1 under PCC-Action-Set configuration instance for PCC-Service configuration:
associate monitoring-key 102 usage-monitor usage_mon1
authorize
This command sets an action to change the various authorization parameters used for the IP-CAN session
under PCC-Action-Set configuration instance for PCC-Service configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
Syntax Description [no] authorize {apn-mbr | default-eps-bearer | qci qci_value} qos-profile qos_prof_name
no
Removes the configured action for QoS Profile authorization parameters from PCC-Action-Set configuration
instance for PCC-Service configuration.
apn-mbr
Sets the action for QoS authorization of Maximum Bit Rate (MBR) at APN through a pre-configured QoS
profile name qos_prof_name which is to use for the authorization at different level in an PCC-Service instance.
This keyword is applicable only to EPS access type. MBR Download and MBR Upload values of the QoS
profile is used with this authorization.
default-eps-bearer
Sets the action QoS authorization of the Default EPS bearer in an EPS access network through a pre-configured
QoS profile name qos_prof_name which is to use for the authorization at different level in an PCC-Service
instance.
This keyword is applicable only to EPS access type. QoS Class Identifier (QCI) and Allocation and Retention
Priority (ARP) values of the QoS profile is used with this authorization.
qci qci_value
Sets the action QoS authorization per QCI in case of PCEF binding through a pre-configured QoS profile
name qos_prof_name which is to use for the authorization at different level in an PCC-Service instance.
qci_value must be an integer value between 1 through 255.
Important This keyword is applicable only in case of PCEF binding. QCI defined in QoS profile is not used when
this keyword is active, so this keyword is repeated per QCI that has to be authorized by IPCF. For each
QCI authorization Maximum Bit Rate Upload/Download (MBR DL/MBR UL) or Guaranteed Bit Rate
Upload/Download (GBR DL/MBR UL) and ARP values of the QoS profile is used with this authorization.
qos-profile qos_prof_name
This keyword associate the action configured for authorization with a pre-configured PCC-QoS-Profile named
qos_prof_name and uses configured values from specific PCC-QoS-Profile during authorization.
qos_prof_name specifies the pre-configured QoS profile name which is to use for the authorization at different
level in an PCC-Service instance.
Usage Guidelines Use this command to define an action for the authorization parameters in PCC-QoS-Profile which is to be
used under PCC-Action-Set configuration instance for PCC-Service configuration.
Examples Following command sets the action for QoS authorization for APN MBR with PCC-QoS-Profile apn_qos_prof1
under PCC-Action-Set configuration instance for PCC-Service configuration.
authorize apn-mbr qos-profile apn_qos_prof1
dissociate monitoring-key
This command dissociates a Monitoring Key id with a PCC-Usage-Monitor in PCC-Action-Set configuration
for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the dissociated Monitoring Key id configuration from a PCC-Action-Set configuration instance for
PCC-Service configuration.
mon_key_id
Specifies the Monitoring Key identifier to be dissociated with PCC-Usage-Monitor under PCC-Action-Set
configuration instance for PCC-Service configuration.
mon_key_id must be an integer between 1 through 65535.
usage-monitor usage_mon_name
Specifies the PCC-Usage-Monitor need to be dissociated with Monitoring Key mon_key_id under
PCC-Action-Set configuration instance for PCC-Service configuration.
usage_mon_name is a pre-configured PCC-Usage-Monitor instance in PCC-Service Profile Configuration
for PCC-Service configuration.
Usage Guidelines Use this command to dissociate a Monitoring Key with a PCC-Usage-Monitor in PCC-Action-Set configuration
for PCC-Service instance.
There is a Many-To-Many relationship between Usage-Monitor and Monitoring-Key. Operator can change
this relationship using required commands.
When usage is reported to IPCF for a particular Monitoring Key, the usage is added to all the Usage Monitoring
to which the Monitoring Key is associated.
Operator can reassociate the relationship between Monitoring Key and Usage Monitor by associate
monitoring-key command.
Examples The following command dissociates a associated pair of Monitoring Key id 102 and PCC-Usage-Monitor
named usage_mon1 under PCC-Action-Set configuration instance for PCC-Service configuration:
dissociate monitoring-key 102 usage-monitor usage_mon1
dynamic-rule-install
This command sets an action to install a PCC Dynamic rule for the specified PCC-Data-service in
PCC-Action-Set configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the Dynamic Rule install action configured for specific PCC-Data-Service from a PCC-Action-Set
configuration instance for PCC-Service configuration.
data-service data_svc_name
Specifies the name of the pre-configured PCC-Data-Service for which Dynamic Rules to be installed under
PCC-Action-Set configuration instance for PCC-Service configuration.
data_svc_name is a pre-configured PCC-Data-Service instance in PCC-Service Configuration mode.
• duration sets the metering method as based on duration of session usage for offline charging in a
PCC-Data-service instance on IPCF node.
• volume sets the metering method as based on the volume of data usage for offline charging in a
PCC-Data-service instance on IPCF node.
monitoring-key mon_key_id
Specifies the pre-defined Monitoring Key identifier which is to be used in Dynamic Rule for specific
PCC-Data-service under PCC-Action-Set configuration instance for PCC-Service configuration.
mon_key_id is a pre-defined Monitoring Key identifier in PCC-Service Configuration mode.
precedence preced_value
Specifies the precedence value for the Dynamic Rule for specific PCC-Data-service under PCC-Action-Set
configuration instance for PCC-Service configuration.
preced_value must be an integer between 1 through 65535.
qos-profile qos_prof_name
Specifies the name of the pre-configured PCC-QoS-Profile which is to be used in Dynamic Rule for specific
PCC-Data-Service under PCC-Action-Set configuration instance for PCC-Service configuration.
qos_prof_name is a pre-configured PCC-QoS-Profile instance in PCC-Service Configuration mode.
rating-group rating_grp_id
Specifies the pre-defined Rating Group identifier which is to be used in Dynamic Rule for specific
PCC-Data-service under PCC-Action-Set configuration instance for PCC-Service configuration.
rating_grp_id is a pre-defined Rating Group identifier in PCC-Service Configuration mode.
disabled: disables the Gate status in downlink and uplink direction which is to be used in Dynamic Rule for
specific PCC-Data-Service under PCC-Action-Set configuration instance for PCC-Service configuration.
enabled: Enables the Gate status in downlink and uplink direction which is to be used in Dynamic Rule for
specific PCC-Data-Service under PCC-Action-Set configuration instance for PCC-Service configuration. This
is the default status of Gate.
enabled-downlink: Enables the Gate status in downlink direction which is to be used in Dynamic Rule for
specific PCC-Data-Service under PCC-Action-Set configuration instance for PCC-Service configuration.
enabled-uplink: Enables the Gate status in uplink direction which is to be used in Dynamic Rule for specific
PCC-Data-Service under PCC-Action-Set configuration instance for PCC-Service configuration.
Usage Guidelines Use this command to set an action to install a PCC Dynamic rule for the specified PCC-Data-Service in
PCC-Action-Set configuration for PCC-Service instance
Operator can override parameters such as, QoS profile, Precedence, Gate-status, Monitoring Key and
PCC-Rating-id.
The same command is used to modify already installed PCC Dynamic-rule for the PCC-Data-Service by
overriding required parameters only.
Additionally defer-by interval keyword is used to configure relative time by which the corresponding dynamic
rule installation is deferred.
IPCF handles operation of PCC Rule and activate/deactivate/install/modify/remove the PCC rules at PCEF
through this configuration. PCC rule operation may fail on PCEF due to various reasons. In such failure cases
PCEF sends back a Charging Rule Report containing name of the failed PCC rule and corresponding failure
cause.
The IPCF handles these charging rule report and take appropriate actions based on configuration done through
failure-policy keyword.
Charging Rule Report comes through CCA or RAA messages in a call flow used for handling the
charging-rule-report.
IPCF supports following charging rule failure codes in report:
• Out-of-credit
• Reallocation-of-credit
• Unknown rule name
• Invalid Rating Group
• Invalid Service Identifier
• GW/PCEF Malfunction
• Limited Resources
• Max No. of Bearers Reached
• Unknown Bearer Id
• Missing Bearer Id
• Missing Flow Description
• Resource Allocation Failure
• QoS Validation Failure
Charging rule status can any one of the following in this scenario:
• Active
• Inactive
• Temporarily Inactive
A charging rule report can occur in CCR message multiple times and maximum of 16 charging rule reports
per CCR message is supported by IPCF.
Examples The following command sets the action for PCC-Data-Service named temp_data1 for Dynamic Rule install
with PCC-QoS-Profile named temp_qos1 having precedence 22 and Gate-status as Enabled under
PCC-Action-Set configuration instance for PCC-Service configuration:
dynamic-rule-install data-service temp_data1 qos-profile temp_qos1 precedence 22
dynamic-rule-uninstall
This command sets an action to uninstall a Dynamic Rule for the specified PCC-Data-Service in PCC-Action-Set
configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the Dynamic Rule Uninstall action configured for specific PCC-Data-Service from a PCC-Action-Set
configuration instance for PCC-Service configuration.
data-service data_svc_name
Specifies the name of the pre-configured PCC-Data-Service for which Dynamic Rules to be uninstalled under
PCC-Action-Set configuration instance for PCC-Service configuration.
data_svc_name is a pre-configured PCC-Data-Service instance in PCC-Service Configuration mode.
• SS specifies the seconds to defer the action trigger and must be an integer between 00 through 59.
Usage Guidelines Use this command to set an action to uninstall a PCC Dynamic rule for the specified PCC-Data-Service in
PCC-Action-Set configuration for PCC-Service instance
Additionally defer-by interval keyword is used to configure relative time by which the corresponding dynamic
rule uninstallation is deferred.
Examples The following command sets the action of Dynamic Rule uninstall for PCC-Data-Service named temp_data1
under PCC-Action-Set configuration instance for PCC-Service configuration:
dynamic-rule-uninstall data-service temp_data1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
log-event
This command allows operator to specify a string to be logged at Subscriber Service Controller (SSC) when
the corresponding action set is triggered under PCC-Action-Set configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the notification message configured for specific action under a PCC-Action-Set configuration
instance for PCC-Service configuration.
msg-text message_string
Specifies the message notification text string to be logged at the SSC when a particular PCC-Action-Set
triggered for PCC-Service instance.
message_string is a string of alphanumerical characters of 1 through 255 characters
Usage Guidelines Use this command to allow operator to specify a string to be logged at Subscriber Service Controller (SSC)
when the corresponding action set is triggered under PCC-Action-Set configuration for PCC-Service instance.
Examples The following command sets the notification message for an action under PCC-Action-Set configuration
instance for PCC-Service configuration:
log-event msg-text "This Action is Applicable for EPS Session Only."
notify-user
This command allows operator to specify a string template-id at SSC under PCC-Action-Set configuration
for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the user notification message id configured for specific action under a PCC-Action-Set configuration
instance for PCC-Service configuration.
message-id message_id
Specifies the message template id stored at SSC to be used for user notification for action under PCC-Action-Set
configuration for PCC-Service instance.
message_id is a string of alphanumerical characters of 1 through 255 characters
parameter value_pair
Specifies the parameters in the form of name value pairs separated by comma. A maximum of 16 name value
pairs can be defined for action under PCC-Action-Set configuration for PCC-Service instance.
The name value pairs is encoded into xml data like other parameters and sent out to SSC for user notification.
value_pair is a string of alphanumerical characters of 1 through 255 characters which can accommodate up
to 16 name value pair separated by comma (,).
Usage Guidelines Use this command to allow operator to specify a string template-id under PCC-Action-Set configuration for
PCC-Service instance.
The template description exists at SSC and when the corresponding action set is hit, the subscriber is notified
with the specified template configured via this command.
The action set makes a XML remote procedure call towards SSC. The XML contains information like IMSI,
template-id of the SMS/E-mail template and other details like MSISDN, NAI if available. To give more
flexibility to the operator, now the this command accepts name value pairs, which are also sent along with
the said data.
Examples The following command sets the message id "Invalid User." for user notification message to an action under
PCC-Action-Set configuration instance for PCC-Service configuration:
notify-user message-id "Invalid User."
offline-charging-server
This command sets the action to change the offline charging server applicable to an IP-CAN session under
PCC-Action-Set configuration instance for PCC-Service configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action set to change the offline charging server under PCC-Action-Set configuration instance
for PCC-Service configuration.
primary pri_serv_address
Identifies the name of the primary offline charging server which is to be used for action to change under
PCC-Action-Set configuration instance for PCC-Service configuration.
The pri_serv_address must be the address of an offline charging server in Diameter URI format (FQDN [port]
[transport] [protocol]).
secondary sec_serv_address
Identifies the name of the secondary offline charging server which is to be used for action to change under
PCC-Action-Set configuration instance for PCC-Service configuration.
The sec_serv_address must be the address of an offline charging server in Diameter URI format (FQDN [port]
[transport] [protocol]).
Usage Guidelines Use this command to set an action to change the offline charging server applicable to an IP-CAN session
under PCC-Action-Set configuration instance for PCC-Service configuration.
Examples Following command configures an action change the offline charging server to primary server
aaa://host.abc.com:6666;transport=tcp;protocol=diameter and secondary server
aaa://host.xyz.com:6666;transport=sctp;protocol=radius under PCC-Action-Set configuration instance for
PCC-Service configuration:
offline-charging-server primary aaa://host.abc.com:6666;transport=tcp;protocol=diameter secondary
aaa://host.xyz.com:6666;transport=sctp;protocol=radius
online-charging-server
This command sets the action to change the online charging server applicable to an IP-CAN session under
PCC-Action-Set configuration instance for PCC-Service configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action set to change the online charging server under PCC-Action-Set configuration instance
for PCC-Service configuration.
primary pri_serv_address
Identifies the name of the primary online charging server which is to be used for action to change under
PCC-Action-Set configuration instance for PCC-Service configuration.
The pri_serv_address must be the address of an online charging server in Diameter URI format (FQDN [port]
[transport] [protocol]).
secondary sec_serv_address
Identifies the name of the secondary online charging server which is to be used for action to change under
PCC-Action-Set configuration instance for PCC-Service configuration.
The sec_serv_address must be the address of an online charging server in Diameter URI format (FQDN [port]
[transport] [protocol]).
Usage Guidelines Use this command to set an action to change the online charging server applicable to an IP-CAN session under
PCC-Action-Set configuration instance for PCC-Service configuration.
Examples Following command configures an action change the online charging server to primary server
aaa://host.abc.com:6666;transport=tcp;protocol=diameter and secondary server
aaa://host.xyz.com:6666;transport=sctp;protocol=radius under PCC-Action-Set configuration instance for
PCC-Service configuration:
online-charging-server primary aaa://host.abc.com:6666;transport=tcp;protocol=diameter secondary
aaa://host.xyz.com:6666;transport=sctp;protocol=radius
request-usage-report monitoring-key
This command configures the action to allow operator to explicitly request usage report for the specified
Monitoring Key or all Monitoring Keys under PCC-Action-Set configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the usage request action configuration for Monitoring Key id from a PCC-Action-Set configuration
instance for PCC-Service configuration.
mon_key_id
Specifies the Monitoring Key identifier for which usage report action is to be set under PCC-Action-Set
configuration instance for PCC-Service configuration.
mon_key_id must be an integer between 1 through 65535.
any
Specifies the Monitoring Key identifier to "ANY" value for which usage report action is to be set under
PCC-Action-Set configuration instance for PCC-Service configuration.
Usage Guidelines Use this command to configure the action to allow operator to explicitly request usage report for the specified
Monitoring Key or all Monitoring Keys under PCC-Action-Set configuration for PCC-Service instance.
Examples The following command sets an action to request the usage report for Monitoring Key id 102 under
PCC-Action-Set configuration instance for PCC-Service configuration:
request-usage-report monitoring-key 102
rule-activate
This command sets an action to activate a pre-configured Rule in PCC-Action-Set configuration for PCC-Service
instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to activate a pre-configured Rule from a PCC-Action-Set configuration instance for
PCC-Service configuration.
rule_name
Specifies the name of the pre-configured Rule on PCEF for an activation action is set under PCC-Action-Set
configuration instance for PCC-Service configuration.
rule_name is a pre-configured Rule in PCEF.
• SS specifies the seconds to defer the action trigger and must be an integer between 00 through 59.
Usage Guidelines Use this command to set an action to activate a pre-configured Rule in PCC-Action-Set configuration for
PCC-Service instance.
Additionally defer-by interval keyword is used to configure relative time by which the corresponding rule
activation is deferred.
Examples The following command sets an action to activate for Rule rule1 under PCC-Action-Set configuration instance
for PCC-Service configuration:
rule-activate rule1
rule-deactivate
This command sets an action to deactivate a pre-configured Rule in PCC-Action-Set configuration for
PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to deactivate a pre-configured Rule from a PCC-Action-Set configuration instance for
PCC-Service configuration.
rule_name
Specifies the name of the pre-configured Rule on PCEF for a deactivation action is set under PCC-Action-Set
configuration instance for PCC-Service configuration.
rule_name is a pre-configured Rule in PCEF.
• SS specifies the seconds to defer the action trigger and must be an integer between 00 through 59.
Usage Guidelines Use this command to set an action to deactivate a pre-configured Rule in PCC-Action-Set configuration for
PCC-Service instance.
Additionally defer-by interval keyword is used to configure relative time by which the corresponding rule
deactivation is deferred.
Examples The following command sets an action to deactivate for Rule rule1 under PCC-Action-Set configuration
instance for PCC-Service configuration:
rule-deactivate rule1
rulebase-activate
This command sets an action to activate a pre-configured Rulebase in PCC-Action-Set configuration for
PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to activate a pre-configured Rulebase from a PCC-Action-Set configuration instance for
PCC-Service configuration.
rulebase_name
Specifies the name of the pre-configured Rulebase on PCEF for an activation action is set under PCC-Action-Set
configuration instance for PCC-Service configuration.
rulebase_name is a pre-configured Rulebase in PCEF.
• SS specifies the seconds to defer the action trigger and must be an integer between 00 through 59.
Usage Guidelines Use this command to set an action to activate a pre-configured Rulebase in PCC-Action-Set configuration for
PCC-Service instance.
Additionally defer-by interval keyword is used to configure relative time by which the corresponding Rulebase
activation is deferred.
Examples The following command sets an action to activate for Rulebase rulebase1 under PCC-Action-Set configuration
instance for PCC-Service configuration:
rulebase-activate rulebase1
rulebase-deactivate
This command sets an action to deactivate a pre-configured Rulebase in PCC-Action-Set configuration for
PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to deactivate a pre-configured Rulebase from a PCC-Action-Set configuration instance
for PCC-Service configuration.
rulebase_name
Specifies the name of the pre-configured Rulebase on PCEF for a deactivation action is set under
PCC-Action-Set configuration instance for PCC-Service configuration.
rule_name is a pre-configured Rulebase in PCEF.
• SS specifies the seconds to defer the action trigger and must be an integer between 00 through 59.
Usage Guidelines Use this command to set an action to deactivate a pre-configured Rulebase in PCC-Action-Set configuration
for PCC-Service instance.
Additionally defer-by interval keyword is used to configure relative time by which the corresponding Rulebase
deactivation is deferred.
Examples The following command sets an action to deactivate for Rulebase rulebase1 under PCC-Action-Set configuration
instance for PCC-Service configuration:
rulebase-deactivate rulebase1
service-tag
This command sets an action to activate/deactivate a pre-configured Service Tag rule in PCC-Action-Set
configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to activate/deactivate a pre-configured Service Tag rule in PCC-Action-Set configuration
for PCC-Service instance.
svc_tag
Specifies the name of the pre-configured Service Tag rule for an activation/deactivation is set under
PCC-Action-Set configuration instance for PCC-Service configuration.
svc_tag is a pre-configured Service Tag name.
activate-rule
Specifies that action to be set for Service Tag activation rule under PCC-Action-Set configuration instance
for PCC-Service configuration.
deactivate-rule
Specifies that action to be set for Service Tag deactivation rule under PCC-Action-Set configuration instance
for PCC-Service configuration.
Usage Guidelines Use this command to set an action to activate/deactivate a pre-configured Service Tag rule in PCC-Action-Set
configuration for PCC-Service instance.
Examples The following command sets the an activation rule for Service Tag named service_1 under PCC-Action-Set
configuration instance for PCC-Service configuration:
service-tag service_1 activate-rule
terminate-session
This command sets an action to terminate a Bearer based on bearer-id or unique combination of QCI and
ARP received through current CCR message in PCC-Action-Set configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to terminate a Bearer based on bearer-id from a PCC-Action-Set configuration instance
for PCC-Service configuration.
Usage Guidelines Use this command to set an action to terminate a Bearer based on bearer-id or unique combination of QCI
and ARP received through current CCR message in PCC-Action-Set configuration for PCC-Service instance.
Termination of Bearer is possible only in case of PCRF binding and limited to Dynamic rules.
Important This action is only applicable to IP-CAN sessions with access type as GPRS. When terminate Bearer is
initiated, IPCF triggers Bearer Termination Procedure for the bearer-id received through current CCR
message.
Caution This command triggers termination of Gx and SPR sub session active under PCC-Service instance.
Examples The following command sets an action to terminate a Bearer based on bearer-id received through current
CCR message under PCC-Action-Set configuration instance for PCC-Service configuration:
terminate-session
usage-monitor
This command sets an action to allow operator to stop, reset or start the counting for a PCC-Usage-Monitor
in PCC-Action-Set configuration for PCC-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Action-Set
Configuration
configure > context context_name > pcc-service service_name > action-set action_set_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-action-set)#
no
Removes the action to allow operator to stop, reset or start the counting for a PCC-Usage-Monitor in
PCC-Action-Set configuration for PCC-Service instance.
usage_mon_name
Specifies the name of the pre-configured PCC-Usage-Monitor for which counter action is configured under
PCC-Action-Set configuration instance for PCC-Service configuration.
usage_mon_name is a pre-configured PCC-Usage-Monitor.
reset-counter
Resets the usage counts for PCC-Usage-Monitor under PCC-Action-Set configuration instance for PCC-Service
configuration.
start-counter
Starts the accumulation of usage counts for PCC-Usage-Monitor under PCC-Action-Set configuration instance
for PCC-Service configuration.
stop-counter
Stops the accumulation of usage counts for PCC-Usage-Monitor under PCC-Action-Set configuration instance
for PCC-Service configuration.
Usage Guidelines Use this command to set an action to allow operator to stop, reset or start the counting for a usage monitor in
PCC-Action-Set configuration for PCC-Service instance.
Examples The following command sets the an action to stop the accumulation of usage counts for PCC-Usage-Monitor
named usage_1 under PCC-Action-Set configuration instance for PCC-Service configuration:
usage-monitor usage_1 stop-counter
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-AF-Service Configuration mode provides a mechanism to IPCF to manage the external interfaces
required for media and application function management. The PCC-AF-Service manages Rx interface which
would be based on the dictionary used.
Command Modes Exec > Global Configuration > Context Configuration > PCC AF Service Configuration
configure > context context_name > pcc-af-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsapp-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate pcc-service
This command is used to associate a pre-configured PCC-Service with a PCC-AF-Service for IPCF
configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC AF Service Configuration
configure > context context_name > pcc-af-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsapp-service)#
no
Removes/disassociate the configured PCC-service from this PCC-AF-Service instance configured for IPCF
configuration.
pcc_svc_name
Specifies the name of a pre-configured PCC-service configured in Context Configuration mode for IPCF
configuration.
The pcc_svc_name is name of a predefined PCC-Service instance and must be an alphanumerical string from
1 through 63 characters.
Usage Guidelines Use this command to associate a pre-configured PCC-Service instance for IPCF configuration.
Important For more information on PCC-Service configuration, refer PCC-Service Configuration Mode Commands.
diameter dictionary
This command is used to assign a Diameter dictionary for Rx messaging with a PCC-AF-Service for IPCF
configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC AF Service Configuration
configure > context context_name > pcc-af-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsapp-service)#
default
Sets the Diameter Rx dictionary to default dictionary r8-standard (3GPP Rel. 8 standard) for a PCC-AF-Service
instance configured for IPCF configuration.
r8-standard
Default: Enabled
Sets the Diameter Rx dictionary to be used by a PCC-AF-Service instance configured for IPCF configuration
over Rx interface to 3GPP Rel. 8 standard.
standard
Default: Disabled
Sets the Diameter Rx dictionary to be used by a PCC-AF-Service instance configured for IPCF configuration
over Rx interface to 3GPP Rel. 7 standard.
Usage Guidelines Use this command to configure the PCC-AF-Service to determine which of the 3GPP dictionary to be used
for Rx interface messaging for media and application function management.
Examples Following command sets the PCC-AF-Service to use 3GPP Rel. 8 standard dictionary for Rx interface and
application function management related messaging in a PCC-AF-Service.
default diameter dictionary
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC AF Service Configuration
configure > context context_name > pcc-af-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsapp-service)#
no
Removes the associated Diameter Origin Endpoint configuration from PCC-AF-Service instance configured
for IPCF configuration.
any
Sets the PCC-AF-Service instance to use any available AF node over Rx interface for AF support.
dia_endpoint_name
The dia_endpoint_name is a predefined Diameter origin endpoint node and must be an alphanumerical string
from 1 through 63 characters.
Usage Guidelines Use this command to bind the AF node over Rx interface by associating a pre-configured Diameter Origin
Endpoint with a PCC-AF-Service.
The Diameter origin endpoint must be a pre-configured instance in the Context Configuration Mode. For
more information on Diameter origin endpoint configuration, refer Diameter Endpoint Configuration Mode
Commands chapter.
Examples Following command associates a pre-configured Diameter endpoint node configuration named af_pcscf1 with
a PCC-AF-Service for AF profile management.
diameter origin endpoint af_pcscf1
Following command removes the pre-associated Diameter endpoint node configuration named af_pcscf1 with
a PCC-AF-Service.
no diameter origin endpoint
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Condition-Group Configuration Mode is used to configure the various rating parameters under a
logical identifier name in the PCC-Service. A PCC-Condition-Group is a collection of conditions that identify
a network or state constraint represented as a logical expression. A maximum of 128 conditions can be
configured in one PCC-Condition-Group.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
af-application-id
This command defines a condition based on the application id of an Application Function service through the
Rx interface over which the IPCF receives media information for the application usage in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the AF-Application id.
operator must be one of the following:
• !=: Does not equal
• =: Equals
af_app_id
Specifies the identity string for AF-Application identifier over Rx interface.
af_app_id must be an alphanumeric string of 1 to 256 characters.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with application id of an Application Function service
condition validation in an IP-CAN session.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with application id of an Application
Function service condition validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the application id of an Application Function service through
the Rx interface over which the IPCF receives media information for the application usage in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the AF
Application id is not equal to pcc_af_1:
af-application-id != pcc_af_1
af-media-codec
This command defines a condition based on the media Codec used by AF application in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] af-media-coded operator value {g722 | g726 | ilbc | pcma | pcmu}
[ no ] af-media-coded operator profile-attribute spr_attr_value
[ no ] af-media-coded operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the AF-media Codec.
operator must be one of the following:
• !=: Does not equal
• =: Equals
g722
Specifies the media Codec as G.722 used for user traffic from AF server over Rx interface.
G.722 is a ITU-T standard 7 kHz wideband speech codec operating at 48, 56 and 64 kbit/s.
g726
Specifies the media codec as G.726 used for user traffic from AF server over Rx interface.
G.726 is an ITU-T ADPCM speech codec standard covering the transmission of voice at rates of 16, 24, 32,
and 40 kbit/s.
ilbc
Specifies the media codec as Internet Low Bitrate Codec (iLBC) used for user traffic from AF server over
Rx interface.
iLBC is a narrowband speech codec suitable for VoIP applications, streaming audio, archival and messaging.
The encoded blocks encapsulated in Real-time Transport Protocol (RTP) protocol for transport.
pcma
Specifies the media codec as Pulse Code Modulation A-law scaling (PCMA) used for user traffic from AF
server over Rx interface.
PCMA is an ITU-T Recommendation G.711 audio data encoding in eight bits per sample, after A-law
logarithmic scaling.
pcmu
Specifies the media codec as Pulse Code Modulation mu-law scaling (PCMU) used for user traffic from AF
server over Rx interface.
PCMU is an ITU-T Recommendation G.711 audio data encoding in eight bits per sample, after mu-law
logarithmic scaling.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with application media codec of an Application Function
service condition validation in an IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with application media codec of an
Application Function service condition validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the media codecs used by AF service through the Rx interface
over which the IPCF receives media information for the application usage in an IP-CAN session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the AF
Media codec is equal to PCMA:
af-media-codec = pcma
af-media-type
This command defines a condition based on the media type used by AF application in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] af-media-type operator value {application | audio | control | data | message | other | text
| video}
[ no ] af-media-type operator profile-attribute spr_attr_value
[ no ] af-media-type operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the media type in user traffic.
operator must be one of the following:
• !=: Does not equal
• =: Equals
application
Specifies the media type as Application in user traffic from AF server over Rx interface.
audio
Specifies the media type as Audio in user traffic from AF server over Rx interface.
control
Specifies the media type as Control in user traffic from AF server over Rx interface.
data
Specifies the media type as Data in user traffic from AF server over Rx interface.
message
Specifies the media type as Message in user traffic from AF server over Rx interface.
text
Specifies the media type as Text in user traffic from AF server over Rx interface.
video
Specifies the media type as Video in user traffic from AF server over Rx interface.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with application media type of an Application Function
service condition validation in an IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with application media type of an Application
Function service condition validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the media type used by AF service through the Rx interface
over which the IPCF receives media information for the application usage in an IP-CAN session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the AF
Media type is equal to Video:
af-media-type = video
af-service-urn
This command defines a condition based on the service Uniform Resource Names (URNs) used by AF
application in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• <=: Lesser than or equals
• =: Equals
• >=: Greater than or equals
urn_string
Specifies the URN in user traffic from AF server over Rx interface.
urn_string must be an alphanumeric string of 1 through 256 characters.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with URN used by AF service condition validation in an
IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with URN used by AF service condition
validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the URN used by AF service through the Rx interface over
which the IPCF receives media information for the application usage in an IP-CAN session.
Uniform Resource Names (URNs) serves as persistent, location-independent resource identifiers and are
designed to make it easy to map other namespaces into URN-space.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the AF
service URN is equal to ietf:rfc:4003:
af-service-urn = ietf:rfc:4003
an-gw-address
This command defines a condition based on the IP address of Access Node Gateway (AN-GW) on which
subscriber is attached in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
in-range
Specifies the condition to accept the IP address which are in range as argument to define the Access Node
Gateway address which is used by subscriber in IP-CAN session.
!in-range
Specifies the condition to accept the IP address which are NOT in range as argument to define the Access
Node Gateway address which is used by subscriber in IP-CAN session.
range_start_ip
Specifies the starting IP address which is used for defining the range of AN-GW IP addresses which is used
by subscriber in IP-CAN session.
range_start_ip is an IP address and must be lesser than end_start_ip address.
range_end_ip
Specifies the ending IP address which is used for defining the range of AN-GW IP addresses which is used
by subscriber in IP-CAN session.
range_end_ip is an IP address and must be greater than end_start_ip address.
angw_ip/mask
Specifies the IP address of the Access Node Gateway which is used by subscriber in IP-CAN session along
with IP mask as well.
angw_ip must be an IP address in IPv4 or IPv6 notation.
mask Specifies the IP address mask bits to determine the number of IP addresses of AN-GW in condition.
mask must be specified using the standard IPv4 dotted decimal notation.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with AN gateway IP address condition validation in an
IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with AN gateway IP address condition
validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the AN-GW IP address used by subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the condition based on AN-GW IP address
1.2.3.4 in an IP-CAN session:
an-gw-address = 1.2.3.4
authorized-qci
This command defines a condition based on the authorized QoS Class Identifier used in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• <=: Lesser than or equals
• =: Equals
• >=: Greater than or equals
qci
Specifies the authorized QoS class identifier in user traffic for condition validation.
qci must be an integer between 1 through 255.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with authorized QoS class identifier condition validation
in an IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with authorized QoS class identifier condition
validation in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the authorized QCI used by the subscriber in an IP-CAN
session.
The condition evaluates to TRUE, if Bearer Control Mode is UE and network (PCEF bearer binding) and
IPCF has authorized the specified QCI earlier.
QCI is a number which describes the error rate and delay that are associated with the service. It includes bearer
parameters including scheduling weights and queue management thresholds.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the authorized
QCI is greater than or equal to 4:
authorized-qci >= 4
base-station-id
This command defines a condition based on the various parameters used in base-station id by subscriber in
an IP-CAN session between PCEF (PDSN) and IPCF over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [no] base-station-id {[sid operator sys_identifier] | range start_range to end_range]] [nid operator
netwrk_identifier ] | range start_range to end_range ]] [ cellid operator cell_identifier ] | range
start_range to end_range ]]}
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• <: Lesser than
• <=: Lesser than or equals
• =: Equals
• >: Greater than
• >=: Greater than or equals
Usage Guidelines User this command to define a condition based on the parameters used in base-station id of subscriber, which
is composition of SID, Cell Id, and/or Network Id, in an IP-CAN session between PCEF (PDSN) and IPCF
over Gx interface.
This condition is defined for PCC functionality support to CDMA users over Gx interface.
Examples The following command creates a condition definition to analyze the condition based on base-station id of
subscriber where SID is 1001 and cell id is in range of 2001 to 2069 in an IP-CAN session between PDSN
and IPCF having network id as 3989:
base-station-id sid = 1001 nid = 3989 cellid = range 2001 to 2069
bearer-count
This command defines a condition based on the number of bearers allowed in a subscriber session on IPCF
to accept bearer-count in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• <=: Lesser than or equals
• =: Equals
• >=: Greater than or equals
num_bearer
Specifies the number of bearers in user traffic for condition validation.
num_bearer indicates a condition based on the number of bearer established in a IP-CAN session and must
be an integer between 1 through 11.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with number of bearers allowed for a subscriber in an
IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with number of bearers allowed for a
subscriber in an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the number of bearers allowed for a subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for bearers
allowed in a subscriber session is equal to 4:
bearer-count = 4
connectivity-access-network
This command defines a condition based on the access network type used by subscriber in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] connectivity-access-network operator value {3gpp-gprs | 3gpp2 | 3gpp2-eps | docsis | non-3gpp-eps
| wimax | xdsl }
[ no ] connectivity-access-network operator profile-attribute spr_attr_value
[ no ] connectivity-access-network operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the media type in user traffic.
operator must be one of the following:
• !=: Does not equal
• =: Equals
3gpp-gprs
Specifies the network access connectivity type as 3GPP-GPRS for user traffic in condition definition.
3gpp2
Specifies the network access connectivity type as 3GPP2 (CDMA) for user traffic in condition definition.
3gpp2-eps
Specifies the network access connectivity type as 3GPP2-EPS for user traffic in condition definition.
docsis
Specifies the network access connectivity type as Data Over Cable Service Interface Specification (DOCSIS)
for user traffic in condition definition.
non-3gpp-eps
Specifies the network access connectivity type as non-3GPP-EPS to connect with Gxa based HSGW for user
traffic in condition definition.
wimax
Specifies the network access connectivity type as Wi-MAX for user traffic in condition definition.
xdsl
Specifies the network access connectivity type as xDSL (ADSL/SDSL) or user traffic in condition definition.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with type of connectivity used for network access by
subscriber in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with type of connectivity used for network
access by subscriber in an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the type of connectivity used for network access by subscriber
in an IP-CAN session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the network
type is equal to Wi-MAX:
connectivity-access-network = wimax
end
Exits the current configuration mode and returns to the Exec mode.
Product All
eval-condition-group
This command defines a condition based on the TRUE or FALSE setting of a configured PCC-Condition-Group
for subscriber in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
cond_grp_name
Specifies the name of the configured PCC-Condition-Group which is used for evaluation with in this
PCC-Condition-Group for user traffic.
cond_grp_name must be a pre-configured PCC-Condition-Group in this configuration mode.
operator
Specifies how to logically match the media type in user traffic.
operator must be one of the following:
• !=: Does not equal
• =: Equals
FALSE
Sets the evaluation condition for specified PCC-Condition-Group to FALSE.
With this keyword system rejects all conditions defined in specific PCC-Condition-Group and match the same
in user traffic.
TRUE
Sets the evaluation condition for specified PCC-Condition-Group to TRUE.
With this keyword system accepts all conditions defined in specific PCC-Condition-Group and match the
same in user traffic.
Usage Guidelines Use this command to define a condition based on the acceptance or rejection of specific pre-configured
PCC-Condition-Group for user traffic in an IP-CAN session.
This command allows the operator to use configured PCC-Condition-Group in another PCC-Condition-Group
as subset of PCC-Condition-Group.
Examples The following command creates a condition definition to accept the all conditions defined in
PCC-Condition-Group named af_1:
eval-condition-group af_1 = TRUE
event-time
This command defines a condition based on the event-trigger time as per the time specified by the named
Time definition (Timedef) configured for subscriber session in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• <=: Lesser than or equals
• =: Equals
• >=: Greater than or equals
timedef_name
Specifies the name of the configured time definition (start and end timer of session) which is used for evaluation
with this PCC-Condition-Group for user traffic.
timedef_name must be a pre-configured Time definition in PCC-Service Configuration Mode.
Usage Guidelines Use this command to define a condition based on the start and end time defined in a Time Definition
configuration in PCC-Service Configuration Mode for user traffic in an IP-CAN session.
Event trigger in this command indicates a condition when the event-trigger time was as per the time specified
by the named Time Definition.
This command allows the operator to use configured time period as event trigger for this PCC-Condition-Group.
Examples The following command creates a condition definition to trigger the condition based on the start and end time
defined in Timedef named timedef_night:
event-time = timedef_night
event-trigger
This command defines a condition based on the event triggers due to various conditions for subscriber in an
IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the media type in user traffic.
operator must be one of the following:
• !=: Does not equal
• =: Equals
event_trigger
Specifies the network events as condition definition for user traffic in an IP-CAN session.
Following event triggers are supported with this command:
• an-gw-change
• bearer-qos-change
• bearer-setup
• bearer-termination
• default-eps-bearer-qos-change
• ip-can-change
• loss-of-bearer
• out-of-credit
• pgw-trace-control
• plmn-change
• qos-change
• qos-change-exceeding-authorization
• rai-change
• rat-change
• reallocation-of-credit
• recover-of-bearer
• resource-modification-request
• revalidation-timeout
• session-setup
• session-termination
• sgsn-change
• successful-resource-allocation
• tft-change
• ue-ip-address-allocate
• ue-ip-address-release
• ue-time-zone-change
• user-location-change
Usage Guidelines Use this command to define a condition based on the event triggered in network for user traffic.
Examples The following command creates a condition definition when there is a change in traffic flow template for
subscriber:
event-trigger = tft-change
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
imsi
This command defines a condition based on the International Mobile Station Identification number (IMSI)
of a subscriber in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
imsi
Specifies the IMSI of subscriber to be used for condition validation in an IP-CAN session.
imsi must be a string of between 8 to 15 digits which starts with 3 digit of MCC then 2 to 3 digit of MNC.
Wildcard characters $ and * are allowed. The * wildcard matches multiple characters and the $ wildcard
matches a single character. If you do not want the wildcard characters interpreted as a wildcard enclose them
in single quotes ( ' ). For example; '$'.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with IMSI in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with IMSI in an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the IMSI of a subscriber used in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on subscriber IMSI as
12334566434 in an IP-CAN session:
imsi = 12334566434
msisdn
This command defines a condition based on the Mobile Station International Subscriber Directory Number
(MSISDN) of a subscriber in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
imsi
Specifies the MSISDN of subscriber to be used for condition validation in an IP-CAN session.
msisdn must be a string of between 1 to 16 digits which contains CC + NDC/NPA + SN.
Wildcard characters $ and * are allowed. The * wildcard matches multiple characters and the $ wildcard
matches a single character. If you do not want the wildcard characters interpreted as a wildcard enclose them
in single quotes ( ' ). For example; '$'.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with MSISDN in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with MSISDN in an IP-CAN session as
condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the MS ISDN of a subscriber used in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on subscriber MSISDN
as 380561234567 in an IP-CAN session:
msisdn = 380561234567
multi-line-or
This command enables/disables the "OR" relation across all conditions exist in PCC-Condition-Group
Configuration Mode.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Sets the PCC-Condition-Group to "AND" relation across all conditions exist in a PCC-Condition-Group
Configuration Mode.
Usage Guidelines Use this command to enable the "OR" relation across all conditions exist in PCC-Condition-Group Configuration
Mode.
In absence of this command the default relation of "AND" applies across all conditions exist in
PCC-Condition-Group Configuration Mode.
Examples The following command enables the "OR" relation across all conditions exist in PCC-Condition-Group
Configuration Mode:
multi-line-or
The following command enables the default "AND" relation across all conditions exist in PCC-Condition-Group
Configuration Mode:
no multi-line-or
nai
This command defines a condition based on the Network Access Identifier (NAI) of a subscriber in an IP-CAN
session between PCEF (PDSN) and IPCF over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] nai operator {username user_name [domain domain] | domain domain}
[ no ] nai operator profile-attribute spr_attr_value
[ no ] nai operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
username user_name
Specifies the subscriber user name to be used for condition validation in an IP-CAN session.
user_name must be an alpha and/or numeric string of 1 through 128 characters in length. The user name can
contain all special characters.
domain domain
Specifies the domain (Realm) of subscriber to be used for condition validation in an IP-CAN session.
domain must be an alpha and/or numeric string of 1 through 128 characters in length. The domain name can
contain all special characters.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with NAI in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with NAI in an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines User this command to define a condition based on the NAI, which contains of user name and domain name,
of a subscriber in an IP-CAN session between PCEF (PDSN) and IPCF over Gx interface.
This condition is defined for PCC functionality support to CDMA users over Gx interface.
Examples The following command creates a condition definition to analyze the condition based on subscriber user name
as cdma2000_subs1 in an IP-CAN session between PDSN and IPCF having xyz.com as domain:
nai = username cdma2000_subs1 domain xyz.com
out-of-credit rulename
This command defines a condition based on the exhaustion of credit for subscriber Rulename at PCEF in an
IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
rule_name
Specifies the name of the Rulename for which out of credit condition is to match in an IP-CAN session.
rule_name is a pre-configured Rulename on PCEF and must be an alphanumeric string of from 1 through 63
characters.
Usage Guidelines Use this command to define a condition based on the exhaustion of credit for subscriber Rulename at PCEF
in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on the exhaustion of
credit for subscriber Rulename rule_101 at PCEF in an IP-CAN session:
out-of-credit rulename = rule_101
out-of-credit rulebase-name
This command defines a condition based on the exhaustion of credit for subscriber Rulebase name at PCEF
in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
rulebase_name
Specifies the name of the Rulebase for which out of credit condition is to match in an IP-CAN session.
rulebase_name is a pre-configured Rulebase on PCEF and must be an alphanumeric string of from 1 through
63 characters.
Usage Guidelines Use this command to define a condition based on the exhaustion of credit for subscriber Rulebase name at
PCEF in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on the exhaustion of
credit for subscriber Rulebase name rulebase_101 at PCEF in an IP-CAN session:
out-of-credit rulebase-name = rulebase_101
pcef-address
This command defines a condition based on the IP address of Policy and Charging Enforcement Function
(PCEF) which is served by IPCF and through which subscriber is attached to an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] pcef-address {operator value pcef_ip/mask | {in-range | !in-range} value range_start_ip to
range_end_ip}
[ no ] pcef-address operator profile-attribute spr_attr_value
[ no ] pcef-address operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
in-range
Specifies the condition to accept the IP address which are in range as argument to define the PCEF address
which is used by subscriber in IP-CAN session.
!in-range
Specifies the condition to accept the IP address which are NOT in range as argument to define the PCEF
address which is used by subscriber in IP-CAN session.
range_start_ip
Specifies the starting IP address which is used for defining the range of PCEF IP addresses which is used by
subscriber in IP-CAN session.
range_start_ip is an IP address and must be lesser than end_start_ip address.
range_end_ip
Specifies the ending IP address which is used for defining the range of PCEF IP addresses which is used by
subscriber in IP-CAN session.
range_end_ip is an IP address and must be greater than end_start_ip address.
pcef_ip/mask
Specifies the IP address of the PCEF which is used by subscriber in IP-CAN session along with IP mask as
well.
pcef_ip must be an IP address in IPv4 or IPv6 notation.
mask Specifies the IP address mask bits to determine the number of IP addresses of PCEF in condition. mask
must be specified using the standard IPv4 dotted decimal notation.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with PCEF address in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with PCEF address in an IP-CAN session
as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the PCEF IP address used by subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the condition based on PCEF P address
1.2.3.4 in an IP-CAN session:
pcef-address = 1.2.3.4
pdn-id
This command defines a condition based on the PDN or calling station id in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
case-insensitive
This optional keyword sets the condition to not to consider the case of argument phrase for condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
• contains: Contains
• !contains: Does not contain
• starts-with: Starts with
• !starts-with: Does not start with
• ends-with: Ends with
pdn_id
Specifies the PDN or calling station id to be used for condition validation in an IP-CAN session.
pdn_id must be an alphanumeric string of between 1 to 128 characters.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with PDN id in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with PDN id in an IP-CAN session as
condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the PDN or calling station id used in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on calling station id
ggsn_all_102 in an IP-CAN session:
pdn-id = ggsn_all_102
profile-attribute
This command defines a condition based on the matching between subscriber profile attribute value and SPR
attribute value in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
parameter subs_prof_attr_value
Specifies the attribute parameter value to match with SPR attribute for condition validation in an IP-CAN
session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
spr_attr_value
Specifies the attribute value in SPR to match with Subscriber profile attribute for condition validation in an
IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
Usage Guidelines Use this command to define a condition based on the matching parameters between subscriber profile attribute
and SPR attribute parameters used in an IP-CAN session and provides a generic way of specifying and
triggering actions based on any of the subscriber profile attribute received from SPR interactions.
Examples The following command creates a condition definition to analyze the condition based on matching of subscriber
profile attribute value subs_gold_102 with SPR attribute value spr_gold_102in an IP-CAN session:
profile-attribute parameter subs_gold_102 = spr_gold_102
radio-access-technology
This command defines a condition based on the radio access technology used by subscriber in an IP-CAN
session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the media type in user traffic.
operator must be one of the following:
• !=: Does not equal
• =: Equals
RAT
Specifies the radio access technology used by subscriber to access the network as condition definition for user
traffic in an IP-CAN session.
Following RAT are supported with this command:
• cdma2000-1x: 3GPP2 CDMA 2000 - 1x RTT
• eutran: Evolved Universal Terrestrial Radio Access Network (eUTRAN)
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with type of RAT in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with type of RAT in an IP-CAN session
as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the type of radio technology used for network access by
subscriber in an IP-CAN session.
Examples The following command creates a condition definition to analyze the PCC service user traffic for the radio
access technology type as WLAN:
radio-access-technology = wlan
sgsn-ip
This command defines a condition based on the IP address of SGSN on which subscriber is attached in an
IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] sgsn-ip {operator value sgsn_ip/mask | {in-range | !in-range} range_start_ip to range_end_ip}
[ no ] sgsn-ip operator profile-attribute spr_attr_value
[ no ] sgsn-ip operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
in-range
Specifies the condition to accept the IP address which are in range as argument to define the SGSN address
which is used by subscriber in IP-CAN session.
!in-range
Specifies the condition to accept the IP address which are NOT in range as argument to define the SGSN
address which is used by subscriber in IP-CAN session.
range_start_ip
Specifies the starting IP address which is used for defining the range of SGSN IP addresses which is used by
subscriber in IP-CAN session.
range_start_ip is an IP address and must be lesser than end_start_ip address.
range_end_ip
Specifies the ending IP address which is used for defining the range of SGSN IP addresses which is used by
subscriber in IP-CAN session.
range_end_ip is an IP address and must be greater than end_start_ip address.
sgsn_ip/mask
Specifies the IP address of the SGSN which is used by subscriber in IP-CAN session along with IP mask as
well.
sgsn_ip must be an IP address in IPv4 or IPv6 notation.
mask Specifies the IP address mask bits to determine the number of IP addresses of SGSN in condition. mask
must be specified using the standard IPv4 dotted decimal notation.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with SGSN IP address in an IP-CAN session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with SGSN IP address in an IP-CAN session
as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the SGSN IP address used by subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the condition based on SGSN IP address
1.2.3.4 in an IP-CAN session:
sgsn-ip = 1.2.3.4
sgsn-mcc-mnc
This command defines a condition based on the PLMN (MCC+MNC) of SGSN on which subscriber is attached
in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] sgsn-mcc-mnc operator {mcc mcc_value [mnc mnc_value] | [mcc mcc_value] mnc mnc_value}
[ no ] sgsn-mcc-mnc operator profile-attribute spr_attr_value
[ no ] sgsn-mcc-mnc operator subscription-attribute subs_prof_attr_value
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
mcc mcc_value
Specifies the Mobile Country Code in PLMN of SGSN which is used by subscriber in IP-CAN session.
mcc_value must be an integer between 101 and 998.
mnc mnc_value
Specifies the Mobile Network Code in PLMN of SGSN which is used by subscriber in IP-CAN session.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with PLMN of SGSN used by subscriber in an IP-CAN
session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with PLMN of SGSN used by subscriber
in an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the PLMN of SGSN used by subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the condition based on PLMN of SGSN
with MCC as 102 and MNC as 99in an IP-CAN session:
sgsn-mcc-mnc = mcc 102 mnc 99
subscription-attribute
This command defines a condition based on the matching between subscriber subscription attribute value and
SPR attribute value in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
subsription_attr_value
Specifies the subscriber subscription attribute value to match with SPR attribute for condition validation in
an IP-CAN session.
subsription_attr_value must be an alphanumeric string of from 1 through 31 characters.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
spr_attr_value
Specifies the attribute value in SPR to match with subscriber subscription attribute for condition validation
in an IP-CAN session.
Usage Guidelines Use this command to define a condition based on the matching parameters between subscriber subscription
attribute and SPR attribute value used in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on matching of subscriber
subscription attribute value subs_gold_102 with SPR attribute value spr_gold_101in an IP-CAN session:
subscription-attribute subscribe_gold_102 = spr_gold_101
spr-profile-not-found
This command defines a condition based on the availability of SPR profile for s subscriber in an IP-CAN
session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
Usage Guidelines Use this command to define a condition based on the availability of SPR profile for a subscriber in an IP-CAN
session.
Examples The following command creates a condition definition to analyze the condition if SPR profile is not available
for a subscriber in an IP-CAN session:
spr-profile-not-found
threshold-condition usage-monitor
This command defines a condition based on the threshold conditions in usage of traffic by subscriber session
in an IP-CAN session. It is used to support usage tracking and dynamic Policy control based on subscriber
usage.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [ no ] threshold-condition usage-monitor usage_mon_name { time | vol-or-time | volume } usage operator
{ absolute value volume | subscription-limit | subscription-threshold subs_thres_limit}
no
Removes the specified condition definition.
usage_mon_name
Specifies a unique name for configured usage monitor condition which is used for evaluation with this condition
group for user traffic in IP-CAN session.
usage_mon_name must be an alphanumeric string of 1 through 63 characters.
usage
This keyword specifies the condition as usage for threshold condition which is used for evaluation with this
condition group for user traffic in IP-CAN session.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
subscription-limit
Specifies the threshold limit condition based on Subscriber's subscription limit which is used for evaluation
with this condition group for user traffic in IP-CAN session.
In this condition the system takes Subscriber's subscription limit as defined in subscriber subscription policy
for threshold limit monitoring.
subscription-threshold subs_thres_limit
Specifies the condition definition for threshold limit based on a configured usage monitor threshold named
subs_thres_limit in IP-CAN session.
subs_thres_limit is a pre-configured subscription limit in this configuration mode.
This command allows the operator to use configured subscription limit as subset of a threshold condition.
Usage Guidelines Use this command to define a condition based on the duration of usage of service in seconds or volume usage
in Bytes for user traffic in an IP-CAN session. This usage monitor is used to support usage tracking and
dynamic Policy control based on subscriber usage.
IPCF supports the concept of Monitoring Key. PCEF, when instructed by PCRF, keeps track of usage per
Monitoring Key. PCEF reports the usage when thresholds are reached or requested by PCRF.
To allow operator to have dynamic Policy control, IPCF uses Usage Monitor. This Usage Monitor has attribute
of volume-limit, time-limit or both. Operator can "associate" different Monitoring Keys to these usage monitors.
It can be a Many-To-Many relationship between Usage-Monitor and Monitoring-Key. Operator can use Usage
conditions on Usage Monitors instead of Monitoring Keys directly.
IPCF tracks usage per usage-monitor. Different monitoring keys associated to a Usage-Monitor. When
monitoring key is associated to a Usage-Monitor, the usage reported for that particular monitoring key is
added to all the usage monitoring to which it is associated.
IPCF supports 2 types of usage monitors; it can be defined on IPCF or SSC supplied usage monitors. The
name usage_mon_name identifies an usage-monitor, which is locally defined or received from SSC. The
absolute value applies to both types of usage monitors.
This command allows the operator to use configured threshold condition with another threshold condition as
subset.
A maximum of 8 thresholds can be configured per usage monitor.
Examples The following command creates a threshold condition with usage monitor name threshold1 as usage monitor
to trigger the condition based on the subscription limit as provide in Subscriber policy:
threshold-condition usage-monitor threshold1 usage = subscription-limit
user-access-network
This command defines a condition based on the access location type of the subscriber in an IP-CAN session
as received on Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
value home
Specifies the subscriber network access type as Home for condition validation in an IP-CAN session.
This condition contains all subscribers active in Home networks in this condition.
value roaming
Specifies the subscriber network access type as Roaming for condition validation in an IP-CAN session.
This condition contains all subscribers active in Roaming networks in this condition.
value visiting
Specifies the subscriber network access type as Visiting for condition validation in an IP-CAN session.
This condition contains all subscribers active in Visiting networks in this condition.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with user access network condition validation in an
IP-CAN session.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with user access network condition validation
in an IP-CAN session.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the location of the user as defined in SPR attribute or
subscriber profile attribut or location of user in Home, Roaming, Visiting in an IP-CAN session as received
on Gx interface.
This configuration compares ULI and other location information from Gx data with the Home/roaming/visiting
location as received from SSC. IPCF will also use global data as received from SSC in determining user access
network through Subscriber profile attribute or SPR attribute in profile attribute.
Examples The following command creates a condition definition to analyze the condition based on location of the
subscribes as roaming in an IP-CAN session:
user-access-network = value roaming
user-equipment-info esn
This command defines a condition based on the Electronic Serial Number (ESN) used for the identification
of mobile device (UE) in a Non-3GPP IP-CAN session received over Gx interface between PDSN and IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
esn
Specifies the Electronic Serial Number used for the identification of UE in 64 bit format as condition value
for UE information received over Gx interface in a Non-3GPP IP-CAN session.
esn must be a 15 character long string of Hexadecimal numbers only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with ESN of the UE in an IP-CAN session as condition.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with ESN of the UE in an IP-CAN session
as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the ESN of the UE received over Gx interface in a Non-3GPP
IP-CAN session between PDSN and IPCF.
Examples The following command creates a condition definition to analyze the condition based on the ESN of the UE
as 1234567890120AF in a Non-3GPP IP-CAN session:
user-equipment-info esn = 1234567890120AF
user-equipment-info eui64
This command defines a condition based on the Extended Unique Identifier in 64 bit (EUI-64) used for the
identification of mobile device (UE) in an IP-CAN session received over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
eui64
Specifies the Extended Unique Identifier in 64 bit format as a condition value for UE information received
over Gx interface in an IP-CAN session.
eui64 is a 16 character long string of Hexadecimal numbers only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with EUI-64 value of the UE in an IP-CAN session as
condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with EUI-64 value of the UE in an IP-CAN
session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the EUI-64 value of the UE received over Gx interface in
an IP-CAN session between PCEF and IPCF.
Examples The following command creates a condition definition to analyze the condition based on the EUI-64 of the
UE as 0012989099832345 in an IP-CAN session:
user-equipment-info eui64 = 0012989099832345
user-equipment-info imeisv
This command defines a condition based on the International Mobile Equipment Identity Software Version
(IMEI-SV) used for the identification of mobile device (UE) in an IP-CAN session received over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
imei_sv
Specifies the International Mobile Equipment Identity Software Version (IMEI-SV) as a condition value for
UE information received over Gx interface in an IP-CAN session.
imei_sv is a 16 digit long string of decimal numbers only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with IMEI-SV value of the UE in an IP-CAN session as
condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with IMEI-SV value of the UE in an IP-CAN
session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the IMEI-SV value of the UE received over Gx interface
in an IP-CAN session between PCEF and IPCF.
Examples The following command creates a condition definition to analyze the condition based on the IMEI-SV of the
UE as 0012989099832345 in an IP-CAN session:
user-equipment-info eui64 = 0012989099832345
user-equipment-info mac
This command defines a condition based on the Media Access Control (MAC) address used for the UE
information in an IP-CAN session as received over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
mac_value
Specifies the Media Access Control (MAC) address as condition value for UE information received over Gx
interface in an IP-CAN session.
mac_value is a 17 character long string of Hexadecimal numbers in xx:xx:xx:xx:xx:xx format only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with MAC value of the UE in an IP-CAN session as
condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with MAC value of the UE in an IP-CAN
session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the MAC address value as UE information received over
Gx interface in an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on the MAC value of
the UE as 00:12:98:90:99:83 in an IP-CAN session:
user-equipment-info mac = 00:12:98:90:99:83
user-equipment-info meid
This command defines a condition based on the Mobile Equipment Id (MEID) used for the identification of
CDMA mobile device (UE) in a Non-3GPP IP-CAN session received over Gx interface between PDSN and
IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
me_id
Specifies the Mobile Equipment identifier used for the identification of UE in 64 bit format as condition value
for UE information received over Gx interface in a Non-3GPP IP-CAN session.
me_id must be a 14 character long string of Hexadecimal numbers only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with Mobile Equipement id of the UE in an IP-CAN
session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with Mobile Equipement id of the UE in
an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the MEID of the CDMAUE received over Gx interface in
a Non-3GPP IP-CAN session between PDSN and IPCF.
Examples The following command creates a condition definition to analyze the condition based on the MEID of the
CDMA UE as 123456780120AF in a Non-3GPP IP-CAN session:
user-equipment-info meid = 123456780120AF
user-equipment-info modified-eui64
This command defines a condition based on the modified-Extended Unique Identifier in 64 bit (EUI-64) used
for the identification of mobile device (UE) in an IP-CAN session received over Gx interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
no
Removes the specified condition definition.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
meui64
Specifies the modified Extended Unique Identifier (EUI) in 64 bit format as condition value for UE information
received over Gx interface in an IP-CAN session.
meui64 is a 16 character long string of Hexadecimal numbers only.
profile-attributespr_attr_value
Specifies the profile attribute value in SPR to match with modified EUI-64 value of the UE in an IP-CAN
session as condition.
spr_attr_value must be an alphanumeric string of from 1 through 63 characters.
subscription-attribute subs_prof_attr_value
Specifies the subscriber profile attribute parameter value to match with modified EUI-64 value of the UE in
an IP-CAN session as condition.
subs_prof_attr_value must be an alphanumeric string of from 1 through 31 characters.
Usage Guidelines Use this command to define a condition based on the modified EUI-64 value of the UE received over Gx
interface in an IP-CAN session between PCEF and IPCF.
Examples The following command creates a condition definition to analyze the condition based on the EUI-64 of the
UE as 0012989099832345 in an IP-CAN session:
user-equipment-info modified-eui64 = 0012989099832345
user-location-info
This command defines a condition based on the UE location used in an IP-CAN session as received on Gx
interface.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Condition Group
Configuration
configure > context context_name > pcc-service service_name > condition group group_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-condition-group)#
Syntax Description [no] user-location-info {cgi operator mcc mcc_value mnc mnc_value lac lac_value ci ci_value |
ecgi operator mcc mcc_value mnc mnc_value ecgi ecgi_value | rai operator mcc mcc_value mnc
mnc_value lac lac_value rai rai_value| sai operator mcc mcc_value mnc mnc_value lac lac_value
sac sac_value| tai operator mcc mcc_value mnc mnc_value tai tai_value}
no
Removes the specified condition definition.
cgi
Specifies the Cell Global Identifier in UE location received over Gx interface in an IP-CAN session.
ecgi
Specifies the E-UTRAN Cell Global Identifier in UE location received over Gx interface in an IP-CAN
session.
rai
Specifies the Routing Area Identifier in UE location received over Gx interface in an IP-CAN session.
sai
Specifies the Service Area Identifier in UE location received over Gx interface in an IP-CAN session.
tai
Specifies the Tracking Area Identifier in UE location received over Gx interface in an IP-CAN session.
operator
Specifies how to logically match the information in the analyzed field.
operator must be one of the following:
• !=: Does not equal
• =: Equals
mcc mcc_value
Specifies the Mobile Country Code used in UE location information received over Gx interface in IP-CAN
session.
mcc_value must be an integer between 101 through 998.
mnc mnc_value
Specifies the Mobile Network Code used in UE location information received over Gx interface in IP-CAN
session.
mnc_value must be an integer between 1 through 998.
lac lac_value
Specifies the Location Area Code used in UE location information received over Gx interface in IP-CAN
session.
lac_value must be an integer between 1 through 65535.
ci ci_value
Specifies the Cell Identifier used in UE location information received over Gx interface in IP-CAN session.
ci_value must be an integer between 1 through 65535.
ecgi ecgi_value
Specifies the E-UTRAN Cell Global Identifier used in UE location information received over Gx interface
in IP-CAN session.
ecgi_value must be an integer between 1 through 1048575.
rai rai_value
Specifies the Routing Area Identifier used in UE location information received over Gx interface in IP-CAN
session.
rai_value must be an integer between 1 through 65535.
sac sac_value
Specifies the Service Area Code used in UE location information received over Gx interface in IP-CAN
session.
tai tai_value
Specifies the Tracking Area Code used in UE location information received over Gx interface in IP-CAN
session.
tai_value must be an integer between 1 through 65535.
Usage Guidelines Use this command to define a condition based on the UE location information received over Gx interface in
an IP-CAN session.
Examples The following command creates a condition definition to analyze the condition based on the RAI received
for UE location in an IP-CAN session:
user-location-info rai = mcc 102 mnc 99 lac 1003 rai 3521
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Data-Service Configuration Mode is used to configure the data flow parameters for the media data,
as well as corresponding attributes that are necessary for charging and policy enforcement decisions for the
media parameters represented by its service flows in the PCC-service. A maximum of ten service data flows
can be configured in a PCC-Data-Service instance.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
flow direction in
This command configures the flow service parameters for incoming data flow in PCC-Data-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
Syntax Description [no] flow direction in protocol {ip | tcp | udp} from {src_ip_addr[/maskbit] | any} port
{src_port_num | any} to {dest_ip_addr/maskbit | any} port {dest_port_num | any}
no
Removes the configured data service flow parameters from PCC-Data-Service instance for IPCF configuration.
• any: specifies that flow from any source IP address/port can be analyzed or considered.
• port src_port_num: specifies the specific source port parameter for flow.
src_port_num is the source port number of flow and must be an integer from 1 through 65535.
Usage Guidelines Use this command to configure the flow service parameters for incoming data flow in PCC-Data-Service
instance for IPCF Configuration.
Examples Following command sets the data service parameters for tcp type protocol from any IP/port source to any
IP/port in incoming direction with in a PCC-Data-Service.
flow direction in protocol tcp from any port any to any port any
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
Syntax Description [no] flow direction in protocol {ip | tcp | udp} from {src_ip_addr[ /maskbit] | any} port
{src_port_num | any} to {dest_ip_addr[ /maskbit] | any} port {dest_port_num | any}
no
Removes the configured data service flow parameters from PCC-Data-Service instance on IPCF node.
• any: specifies that flow from any source IP address/port can be analyzed or considered.
• port src_port_num: specifies the specific source port parameter for flow.
src_port_num is the source port number of flow and must be an integer from 1 through 65535.
Usage Guidelines Use this command to configure the flow service parameters for outgoing data flow in PCC-Data-Service
instance on IPCF node.
Examples Following command sets the data service parameters for tcp type protocol from any IP/port source to any
IP/port in outgoing direction with in a PCC-Data-Service.
flow direction out protocol tcp from any port any to any port any
metering-method
This command specifies the metering method to be used by PCEF for offline charging in a PCC-Data-Service
instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured metering method from PCC-Data-Service instance on IPCF node.
In such scenario, the metering method pre-configured at PCEF is considered.
both-duration-volume
Specifies the metering method as based on volume of data usage and duration of session, both for offline
charging in a PCC-Data-Service instance on IPCF node.
duration
Specifies the metering method as based on the duration of session usage for offline charging in a
PCC-Data-Service instance on IPCF node.
volume
Specifies the metering method as based on the volume of data usage for offline charging in a PCC-Data-Service
instance on IPCF node.
Usage Guidelines Use this command to define the metering method to be used for offline charging in a PCC-Data-Service
instance on IPCF node.
Examples Following command sets the metering method as based on volume of data usage and duration of session, both
for offline charging in a PCC-Data-Service instance on IPCF node.
metering-method both-duration-volume
monitoring-key
This command defines the monitoring key under which data is monitored for the PCC-Data-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured monitoring-key from PCC-Data-Service instance on IPCF node.
mon_key_value
Specifies the monitoring key value under which data usage is monitored for the PCC-Data-Service instance
and must be an integer from 1 through 65535.
Usage Guidelines Use this command to define the monitoring key under which data is monitored for the PCC-Data-Service
instance on IPCF node.
Examples Following command sets the monitoring key 123 for data monitoring in PCC-Data-Service instance on IPCF
node.
monitoring-key 123
precedence
This command defines the precedence that is assigned to the Dynamic PCC rule created for a PCC-Data-Service
instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured precedence value from PCC-Data-Service instance on IPCF node.
preced_value
Specifies the precedence that is assigned to the dynamic PCC rule created for a PCC-Data-Service instance
and must be an integer from 1 through 65535.
Usage Guidelines Use this command to define the precedence value that is assigned to the Dynamic PCC rule created for a
PCC-Data-Service instance on IPCF node.
Examples Following command sets the precedence value2 to assign to the dynamic PCC rule created for a
PCC-Data-Service instance on IPCF node.
precedence 2
qos-profile
This command defines the PCC-QoS-Profile which is to use for the PCC-Data-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured PCC-QoS-Profile from PCC-Data-Service instance on IPCF node.
qos_prof_name
Specifies the pre-configured PCC-QoS-Profile name which is to use for the PCC-Data-Service instance.
Usage Guidelines Use this command to define the PCC-QoS-Profile for a PCC-Data-Service instance on IPCF node.
Examples Following command sets the PCC-QoS-Profile ipcf_qos_prof1 for PCC-Data-Service instance on IPCF node.
qos-profile ipcf_qos_prof1
rating-group
This command defines the PCC-Rating-Group that is assigned for a PCC-Data-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured PCC-Rating-Group Id from PCC-Data-Service instance on IPCF node.
rating_grp_id
Specifies the PCC-Rating-Group Id that is assigned to the PCC-Data-Service instance and must be an integer
from 1 through 99.
Usage Guidelines Use this command to define the PCC-Rating Id value for a PCC-Data-Service instance on IPCF node.
Examples Following command sets the Rating Group id 11 for a PCC-Data-Service instance on IPCF node.
rating-group 11
reporting-level
This command specifies the reporting level to be used by PCEF to report the data usage for the related PCC
rule in a PCC-Data-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured usage reporting level from PCC-Data-Service instance on IPCF node.
In such scenario, the reporting level configured at PCEF is considered.
service-identifier-level
Sets the data usage reporting level to be used by PCEF to report the data usage for the related PCC rule at the
service identifier level in a PCC-Data-Service instance on IPCF node.
rating-group-level
Sets the data usage reporting level to be used by PCEF to report the data usage for the related PCC rule at the
Rating-group level in a PCC-Data-Service instance on IPCF node.
Usage Guidelines Use this command to define the reporting level to be used by PCEF to report the data usage for the related
PCC rule in a PCC-Data-Service instance on IPCF node.
If no reporting level is defined, the reporting level preconfigured at PCEF is considered.
Examples Following command sets the reporting level at 'Rating-group' level which is to be used by PCEF to report the
data usage for the related PCC rule in a PCC-Data-Service instance on IPCF node.
reporting-level rating-group-level
service-identifier
This command defines the service identifier for a PCC-Data-Service instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Data Service
Configuration
configure > context context_name > pcc-service service_name > data-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-data-service)#
no
Removes the configured Service Identifier from PCC-Data-Service instance on IPCF node.
svc_id
Specifies the Service Identifier that is assigned to the PCC-Data-Service instance and must be an integer from
1 through 65535.
Usage Guidelines Use this command to define the Service Identifier for a PCC-Data-Service instance on IPCF node.
Examples Following command sets the Service Identifier 1011 for a PCC-Data-Service instance on IPCF node.
service-identifier 1011
Important This configuration mode is supported from StarOS Release 12.1 onward.
Command Modes Exec > Global Configuration > Context Configuration > Event Notification Interface Endpoint Configuration
configure > context context_name > event-notif-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ntfyintf-endpoint)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
address
This command binds an IP address to the local IPCF node which is to be used for event notification processing
with remote event collection server endpoint during IP-CAN session in
PCC-Event-Notification-Interface-Endpoint instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > Event Notification Interface Endpoint Configuration
configure > context context_name > event-notif-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ntfyintf-endpoint)#
ip_address
Specifies the IP address bind with local IPCF node to be used by the event collection server endpoint for event
message processing during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
Usage Guidelines Use this command to bind an IP address to interact with the remote event notification collection server endpoint
to which the event messages are sent for IP-CAN session events.
Examples Following command binds the 1.2.3.4 for event notification message with remote event notification endpoint.
address1.2.3.4
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
peer name
This command binds/associates a remote Event Notification collection server as peer having specified IP
address and optionally port for event notification during IP-CAN session in
PCC-Event-Notification-Interface-Endpoint instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > Event Notification Interface Endpoint Configuration
configure > context context_name > event-notif-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ntfyintf-endpoint)#
name peer_name
Sets the in PCC-Event-Notification-Interface-Endpoint instance to use a particular peer node configured by
a peer name peer_name for event notification interface.
The peer_name is name of the peer node to be sued for event collection and must be an alphanumerical string
from 1 through 31 characters. peer_name allows punctuation marks.
ip-address ip_address
Sets the PCC-Event-Notification-Interface-Endpoint instance to bind the particular peer node name with IP
address ip_address in IPv4 or IPv6 notation for event notification message processing.
The ip_address is an IP address in IPv4/IPv6 notation.
port port_num
This optional keyword sets a particular port number to be used with in the
PCC-Event-Notification-Interface-Endpoint instance to configure a particular peer node having a pre assigned
IP address ip_address in IPv4 or IPv6 notation for event notification message processing.
The port_num must be an integer between 1 and 65535.
Usage Guidelines Use this command to bind/associate a remote Event Notification collection server as peer having specified IP
address and optionally port for event notification during IP-CAN session in the
PCC-Event-Notification-Interface-Endpoint instance.
Multiple peers can be configured using this command and peer selection methods, primary-secondary or
round-robin can be applied using peer select-algorithm command for event notification during IP-CAN
session in this configuration mode.
Examples Following command configures and associates an Event Notification peer node named event_peer_1 having
an IP address 1.2.3.4 with port number as 2345 in an PCC-Event-Notification-Interface-Endpoint instance.
peer name event_peer_1 ip-address 1.2.3.4 port 2345
peer select-algorithm
This command applies the peer selection algorithm to select the configured remote Event Notification collection
server during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > Event Notification Interface Endpoint Configuration
configure > context context_name > event-notif-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ntfyintf-endpoint)#
primary-secondary
Sets the PCC-Event-Notification-Interface-Endpoint peer selection algorithm to select the configured remote
peer servers in primary and secondary method during IP-CAN session in
PCC-Event-Notification-Interface-Endpoint instance.
This mode is applicable only when multiple peers are configured and primary and secondary peer is defined
using peer select-peer command in this configuration mode.
round-robin
Sets the PCC-Event-Notification-Interface-Endpoint peer selection algorithm to select the configured remote
peer servers in round-robin method mode during IP-CAN session in PCC-Event-Notification-Interface-Endpoint
instance.
This mode is applicable only when multiple peers are configured in this configuration mode.
Usage Guidelines Use this command apply the peer selection algorithm to select the configured remote Event Notification
collection server during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
Examples Following command configures the peer selection algorithm to select the configured remote peer in round-robin
method in PCC-Event-Notification-Interface-Endpoint instance.
peer select-algorithm round-robin
Following command configures the peer selection algorithm to select the configured primary and secondary
remote peers in primary-secondary method in PCC-Event-Notification-Interface-Endpoint instance.
peer select-algorithm primary-secondary
peer select-peer
This command sets the configured remote Event Notification collection server as primary and secondary
servers for event notification collection during IP-CAN session in PCC-Event-Notification-Interface-Endpoint
instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > Event Notification Interface Endpoint Configuration
configure > context context_name > event-notif-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ntfyintf-endpoint)#
pri_peer_name
Sets the configured remote PCC-Event-Notification-Interface-Endpoint peer as primary peer for event
notification collection during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
pri_peer_name must be a pre-configured peer name configured with peer name command in this configuration
mode.
secondary sec_peer_name
Sets the configured remote PCC-Event-Notification-Interface-Endpoint peer as secondary peer for event
notification collection during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
sec_peer_name must be a pre-configured peer name configured with peer name command in this configuration
mode.
Usage Guidelines Use this command to set the configured remote Event Notification collection server as primary and secondary
node for event notification collection during IP-CAN session in PCC-Event-Notification-Interface-Endpoint
instance.
This configuration is used when peer selection algorithm is set to primary-secondary using peer
select-algorithm command in this configuration mode.
Examples Following command configures the specified peer event1 as primary and event2 as secondary node for event
notification collection during IP-CAN session in PCC-Event-Notification-Interface-Endpoint instance.
peer select-peer event1 secondary event2
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Policy-Service Configuration mode provides a mechanism for the Intelligent Policy Control
Function (IPCF) to manage the external interfaces required for policy authorization purpose between IPCF
and PCEF Bearer Binding and Event Reporting Function (BBERF). The PCC-Policy-Service manages Gx/Gx
based on the Diameter dictionary used.
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate pcc-service
This command associates a pre-configured PCC-Service with a PCC-Policy-Service for IPCF configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
no
Removes/disassociate the configured PCC-Service from this PCC-Policy-Service instance configured for
IPCF configuration.
pcc_service_name
Specifies the name of a pre-configured PCC-Service configured in Context Configuration mode for IPCF
configuration.
The pcc_service_name is name of a predefined PCC-Service instance and must be an alphanumerical string
from 1 through 63 characters.
Usage Guidelines Use this command to associate a pre-configured PCC-Service instance for IPCF configuration.
Important For more information on PCC-Service configuration, refer PCC-Service Configuration Mode Commands.
diameter dictionary
This command assigns a Diameter dictionary for Gx/Gxa messaging with a PCC-Policy-Service for IPCF
configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the Diameter Gx dictionary to default dictionary standard (3GPP Rel. 8 standard) for a PCC-Policy-Service
instance configured for IPCF configuration.
gxa-standard
Default: Disabled
Sets the Diameter Gxa dictionary to be used by a PCC-Policy-Service instance configured for IPCF
configuration over Gxa interface to 3GPP Rel. 8 standard.
r7-standard
Default: Disabled
Sets the Diameter Gx dictionary to be used by a PCC-Policy-Service instance configured for IPCF configuration
over Gx interface to 3GPP Rel. 7 standard.
standard
Default: Enabled
Sets the Diameter Gx dictionary to be used by a PCC-Policy-Service instance configured for IPCF configuration
over Gx interface to 3GPP Rel. 8 standard.
Usage Guidelines Use this command to configure the PCC-Policy-Service to determine which of the 3GPP dictionary to be used
for Gx or Gxa interface messaging for policy and/or quota management.
Examples Following command sets the PCC-Policy-Service to use 3GPP Rel. 8 standard dictionary for Gx interface
and policy management related messaging in a PCC-Policy-Service.
default diameter dictionary
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
no
Removes the associated Diameter Origin Endpoint configuration from PCC-Policy-Service instance configured
for IPCF configuration.
any
Sets the PCC-Policy-Service instance to use any available PCEF/BBERF node for policy interfaces (Gx/Gx-like)
support.
dia_endpoint_name
The dia_endpoint_name is a predefined Diameter origin endpoint node and must be an alphanumerical string
from 1 through 63 characters.
Usage Guidelines Use this command to bind the PCEF/BBERF node over Gx/Gx-like interface by associating a pre-configured
Diameter Origin Endpoint with a PCC-Policy-Service.
The Diameter origin endpoint must be a pre-configured instance in the Context Configuration Mode. For
more information on Diameter origin endpoint configuration, refer Diameter Endpoint Configuration Mode
Commands chapter.
Examples Following command associates a pre-configured Diameter endpoint node configuration named pcef_1 with
a PCC-Policy-Service for policy profile management.
diameter origin endpointpcef_1
Following command removes the pre-associated Diameter endpoint node configuration named pcef_1 with
a PCC-Policy-Service.
no diameter origin endpoint
ehrpd-access-bcm
This command configures the PCC-Policy-Service to accept the applicable Bearer-Control-Mode for eHRPD
access over Gxa interface on IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from Application Server (AS) for
eHRPD access over Gxa interface on IPCF node.
as-requested
Default: Enabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from Application Server (AS) for
eHRPD access over Gxa interface on IPCF node.
ue-nw
Default: Disabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from UE and/or network element
for eHRPD access over Gxa interface on IPCF node.
ue-only
Default: Disabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from UE only for eHRPD access
over Gxa interface on IPCF node.
Usage Guidelines Use this command to set the PCC-Policy-Service to accept the Bearer-Control-Mode request from AS or UE
or Network for eHRPD access over Gxa interface on IPCF node.
Examples Following command sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from AS for
eHRPD access over Gxa interface on IPCF node.
default ehrpd-access-bcm
Following command sets the PCC-Policy-Service to accept the Bearer-Control-Mode request only from UE
for eHRPD access over Gxa interface on IPCF node.
ehrpd-access-bcm ue-only
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
gprs-access-bcm
This command configures the PCC-Policy-Service to accept the applicable Bearer-Control-Mode for GPRS
access over Gx interface on IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from Application Server (AS) for
GGSN access over Gx interface on IPCF node.
as-requested
Default: Enabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from Application Server (AS) for
GGSN access over Gx interface on IPCF node.
ue-nw
Default: Disabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from UE and/or network element
for GGSN access over Gx interface on IPCF node.
ue-only
Default: Disabled.
Sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from UE only for GGSN access over
Gx interface on IPCF node.
Usage Guidelines Use this command to set the PCC-Policy-Service to accept the Bearer-Control-Mode request from AS or UE
or Network for GGSN access over Gx interface on IPCF node.
Examples Following command sets the PCC-Policy-Service to accept the Bearer-Control-Mode request from AS for
GGSN access over Gx interface on IPCF node.
default gprs-access-bcm
Following command sets the PCC-Policy-Service to accept the Bearer-Control-Mode request only from UE
for GGSN access over Gx interface on IPCF node.
gprs-access-bcm ue-only
max policy-sessions
This command configures the maximum limit of the policy sessions allowed in a PCC-Policy-Service instance
on IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the maximum policy sessions allowed in PCC-Policy-Service instance to default value of 10000 sessions.
max_session
Default: 10000
Specifies the maximum number of policy sessions configured in PCC-Policy-Service to allow to be connected
in PCC-Quota service instance.
max_session must be an integer between 0 and 4000000.
Usage Guidelines Use this command to set the maximum number of policy sessions allowed by a PCC-Policy-Service instance
on IPCF.
Examples Following command sets the maximum number of policy sessions allowed in PCC-Policy-Service instance
to 10000.
default max policy-sessions
subscriber-binding-identifier
This command specifies the subscriber binding identifier to be used by bindmux for binding different subscriber
session to PCC-Policy-Service on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the subscriber binding identifier to default value; i.e. IMSI, to be used by bindmux for binding different
subscriber session to PCC-Policy-Service on IPCF node.
imsi
Default: Enabled.
Sets the subscriber binding identifier as IMSI to be used by bindmux for binding different subscriber session
to PCC-Policy-Service on IPCF node.
msisdn
Default: Disabled.
Sets the subscriber binding identifier as MSISDN to be used by bindmux for binding different subscriber
session to PCC-Policy-Service on IPCF node.
nai
Default: Disabled.
Sets the subscriber binding identifier as Network Access Identifier (NAI) to be used by bindmux for binding
different subscriber session to PCC-Policy-Service on IPCF node.
sip-uri
Default: Disabled.
Sets the subscriber binding identifier as SIP URI (Uniform Resource Identifier) to be used by bindmux for
binding different subscriber session to PCC-Policy-Service on IPCF node.
Usage Guidelines Use this command to configure the bindmux in PCC-Policy-Service instance on IPCF node to use specific
subscriber identifier for binding different subscriber session to IP-CAN session.
Examples The following command sets the PCC-Policy-Service to use IMSI as subscriber binding identifier for IP-CAN
session on an IPCF node.
default subscriber-binding-identifier
subscription-id-absence-action
This command configures the PCC-Policy-Service instance to handle the Initial Credit Control Request
(CCR-I) messages during initial authentication over Gx interface when CCR-I message received by IPCF
node is without a valid Subscription-Id AVP (IMSI, NAI, E164 etc.).
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the PCC-Policy-Service instance to reject the CCR-I message during initial authentication over Gx
interface if received without a valid Subscription-Id AVP (IMSI, NAI, E164 etc.) on IPCF node.
continue
Default: Disabled.
Sets the PCC-Policy-Service instance to accept the CCR-I message and continue with the session if CCR-I
is received without a valid Subscription-Id AVP (IMSI, NAI, E164 etc.) on IPCF node.
In this case, IPCF accepts the CCR-I message and will do the PCC provisioning as per the operator configuration
in associated PCC-Service.
reject
Default: Enabled.
Sets the PCC-Policy-Service instance to reject the CCR-I message and continue with the session if CCR-I is
received without a valid Subscription-Id AVP (IMSI, NAI, E164 etc.) on IPCF node.
In this case, IPCF will send CCA-I message with Result-code as Permanent Error and rejects the session
establishment with PCEF.
Usage Guidelines Use this command to configure the PCC-Policy-Service instance to handle the Initial Credit Control Request
(CCR-I) message processing during the initial authentication over Gx interface if CCR-I message received
by IPCF node has no valid Subscription-Id AVP.
Examples The following command sets the PCC-Policy-Service to reject the CCR-I request and terminations the session
establishment with PCEF.
default subscription-id-absence-action initial-auth
unsolicited-provisioning
This command is used to enable/disable the support for unsolicited time-of-day-based procedures to
PCC-Policy-Service on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Policy Service Configuration
configure > context context_name > pcc-policy-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pccpolicy-service)#
default
Sets the support for unsolicited time-of-day-based procedures to default mode; i.e. disabled, in
PCC-Policy-Service instance on IPCF node.
no
Removes the configured support for unsolicited time-of-day-based procedures in PCC-Policy-Service instance
on IPCF node.
Usage Guidelines Use this command to enable/disable the support for unsolicited time-of-day-based procedures to
PCC-Policy-Service on IPCF node.
By default this command is disabled.
Examples The following command enables the unsolicited time-of-day-based procedures to PCC-Policy-Service on an
IPCF node.
unsolicited-provisioning time-of-day
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Service-Profile Configuration Mode is used to define the business logic used by the operator for
managing the policy requirements and objectives for the network specific to a group of subscribers in the
network. A PCC-Service-Profile manages multiple PCC-Conditions-Groups and associated PCC-Action-Sets
pairs in an ordered manner. A maximum of 32 PCC-Service-Profile can be configured in a PCC-Service
instance.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
default-rulebase-name
This command is used to associate the default PCC-Rulebase with a PCC-Service-Profile which is to use in
Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
no
Removes the configured default PCC-Rulebase from PCC-Service-Profile instance on IPCF node.
rulebase_name
This keyword specifies the default PCC-Rulebase name to be associated with PCC-Service-Profile instance.
rulebase_name is the Rulebase name configured at PCEF and must be an alphanumeric string of 1 through
63 characters.
Usage Guidelines Use this command to associate the default PCC-Rulebase configured on PCEF with a PCC-Service-Profile
which is to use in Subscriber profile in PCC-Service instance on IPCF node.
Examples Following command associates the PCC-Rulebase named pcc_rulebase1 for PCC-Profile instance on IPCF
node.
default-rulebase-name pcc_rulebase1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
eval-priority
This command sets the priority for evaluation of PCC-Condition-Group with corresponding PCC-Action-Set
in a PCC-Service-Profile which is to use in Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
no
Removes the configured evaluation priority for PCC-Condition-Group with corresponding PCC-Action-Set
from PCC-Service-Profile instance on IPCF node.
last
Sets last evaluation priority action-set configured for the PCC-Service-Profile instance.
no removes the last evaluation priority action-set.
priority_value
Specifies the priority to be set for PCC-Condition-Group with corresponding PCC-Action-Set in a
PCC-Service-Profile instance.
priority_value must be an integer from 1 through 1023.
timedef timedef_name
Specifies a pre-configured time definition in PCC-Timedef Configuration mode and to be set for evaluation
priority in a PCC-Service-Profile instance.
condition-group cond_grp_name
Specifies a pre-configured PCC-Condition-Group to be set for evaluation priority in a PCC-Service-Profile
instance.
cond_grp_name is a pre-configured PCC-Condition-Group and must be an alphanumerical string of 1 through
63 characters.
Important An special PCC-Condition-Group "none" can be used to set the default PCC-Condition-Group for
any-match typically used for a default condition for a session which does not match any of the conditions
specified with higher evaluation priority.
action-set actionset_name
Specifies a pre-configured PCC-Action-Set for PCC-Condition-Group to be set for evaluation priority in a
PCC-Service-Profile instance.
actionset_name is a pre-configured PCC-Action-Set and must be an alphanumerical string of 1 through 63
characters.
Usage Guidelines Use this command to set the priority for evaluation of PCC-Condition-Group with corresponding
PCC-Action-Set in a PCC-Service-Profile which is to use in Subscriber profile in PCC-Service instance on
IPCF node.
Additionally timedef is used to accept the Timedefs to support the time-of-day-based procedures to trigger
an evaluation priority. The action is triggered only when the time of session lies in the time span defined in
specific PCC-Timedef timedef_name.
Default eval-priority has the lowest priority in the PCC-Service-Profile and as default eval-priority does not
have any PCC-Condition-Group associated with it, all the actions in the action-set always be applied.
A maximum of 64 PCC-Evaluation-Priorities can be configured in a PCC-Service-Profile.
Examples Following command sets the evaluation priority value as 1 for PCC-Condition-Group cond_1 along with
PCC-Action-Set act_cond1 for PCC-Service-Profile instance on IPCF node:
eval-priority 1 condition-group cond_1 action-set act_cond1
Following command sets the evaluation priority value as 2 for PCC-Condition-Group none for any-match
typically used for a default condition for a session which does not match any of the conditions specified with
higher evaluation priority.along with PCC-Action-Set act_cond1 for PCC-Service-Profile instance on IPCF
node:
eval-priority 1 condition-group none action-set act_cond1
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
service-tag
This command configures the PCC-Service Tags to be used for PCC-Rulename or PCC-Rule-base in a
PCC-Service-Profile which is to use in Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
no
Removes the configured PCC-Service Tags for PCC-Rulename and PCC-Rulebase from PCC-Service-Profile
instance on IPCF node.
svc_tag
Specifies the name of the PCC-Service Tag to be used for PCC-Rulename and PCC-Rulebase in a
PCC-Service-Profile instance.
svc_tag must be an alphanumerical string of 1 through 63 characters.
rule-name rule_name
Specifies a pre-defined PCC-Rulename on PCEF to be used with PCC-Service Tag svc_tag in a
PCC-Service-Profile instance.
rule_name is a pre-defined PCC-Rulename on PCEF and must be an alphanumerical string of 1 through 63
characters.
rulebase-name rulebase_name
Specifies a pre-defined PCC-Rulebase name pre-defined on PCEF to be used with PCC-Service Tag svc_tag
in a PCC-Service-Profile instance.
Usage Guidelines Use this command to set the PCC-Service Tag for PCC-Rulename and PCC-Rulebase which are defined on
PCEF with a PCC-Service-Profile which is to use in Subscriber profile in PCC-Service instance.
Examples Following command sets the PCC-Service Tag named Rule1 for PCC-Rulebase named pcc_rulebase1 for
PCC-Service-Profile instance on IPCF node:
service-tag Rule1 rulebase-name pcc_rulebase1
Following command sets the PCC-Service Tag named Rule11 for PCC-Rulename pcc_rule1 for
PCC-Service-Profile instance on IPCF node:
service-tag Rule11 rule-name pcc_rule1
timeout long-duration
Configures the long duration timeout and inactivity duration for subscriber session before system notifies or
terminates session in PCC Profile instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
no
Removes the configured long duration timeout value and action in PCC Profile instance.
default
Sets the the long duration timeout value to the default value of '0' which disables the long duration timeout
configuration in PCC Profile instance.
long-duration ldt_timeout
Default: 0
Designates the maximum duration of the session, in seconds, before the system automatically reports/terminates
the session.
Specifies the maximum amount of time, in seconds, before the specified timeout action is activated.
ldt_timeout must be a value in the range from 0 through 4294967295.
The special value 0 disables the timeout specified.
Specifies the action to be taken on expiry of long duration timeout duration ldt_timeout set with timeout
long-duration command.
• detection: sets the system to detect the sessions for which long duration timeout timer is exceeded and
sends the SNMP TRAP and CORBA notification. This is the default behavior.
• disconnect: sets the system to send SNMP TRAP and CORBA notification and disconnect the subscriber
session once the long duration timeout timer is expired.
Usage Guidelines Use this command to set the long duration timeout period and actions to be taken on expiry of duration of
timer for subscriber session.
Important Reduce the timeout duration to free session resources faster for use by new requests.
Important In case of long-duration timeout configured at PCC Service Configuration mode as well as at the
PCC-Profile Configuration mode level, the long-duration timeout and action set in PCC-Profile
Configuration mode will prevail. This enables defining session behavior as per profile provisioning.
Examples Following command sets the system to detect the subscriber sessions that exceeds the long duration timer of
6000 seconds and sends SNMP TRAP and CORBA notification:
timeout long-duration 6000 action detection
Following command sets the system to detect and disconnect the subscriber sessions that exceeds the long
duration timer of 6000 seconds and disconnect the session after sending SNMP TRAP and CORBA notification:
timeout long-duration 6000 action disconnect
usage-monitor
This command creates/modifies/deletes the PCC-Usage-Monitor Configuration instance to track the usage
volume across the PCC-services based on the usage monitor settings in a PCC-service instance for IPCF
configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
no
Removes the configured PCC-Usage-Monitor from PCC-Service-Profile instance for IPCF configuration.
usage_mon_name
Identifies the name of the PCC-Usage-Monitor instance which is to be created or modified through this
command.
The usage_mon_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Caution If this keyword option is used with no usage-monitor usage_mon_name command, the PCC-Usage-Monitor
instance named usage_mon_name is deleted with all configured parameters without prompting any warning
or confirmation.
Usage Guidelines Use this command to create and configure a PCC-Usage-Monitor for PCC-Service-Profile in a PCC-service
instance of IPCF configuration.
A maximum number of 8 PCC-Usage-Monitors can be configured per PCC-Service-Profile.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-profile-usage-mon)#
The commands configured in this mode are defined in the PCC-Usage-Monitor Configuration Mode Commands
chapter of Command Line Interface Reference.
Caution This is a critical configuration. The PCC-Usage-Monitor for volume usage can not be configured without
this configuration. Any change to this configuration would lead to removing or disabling configuration
parameters defined here.
Examples Following command configures the PCC-Usage-Monitor named pcc_usage1 to track the usage of service
with in a PCC-Service-Profile instance.
usage-monitor pcc_usage1
unknown-services-treatment
This command configures the PCC-Service for handling of unknown services at IPCF which is to be used in
Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Service Profile
Configuration
configure > context context_name > pcc-service service_name > profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-profile)#
default
Sets the configured PCC-Service for handling of unknown services at IPCF to default value of "Not allowed"
which is to be used in Subscriber profile in PCC-Service instance on IPCF node.
not-allowed
Sets the PCC-Service-Profile instance to reject the packet filters and does not installs any dynamic rule when
IPCF receives any request to authorize Packet Filters from PCEF and no matching service flow is found in
data service list.
qos-profile qos_prof_name
Specifies a pre-defined PCC-QoS profile name to be used to create dynamic rule when IPCF receives any
request to authorize Packet Filters from PCEF and no matching service flow is found in data service list.
qos_prof_name is a pre-defined PCC-QoS Profile and must be an alphanumerical string of 1 through 63
characters.
start_preced is an integer between 1 through 65535 and must be less than end_preced value where end_preced
is an integer between 1 through 65535 and must be more than start_preced value
Usage Guidelines Use this command to set the PCC-Service Tag for PCC-Rulename and PCC-Rulebase which are defined on
PCEF with a PCC-Service-Profile which is to use in Subscriber profile in PCC-Service instance.
Whenever IPCF receives any request to authorize Packet Filters from PCEF, it does a lookup in data service
list to find a match. If No service flow is found matching then the requested filters are treated as unknown
service request and handled as per the mentioned configuration.
When unknown-service-treatment is set to not-allowed, then Packet Filters are rejected and no dynamic rule
is installed. Otherwise, dynamic rule is created using the requested packet filters, data rates mentioned in the
QoS profile name qos_prof_name and precedence value derived from the configured values.
The precedence configuration works in following manner:
• If precedence limits are configured as 1000 to 2000 with order ascending then precedence of subsequent
dynamic rules will go from 1000 to 2000.
• If precedence limits are configured as 1000 to 2000 with order descending then precedence of subsequent
dynamic rules will go from 2000 to 1000.
Examples Following command sets the PCC-Service for handling of unknown services for PCC-Service-Profile instance
on IPCF node to default action of not allowed:
default unknown-services-treatment
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-QoS-Profile Configuration Mode is used to define the QoS logic used by the operator for managing
the QoS policy requirements and objectives for the network specific to a group of subscribers in the network.
A QoS Profile represents a resource requirement identified by means of the corresponding QoS attributes
like QCI, MBR, GBR, ARP etc.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC QoS Profile
Configuration
configure > context context_name > pcc-service service_name > qos-profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-qos-profile)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
arp-priority
This command is used to define the Allocation and Retention Priority (ARP) values of the QoS profile in
PCC-QoS-Profile which is to use in Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC QoS Profile
Configuration
configure > context context_name > pcc-service service_name > qos-profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-qos-profile)#
Syntax Description [no] arp-priority arp_priority pre-emption {capable | not-capable} {not-vulnerable | vulnerable}
no
Removes the configured ARP priority set for PCC-QoS-Profile for PCC-Service instance on IPCF node.
arp_priority
Specifies the priority value for ARP in a PCC-QoS-Profile which is to use in Subscriber profile in PCC-Service
instance on IPCF node.
arp_priority must be an integer from 1 through 15.
{not-vulnerable | vulnerable}
Sets the Pre-emption vulnerability related parameters with ARP priority in PCC-QoS-Profile name which is
to use in Subscriber profile in PCC-Service instance on IPCF node.
Pre-emption vulnerability determines whether a bearer is applicable for dropping by a pre-emption capable
bearer with a higher ARP priority value.
not-vulnerable: This keyword indicates that the resources assigned to the service data flow shall not be
pre-empted and allocated to a service data flow with a higher priority level.
vulnerable: This keyword indicates that the resources assigned to the service data flow can be pre-empted
and allocated to a service data flow with a higher priority level.
Usage Guidelines Use this command to define the ARP priority and pre-empt parameters in PCC-QoS-Profile which is to be
used in Subscriber profile in PCC-Service instance on IPCF node.
ARP controls how the IPCF reacts when there are insufficient resources to establish the new RAB. Typically
it manages it by; 1) Deny the RAB request and 2) Preempt an existing RAB and accept the new RAB request.
Examples Following command sets the ARP Priority 2 with preemption capability and vulnerability in PCC-QoS-Profile
instance on IPCF node.
arp-priority 2 pre-emption capable vulnerable
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
guaranteed-bitrate
This command defines the Guaranteed Bit Rate (GBR) value in bits per second for downlink and uplink traffic
in PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC QoS Profile
Configuration
configure > context context_name > pcc-service service_name > qos-profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-qos-profile)#
no
Removes the configured GBR value set for PCC-QoS-Profile for PCC-Service instance on IPCF node.
downlink downlink_gbr
Sets the Guaranteed Bit Rate allowed in downlink direction (from PCEF to UE) in bits per second for a
PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
downlink_gbr must be an integer from 0 through 104857600. A 'zero' value disables the downlink in specified
PCC-QoS-Profile.
uplink uplink_gbr
Sets the Guaranteed Bit Rate allowed in uplink direction (from PCEF to PDN) in bits per second for a
PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
uplink_gbr must be an integer from 0 through 104857600. A 'zero' value disables the uplink in specified
PCC-QoS-Profile.
Usage Guidelines Use this command to define the Guaranteed Bit Rate value in bits per second for downlink and uplink traffic
in PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
Examples Following command sets the 1024 bits per seconds as uplink GBR and 2048 bits per second as downlink GBR
in PCC-QoS-Profile instance on IPCF node.
guaranteed-bitrate downlink 2048 uplink 1024
max-bitrate
This command defines the Maximum Bit Rate (MBR) value in bits per second for downlink and uplink traffic
in PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC QoS Profile
Configuration
configure > context context_name > pcc-service service_name > qos-profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-qos-profile)#
no
Removes the configured Maximum Bit Rate value set for PCC-QoS-Profile for PCC-Service instance on IPCF
node.
downlink downlink_mbr
Sets the Maximum Bit Rate allowed in downlink direction (from PCEF to UE) in bits per second for a
PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
downlink_mbr must be an integer from 0 through 104857600. A 'zero' value disables the downlink in specified
PCC-QoS-Profile.
uplink uplink_mbr
Sets the Maximum Bit Rate allowed in uplink direction (from PCEF to PDN) in bits per second for a
PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
uplink_mbr must be an integer from 0 through 104857600. A 'zero' value disables the uplink in specified
PCC-QoS-Profile.
Usage Guidelines Use this command to define the Maximum Bit Rate value in bits per second for downlink and uplink traffic
in PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service instance on IPCF node.
Examples Following command sets the 1024 bits per seconds as uplink MBR and 2048 bits per second as downlink
MBR in PCC-QoS-Profile instance on IPCF node.
max-bitrate downlink 2048 uplink 1024
qci
This command sets the QoS Class Identifier (QCI) for PCC-QoS-Profile which is to use for Subscriber profile
in PCC-Service instance on IPCF node.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC QoS Profile
Configuration
configure > context context_name > pcc-service service_name > qos-profile profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-qos-profile)#
no
Removes the configured QCI value set for PCC-QoS-Profile for PCC-Service instance on IPCF node.
qci qci_id
Sets the QoS Class Identifier for a PCC-QoS-Profile which is to use for Subscriber profile in PCC-Service
instance on IPCF node.
qci_id must be an integer from 1 through 255.
Usage Guidelines Use this command to set the QoS Class Identifier for PCC-QoS-Profile which is to use for Subscriber profile
in PCC-Service instance on IPCF node.
Examples Following command sets the QCI 101 for PCC-QoS-Profile instance on IPCF node.
qci 101
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Quota Service Configuration mode provides a mechanism for Intelligent Policy Control Function
(IPCF) to manage the external interfaces required for quota management purpose. The PCC-Quota service
uses Gx interface towards PCEF/DPI node for Volume-Reporting-Over-Gx (VRoGx) for Quota management
and messaging based on a Diameter dictionary. This mode exists within Context Configuration mode.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Quota Service
Configuration
configure > context context_name > pcc-quota-service service_name
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate pcc-service
This command is used to associate a pre-configured PCC-Service with a PCC-Quota service for IPCF
configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Quota Service
Configuration
configure > context context_name > pcc-quota-service service_name
no
Removes/disassociate the configured PCC-service from this PCC-Quota service instance configured for IPCF
configuration.
pcc_service_name
Specifies the name of a pre-configured PCC-service configured in Context Configuration mode for IPCF
configuration.
The pcc_service_name is name of a predefined PCC-Service instance and must be an alphanumerical string
from 1 through 63 characters.
Usage Guidelines Use this command to associate a pre-configured PCC-Service instance for IPCF configuration.
Important For more information on PCC-Service configuration, refer PCC-Service Configuration Mode Commands.
Examples Following command binds a PCC-Service named pcc_svc1 with in a PCC-Quota service.
associate pcc-service pcc_svc1
Following command removes an associated PCC-Service named pcc_svc1 from a PCC-Quota service.
no associate pcc-service pcc_svc1
diameter dictionary
This command is used to assign a 3GPP Rel. 8 Gx standard Diameter dictionary for VRoGx messaging with
a PCC-Quota service for IPCF configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Quota Service
Configuration
configure > context context_name > pcc-quota-service service_name
no
Removes the assigned a 3GPP Rel. 8 Gx standard Diameter dictionary for VRoGx messaging with a PCC-Quota
service for IPCF configuration.
Usage Guidelines Use this command to assign a 3GPP Rel. 8 Gx standard Diameter dictionary for VRoGx messaging in
PCC-Quota service for quota management.
Examples Following command sets the PCC-Quota service to use 3GPP Rel. 8 standard dictionary over Gx interface
and VRoGx supported quota management related messaging in a PCC-Quota service:
diameter dictionary standard
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Quota Service
Configuration
configure > context context_name > pcc-quota-service service_name
no
Removes the associated Diameter Origin Endpoint configuration from PCC-Quota service instance configured
for IPCF configuration.
any
Sets the PCC-Quota service instance to use any available PCEF/DPI node over Sp interface for quota
management.
dia_endpoint_name
The dia_endpoint_name is a predefined Diameter origin endpoint node and must be an alphanumerical string
from 1 through 63 characters.
Usage Guidelines Use this command to bind the SSC/SPR node over Sp interface by associating a pre-configured Diameter
Origin Endpoint with a PCC-Quota service.
The Diameter origin endpoint must be a pre-configured instance in the Context Configuration Mode. For
more information on Diameter origin endpoint configuration, refer Diameter Endpoint Configuration Mode
Commands chapter.
Examples Following command associates a pre-configured Diameter endpoint node configuration named ssc1 with a
PCC-Quota service for subscriber quota management.
diameter origin endpointssc1
Following command removes the pre-associated Diameter endpoint node configuration named ssc1 with a
PCC-Quota service.
no diameter origin endpoint
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
max total-charging-sessions
This command is used configure the maximum limit of the charging sessions allowed in a PCC-Quota service
instance on IPCF.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC Quota Service
Configuration
configure > context context_name > pcc-quota-service service_name
default
Sets the maximum charging sessions allowed in PCC-Quota service instance to default value of charging
sessions.
max_session
Default: 1
Specifies the maximum number of charging sessions configured in PCC-Quota service to allow to be connected
in PCC-Quota service instance.
max_session must be an integer between 0 and 113.
Usage Guidelines Use this command to set the maximum number of charging sessions allowed by a PCC-Quota service instance
on IPCF.
Examples Following command sets the maximum number of charging sessions allowed in PCC-Quota service instance
to 10000.
default max total-charging-sessions
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-Sp-Endpoint Configuration mode provides a mechanism for Intelligent Policy Control Function
(IPCF) to support the Sp interface endpoint. It represents a client end for SSC interactions. The
PCC-Sp-Endpoint configuration mode facilitates the configuration of Sp interface, and manages the connection
and operational parameters related to its peer.
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
access-type
This command is used to define the type of access protocol to be used with a PCC-Sp-Endpoint instance for
IPCF configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
default
Sets access type protocol to be used by endpoints to default protocol of Diameter protocol in a
PCC-Sp-Endpoint instance for IPCF configuration.
diameter
Default: Enabled
Sets access type protocol to be used by endpoints to Diameter protocol in a PCC-Sp-Endpoint instance for
IPCF configuration.
ldap
Default: Disabled
Sets access type protocol to be used by endpoints to Lightweight Directory Access Protocol (LDAP) in a
PCC-Sp-Endpoint instance for IPCF configuration.
Usage Guidelines Use this command to define the type of access protocol to be used with a PCC-Sp-Endpoint instance for IPCF
configuration.
Examples Following command sets the access type of protocol to Diameter for a PCC-Sp-Endpoint instance.
default access-type
diameter dictionary
This command is used to assign a Diameter dictionary for interaction with SSC and messaging over Sp
interface in a PCC-Sp-Endpoint instance of IPCF configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
default
Sets the Diameter Sh dictionary to default dictionary sh-custon-starent (3GPP Rel. 8 Sh standard) for Sh
interaction and messaging over Sp interface in a PCC-Sp-Endpoint instance of IPCF configuration.
sh-custom-starent
Default: Enabled
Sets the Diameter Sh dictionary to default dictionary sh-custom-starent for Sh interaction and messaging
over Sp interface in a PCC-Sp-Endpoint instance of IPCF configuration.
sh-custom2
Default: Disabled
Sets the Diameter Sh dictionary to sh-custom2with Cisco ULI AVP support for Sh interaction and messaging
over Sp interface in a PCC-Sp-Endpoint instance of IPCF configuration.
sh-standard
Default: Disabled
Sets the Diameter Sh dictionary to default dictionary sh-standard (3GPP Rel. 8 standard) for Sh interaction
and messaging over Sp interface in a PCC-Sp-Endpoint instance of IPCF configuration.
Usage Guidelines Use this command to assign a Diameter dictionary for Sh interaction and messaging over Sp interface in a
PCC-Sp-Endpoint instance of IPCF configuration.
Examples Following command sets the Diameter dictionary for Sh interaction and messaging over Sp interface in a
PCC-Sp-Endpoint instance of IPCF configuration to 3GPP Rel. 8 standard.
diameter dictionary sh-standard
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
no
Removes the associated Diameter Origin Endpoint configuration from PCC-Sp-Endpoint instance configured
for IPCF configuration.
ssc_node_name
The ssc_node_name is a predefined Diameter origin endpoint node (SSC) and must be an alphanumerical
string from 1 through 63 characters.
Usage Guidelines Use this command to bind the SSC node over Sp interface by associating a pre-configured Diameter Origin
Endpoint configuration with a PCC-Sp-Endpoint instance.
The Diameter origin endpoint must be a pre-configured instance in the Context Configuration Mode. For
more information on Diameter origin endpoint configuration, refer Diameter Endpoint Configuration Mode
Commands chapter.
Examples Following command associates a pre-configured Diameter endpoint node configuration named ssc_1 with a
PCC-Sp-Endpoint instance for subscriber policy profile management.
diameter origin endpoint ssc_1
Following command removes the pre-associated Diameter endpoint node configuration named ssc_1 from a
PCC-Sp-Endpoint instance.
no diameter origin endpoint
diameter peer-select
This command nominates primary and secondary Diameter peers amongst the peers configured under Diameter
Endpoint Configuration instance which is associated with a PCC-Sp-Endpoint configuration.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
Syntax Description diameter peer-select peer pri_peer_name [row-precedence row_prec_name selection-condition imsi
{mcc mcc-string [mnc mnc-string] [msin msin-string] | mnc mnc-string [msin msin-string] | msin
msin-string}] [realm realm_name] [secondary-peer sec_peer_name [realm sec_realm_name]]
no diameter peer-select
no diameter peer-select row-precedence
no
To remove all the configuration parameters.
To remove a particular selection-condition, a row-precedence value is specified.
peer pri_peer_name
Sets a configured Diameter peer, which is configured in Diameter Endpoint configuration associated with
PCC-Sp-Endpoint instance configuration, as primary in Diameter peer selection process for IPCF configuration.
The pri_peer_name is a pre-configured Diameter peer in Diameter Endpoint configuration which is associated
with a PCC-Sp-Endpoint configuration and must be an alphanumerical string from 1 through 63 characters.
pri_peer_name allows punctuation marks.
row-precedence row_prec_name
The row-precedence decides order of evaluation of the selection conditions.
The row_prec_name is an integer between 1 and 63. Lower the value, higher is the priority of evaluation.
selection-condition
The selection-condition parameter is used to define Diameter SPR peer selection conditon.
imsi
It indicates that the selection condition is based on subscriber IMSI.
mcc mcc-string
It indicates that the selection condition is based on MCC component of subscriber IMSI.
The mcc-string can have one of the following formats: mccval1 or mccval1-mccval2. The first format compares
single MCC value and second one incorporates the range with mccval1 <= mccval2.
The values for both mccval1 and mccval2 must be between 100 to 999.
mnc mnc-string
It indicates that the selection condition is based on MNC component of subscriber IMSI.
The mnc-string can have one of the following formats: mncval1 or mncval1-mncval2. The first format compares
single MNC value and second one incorporates the range with mncval1 <= mncval2.
The values for both mncval1 and mncval2 must be between 1 to 999.
msin msin-string
It indicates that the condition is based on MSIN component of subscriber IMSI.
The msin-string can have following format: msinval1-msinval2. The format incorporates the range with
mccval1 <= mccval2.
The values for both mccval1 and mccval2 must be of maximum of 10 digits.
realm realm-name
This keyword optionally defines the realm (domain) of a configured primary Diameter peer, which is
configured in Diameter Endpoint configuration associated with PCC-Sp-Endpoint instance configuration, in
Diameter peer selection process for IPCF configuration.
realm_name is the realm (domain) of the associated primary Diameter peer in Diameter Endpoint configuration
which associated with a PCC-Sp-Endpoint configuration. The realm_name must be an alpha and/or numeric
string of 1 to 127 characters. The realm may typically be a company or service name and it allows punctuation
marks.
secondary-peer sec_peer_name
Sets a configured Diameter peer, which is configured in Diameter Endpoint configuration associated with
PCC-Sp-Endpoint instance configuration, as secondary in Diameter peer selection process for IPCF
configuration.
The sec_peer_name is a pre-configured Diameter peer in Diameter Endpoint configuration which is associated
with a PCC-Sp-Endpoint configuration and must be an alphanumerical string from 1 through 63 characters.
sec_peer_name allows punctuation marks.
realm sec_realm-name
This keyword optionally defines the realm (domain) of a configured secondary Diameter peer, which is
configured in Diameter Endpoint configuration associated with PCC-Sp-Endpoint instance configuration, in
Diameter peer selection process for IPCF configuration.
sec_realm_name is the realm (domain) of the associated primary Diameter peer in Diameter Endpoint
configuration which associated with a PCC-Sp-Endpoint configuration. The sec_realm_name must be an
alpha and/or numeric string of from 1 to 127 characters. The realm may typically be a company or service
name and it allows punctuation marks.
Usage Guidelines Use this command to nominate primary and secondary Diameter peers amongst the peers configured under
Diameter Endpoint Configuration instance which is associated with a PCC-Sp-Endpoint configuration. When
both primary and secondary are down, the remaining Diameter peers are chosen based on their configured
weight in round robin manner.
When row-precedence and selection-conditions are not defined for peer selection configuration, the row
precedence value is assumed to be 64 (which is the lowest). This is a deafult peer selection when all the other
configured selection conditions fail.
Multiple Diameter peers can be configured in a PCC-Sp-Endpoint instance by entering this command multiple
times.
Examples Assume the operator has two MCC-MNC combinations: 123-456 and 123-457 respectively for subscriber
IMSI values. If operator wishes to divert subscribers with these two different combinations to different SSC
peers (say ssc123456 and ssc123457) then the operator needs to use following commands under
PCC-Sp-Endpoint :
diameter peer-select row-precedence 4 selection-condition imsi mcc 123 mnc 456 peer ssc123456
diameter peer-select row-precedence 6 selection-condition imsi mnc 123 mnc 456 peer ssc223457
Following command nominates a pre-configured Diameter peer dia1 as primary and dia2 as secondary for
Diameter peer selection process in a PCC-Sp-Endpoint configuration instance.
diameter peer-select peer dia1 secondary peer dia2
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
profile-data
This command allows the operator to specify data-reference and service indication AVP values used in
UDR/SNR message for profile data sent over Sp endpoint when access type is set to Diameter.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
default
Specifies the data-reference value used in UDR/SNR message for profile data sent over Sp endpoint to default
value of '0' (zero) and service indication AVP value to profile data when access type is set to Diameter in
PCC-Sp-Endpoint instance.
data-reference data_ref_value
Default: 0
Specifies the data-reference values used in UDR/SNR message for profile data sent over Sp endpoint when
access type is set to Diameter in PCC-Sp-Endpoint instance.
The data reference value data_ref_value must be an integer from 1 through 65535.
service-indication svc_ind_value
Specifies the service indication AVP value used in UDR/SNR message for profile data sent over Sp endpoint
when access type is set to Diameter in PCC-Sp-Endpoint instance.
Important The Service-Indication values are set as per application logic and are supposed to be used only with SSC.
The service indication value svc_ind_value must be a string of alpha and/or numeric characters from 1 to 32
characters.
Usage Guidelines Use this command to allow the operator to specify data-reference and service indication AVP values used in
UDR/SNR message for profile data sent over Sp endpoint when access type is set to Diameter in
PCC-Sp-Endpoint configuration instance.
Default service-indication value varies as per other interface configuration default profile-data key is used
to set the service indication value.
Examples Following command set the data-reference value used in UDR/SNR message for profile data sent over Sp
endpoint to default value of '0' (zero) and service-indication value as per application logic when access type
is set to Diameter in PCC-Sp-Endpoint instance.
default profile-data key
profile-update-notification
This command sets the system to indicate whether SSC and IPCF are capable of supporting profile update
notifications in a PCC-Sp-Endpoint instance.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
default
Set the PCC-Sp-Endpoint instance default mode for supporting profile update notifications in a
PCC-Sp-Endpoint instance and also to use the same in a PCC-Sp-Endpoint configuration instance.
allow
Default: Enabled
Set the PCC-Sp-Endpoint instance to support the profile update notifications and also allow to use the same
in Sp interaction.
disallow
Default: Disabled
Set the PCC-Sp-Endpoint instance to not to support the profile update notifications and also does not allow
to use the same in Sp interaction.
Usage Guidelines Use command to set to indicate whether SSC and IPCF are capable of supporting profile update notifications
in a PCC-Sp-Endpoint instance. It also sets that whether profile update notification should be used or not for
a PCC-Sp-Endpoint configuration instance.
Examples Following command indicates that SSC and IPCF are capable of supporting profile update notifications in a
PCC-Sp-Endpoint instance and also allow to use it for a PCC-Sp-Endpoint configuration instance.
default profile-update-notification
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Sp Endpoint Configuration
configure > context context_name > pcc-sp-endpoint endpoint_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-spendpoint)#
default
Sets the PCC-Sp-Endpoint instance to use subscriber IMSI to identify subscriber uniquely in SPR database
while requesting subscriber data from SSC.
imsi
Default: Enabled
Sets the PCC-Sp-Endpoint instance to use subscriber IMSI to identify subscriber uniquely in SPR database
while requesting subscriber data from SSC.
msisdn
Default: Disabled
Sets the PCC-Sp-Endpoint instance to use subscriber MSISDN to identify subscriber uniquely in SPR database
while requesting subscriber data from SSC.
nai
Default: Disabled
Sets the PCC-Sp-Endpoint instance to use Network Address Identifier as token to identify subscriber uniquely
in SPR database while requesting subscriber data from SSC.
This token facilitates CDMA users for Policy Control and Charging functions.
Usage Guidelines Use command to set the PCC-Sp-Endpoint instance to indicate how a subscriber is uniquely identified in SPR
database on SSC while requesting subscriber data. By default it uses Subscriber IMSI for identification in
SPR database.
For IP-CAN session between PDSN and IPCF the subscriber token NAI facilitates the Policy Control and
Charging functions to subscribers.
Examples Following command sets the PCC-Sp-Endpoint instance to use a subscriber IMSI to uniquely identified in
SPR database at SSC.
default spr subscriber identifier
Important This configuration mode is supported from StarOS Release 15.0 onwards.
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate-addon-state
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured associated addon state from this PCC service addon instance for IPCF configuration.
expired
grace
not-started
started
action-set act_set_name
The act_set_name must be an alphanumerical string from 1 through 63 characters.
notification-template not_temp_name
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
description
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured PCC-Action-Set from this PCC-service instance for IPCF configuration.
act_set_name
Identifies the name of the PCC-Action-Set which is to be created or modified through this command.
The act_set_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
duration
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured PCC-Action-Set from this PCC-service instance for IPCF configuration.
act_set_name
Identifies the name of the PCC-Action-Set which is to be created or modified through this command.
The act_set_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
status active
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured PCC-Action-Set from this PCC-service instance for IPCF configuration.
act_set_name
Identifies the name of the PCC-Action-Set which is to be created or modified through this command.
The act_set_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
time-allowance
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured PCC-Action-Set from this PCC-service instance for IPCF configuration.
act_set_name
Identifies the name of the PCC-Action-Set which is to be created or modified through this command.
The act_set_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
volume-allowance
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > Addon Configuration
configure > context context_name > pcc-service service_name> addon addon_name
no
Removes the configured PCC-Action-Set from this PCC-service instance for IPCF configuration.
act_set_name
Identifies the name of the PCC-Action-Set which is to be created or modified through this command.
The act_set_name must be an alphanumerical string from 1 through 63 characters.
-noconfirm
Indicates that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to create/remove/configure an action-set in a PCC-service instance for IPCF Configuration.
An Action-set indicates the policy and charging as well as event generation related decisions that will get
activated when the corresponding Condition-Group is evaluated to TRUE within a subscriber policy/profile.
A maximum of 512 PCC-Action-Sets can be configured in 1 instance of PCC-Service.
Entering this command results in the following prompt:
[context_name]hostname(config-pcc-action-set)
Important For more information on PCC-Action-Set configuration, refer PCC-Action-Set Configuration Mode
Commands.
Important This configuration mode is supported from StarOS Release 12.1 onward.
The PCC-TimeDef Configuration Mode is used to configure various time definitions (TimeDefs) in the
PCC-service instance. A PCC-TimeDef specifies the start and end time for triggering policy-related procedures
or conditions.
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC TimeDef
Configuration
configure > context context_name > pcc-service service_name > timedef timedef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-timedef)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
start date
This command defines PCC-TimeDefs with a start and end dates with time for an event to trigger a procedure
or condition in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC TimeDef
Configuration
configure > context context_name > pcc-service service_name > timedef timedef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-timedef)#
Syntax Description [no] start date start_date time start_time end date end_date time end_time
no
Removes the specified time definition.
date start_date
Specifies the start date start_date in MM/DD/YY format.
Following format is used for MM/DD/YY in start_date:
• MM specifies the month of the start date from January through December and must be an integer between
1 through 12.
• DD specifies the date of month of the start date 1 through 31 and must be an integer between 1 through
31.
• YY specifies the year of the start date from 2010 through 2037 and must be an integer between 10 through
37.
time start_time
Specifies the start time start_time in HH MIN SS format.
Following format is used for HH MIN SS in start_time:
• HH specifies the hour of the start date in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the start date must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the start date must be an integer between 00 through 59.
time end_time
Specifies the end time end_time in HH MIN SS format.
Following format is used for HH MIN SS in end_time:
• HH specifies the hour of the end date in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the end date must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the end date must be an integer between 00 through 59.
Usage Guidelines Use this command to define a PCC-TimeDef with a start and end dates with time for an event to trigger a
procedure or condition in an IP-CAN session.
Examples The following command defines a PCC-TimeDef with start date as December 31st 2010 at 00 Hrs. 30 mins.
and 00 seconds and end date as January 31st 2011 at 23 Hrs. 59 mins. and 59 seconds:
start date 12/31/10 time 00 30 00 end date 01/31/11 time 23 59 59
start day
This command defines PCC-TimeDefs with a start and end week days with time for an event to trigger a
procedure or condition in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC TimeDef
Configuration
configure > context context_name > pcc-service service_name > timedef timedef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-timedef)#
Syntax Description [no] start day start_weekdays time start_time end day end_weekdays time end_time
no
Removes the specified time definition.
day start_weekdays
Specifies the start day as week days.
Following days are used as start_weekdays:
• friday specifies Friday of the week as start day.
• monday specifies Monday of the week as start day.
• saturday specifies Saturday of the week as start day.
• sunday specifies Sunday of the week as start day.
• thrusday specifies Thursday of the week as start day.
• tuesday specifies Tuesday of the week as start day.
• wednesday specifies Wednesday of the week as start day.
time start_time
Specifies the start time start_time in HH MIN SS format.
time end_time
Specifies the end time end_time in HH MIN SS format.
Following format is used for HH MIN SS in end_time:
• HH specifies the hour of the end day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the end day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the end day must be an integer between 00 through 59.
Usage Guidelines Use this command to define a PCC-TimeDef with a start and end weekdays with time for an event to trigger
a procedure or condition in an IP-CAN session.
Examples The following command defines a PCC-TimeDef with start day as Friday at 00 Hrs. 30 mins. and 00 seconds
and end day as Sunday at 23 Hrs. 59 mins. and 59 seconds:
start day friday time 00 30 00 end day sunday time 23 59 59
start time
This command defines PCC-TimeDefs with a start and end time of a day for an event to trigger a procedure
or condition in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC TimeDef
Configuration
configure > context context_name > pcc-service service_name > timedef timedef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-timedef)#
no
Removes the specified time definition.
time start_time
Specifies the start time start_time in HH MIN SS format.
Following format is used for HH MIN SS in start_time:
• HH specifies the hour of the day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the day must be an integer between 00 through 59.
Usage Guidelines Use this command to define a PCC-TimeDef with a start and end time for an event to trigger a procedure or
condition in an IP-CAN session.
Examples The following command defines a PCC-TimeDef with start at 00 Hrs. 30 mins. and 00 seconds and end at 23
Hrs. 59 mins. and 59 seconds:
start time 00 30 00 end time 23 59 59
time-slot
This command defines PCC-TimeDefs with a start and end week days with time for an event to trigger a
procedure or condition in an IP-CAN session.
Product IPCF
Command Modes Exec > Global Configuration > Context Configuration > PCC Service Configuration > PCC TimeDef
Configuration
configure > context context_name > pcc-service service_name > timedef timedef_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pcc-timedef)#
Syntax Description [ no ] time-slot slot_num{ start datestart_date time start_time end date end_date time end_time | start day
start_weekdays time start_timeend day end_weekdays time end_time | start time start_time end time
end_time
[ no ] time-slot slot_num
no
Removes the specified time slot definition.
slot_num
Specifies the time slot number to be configured with this command and it must be an integer between 1 through
64.
date start_date
Specifies the start date start_date in MM/DD/YY format.
Following format is used for MM/DD/YY in start_date:
• MM specifies the month of the start date from January through December and must be an integer between
1 through 12.
• DD specifies the date of month of the start date 1 through 31 and must be an integer between 1 through
31.
• YY specifies the year of the start date from 2010 through 2037 and must be an integer between 10 through
37.
time start_time
Specifies the start time start_time in HH MIN SS format.
Following format is used for HH MIN SS in start_time:
• HH specifies the hour of the start date in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the start date must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the start date must be an integer between 00 through 59.
time end_time
Specifies the end time end_time in HH MIN SS format.
Following format is used for HH MIN SS in end_time:
• HH specifies the hour of the end date in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the end date must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the end date must be an integer between 00 through 59.
day start_weekdays
Specifies the start day as week days.
Following days are used as start_weekdays:
• friday specifies Friday of the week as start day.
• monday specifies Monday of the week as start day.
• saturday specifies Saturday of the week as start day.
• sunday specifies Sunday of the week as start day.
• thrusday specifies Thursday of the week as start day.
• tuesday specifies Tuesday of the week as start day.
• wednesday specifies Wednesday of the week as start day.
time start_time
Specifies the start time start_time in HH MIN SS format.
Following format is used for HH MIN SS in start_time:
• HH specifies the hour of the start day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the start day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the start day must be an integer between 00 through 59.
time end_time
Specifies the end time end_time in HH MIN SS format.
Following format is used for HH MIN SS in end_time:
• HH specifies the hour of the end day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the end day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the end day must be an integer between 00 through 59.
time start_time
Specifies the start time start_time in HH MIN SS format.
Following format is used for HH MIN SS in start_time:
• HH specifies the hour of the day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the hour of the day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the day must be an integer between 00 through 59.
• HH specifies the end hour of the day in 24 hour format and must be an integer between 00 through 23.
• MIN specifies the minutes of the end hour of the day must be an integer between 00 through 59.
• SS specifies the seconds of the minute of the end hour must be an integer between 00 through 59.
Usage Guidelines Use this command to define a time slot with a start day, start time, and start date optiion for an event to trigger
a procedure or condition in an IP-CAN session. A maximum of 12 time-slots can be configured through this
command.
Examples The following command defines a time slot 2 which will start on Friday at 00:30:00 and ends on Saturday at
23:00:00:
start day friday start time 00 30 00 end day saturday time 23 00 00
Important This configuration mode is customer specific. For more information, contact your Cisco account
representative.
Command Modes Exec > ACS Configuration > Port Control Protocol Service Configuration
active-charging service service_name > pcp-service service_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed (s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
policy-control
This command enters the PCP Policy Control Configuration mode to configure policy control parameters for
PCP service.
Important This command is customer specific. For more information, contact your Cisco account representative.
Product ACS
NAT
PSF
Command Modes Exec > ACS Configuration > Port Control Protocol Service Configuration
active-charging service service_name > pcp-service service_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-service)#
default
Configures this command with the default setting.
Default: Enabled
Usage Guidelines Use this command to enter the PCP Policy Control Configuration Mode to configure the policy control
parameters for the PCP service.
On entering this command, the CLI prompt changes to:
[context_name]hostname(config-pcp-policy-control)#
Also see the PCP Policy Control Configuration Mode Commands chapter.
server
Configures the IP address of the PCP server to receive PCP packets.
Important This command is customer specific. For more information, contact your Cisco account representative.
Product ACS
NAT
PSF
Command Modes Exec > ACS Configuration > Port Control Protocol Service Configuration
active-charging service service_name > pcp-service service_name
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-service)#
server ipv4_address
Specifies the IPv4 address of the server to receive PCP packets.
ipv4_address must be specified using the IPv4 dotted-decimal notation.
port port_number
Specifies the UDP port number where PCP Request messages are received by the PCP service.
port_number must be an integer from 1 through 65535.
Default: 5351
Usage Guidelines Use this command to configure the IPv4 address on which the PCP service will receive PCP packets and the
port on which PCP Request messages will be received from the PCP service.
Example
The following command configures the IPv4 address 1.2.3.4 with port number 5351 for the PCP service:
server ipv4-address 1.2.3.4 port 5351
Important This configuration mode is customer specific. For more information, contact your Cisco account
representative.
Command Modes Exec > ACS Configuration > PCP Configuration > Port Control Protocol Service Policy Control Configuration
active-charging service service_name > pcp-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-policy-control)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed (s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
request-opcode
This command allows you to configure various PCP Request Opcode options.
Important This command is customer specific. For more information, contact your Cisco account representative.
Product ACS
NAT
PSF
Command Modes Exec > ACS Configuration > PCP Configuration > Port Control Protocol Service Policy Control Configuration
active-charging service service_name > pcp-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-policy-control)#
no
Deletes the specific PCP opcode settings.
announce
Configures PCP ANNOUNCE opcode to process Announce Request messages.
peer
Configures PCP PEER opcode to process Peer Request messages.
Usage Guidelines Use this command to configure various PCP Request Opcode options.
response-opcode
This command allows you to configure various PCP Response Opcode options.
Product ACS
NAT
PSF
Command Modes Exec > ACS Configuration > PCP Configuration > Port Control Protocol Service Policy Control Configuration
active-charging service service_name > pcp-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[local]host_name(config-pcp-policy-control)#
Syntax Description response-opcode { map | peer } [ error { long life-time long_life_time | short life-time short_life_time } |
success life-time succ_life_time ] +
{ default | no } response-opcode [ map | peer ] +
default
Configures this command with its default setting.
map
Configures the lifetime for which Map mappings are available.
peer
Configures the lifetime for which Peer mappings are available.
peer
Configures this command with its default setting.
Usage Guidelines Use this command to configure the PCP Response Opcode options.
Example
The following command configures the MAP opcode with lifetime for long and short error cases set to 600
and 30 respectively:
response-opcode map error long life-time 600 short life-time 30
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
aaa attribute
Sets the system attributes for AAA messages.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
Syntax Description aaa attribute { 3gpp2-bsid string | 3gpp2-service-option integer | calling-station-id integer |
3gpp2-serving-pcf ip-address }
no aaa attribute
default aaa attribute 3gpp2-service-option integer
no
Removes a previously configured AAA attribute.
default
Returns the specified aaa attribute to the original default system settings.
3gpp2-bsid string
Specifies the base-station ID and consists of the SID + NID + CELLID.
string must contain 12 hexadecimal upper-case ASCII characters.
3gpp2-service-option integer
Specifies the radius attribute value when sending authentication and accounting messages as an integer from
0 through 32767. Default: 4095
calling-station-id integer
Specifies the calling station phone number as a sequence of 1 through 15 digits.
3gpp2-serving-pcf ip-address
Use this command to generate attribute values without creating a new ASR 5000ASR 5500 image.
Usage Guidelines If the RADIUS protocol is being used, accounting messages can be sent over a AAA interface to the RADIUS
server.
3gpp2-serving-pcf attribute value (if configured) is sent in both RADIUS authentication and accounting
messages. If the attribute value is not configured (or explicitly "not configured" using the no keyword),
RADIUS attributes are still included with just type and length. This is because inclusion/exclusion of RADIUS
attributes are still controlled through the dictionary, not via the CLI.
aaa authentication
Sets the aaa authentication for first and second phase authentication when multiple authentication is configured
on the system.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
Syntax Description aaa authentication { { first-phase | second-phase } | { context-name name aaa-group name } }
no aaa authentication { first-phase | second-phase }
Important First phase authentication is mandatory when multiple authentication is configured on the system.
• context-name name: Specifies the context where the aaa server group is defined as an alphanumeric
string of 1 through 79 characters.
• aaa-group name: Specifies the name of the aaa-group to be used for authentication as an alphanumeric
string of 1 through 79 characters.
• aaa-group name: Specifies the name of the aaa-group to be used for authentication as an alphanumeric
string of 1 through 63 characters.
Usage Guidelines Two phase-authentication happens in IKEv2 setup for setting up the IPSec session. The first authentication
uses Diameter AAA EAP method and second authentication uses RADIUS AAA authentication. The same
AAA context may be used for both authentications. PDIF service allows you to specify only a single AAA
group, which could normally be used for the first authentication method.
A given AAA group only supports either Diameter or RADIUS authentication. If the NAI in the first
authentication is different from NAI in the second authentication each NAI can point to a different domain
profile in the PDIF. Each domain profile may be configured with each AAA group, one for Diameter and the
other for RADIUS.
Examples Use the following to configure first-phase authentication for an aaa group named aaa-10 in the PDIF context:
first-phase context-name pdif aaa-group aaa-10
bind
Binds the service IP address to a crypto template and configures the number of sessions the PDIF can support.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
no
Removes a previously configured binding.
address
Specifies the IP address of the service.
crypto-template string
Specifies the name of the crypto template to be bound to the service as an alphanumeric string of 0 through
127 characters.
max-sessions number
Specifies the maximum number of sessions to be supported by the service as an integer from 0 to 3000000.
Default: 3000000
If the max-sessions value is changed on an existing system, the new value takes effect immediately if it is
higher than the current value. If the new value is lower than the current value, existing sessions remain
established, but no new sessions are permitted until usage falls below the newly-configured value.
Usage Guidelines Binds the IP address used as the connection point for establishing the IKEv2 sessions to the crypto template.
It can also define the number of sessions the PDIF can support.
Examples The following command binds a service with the IP address 13.1.1.1 to the crypto template T1 and sets the
maximum number of sessions to 2000000:
bind address 13.1.1.1 crypto-template T1 max-sessions 200000
default
Sets or restores the default condition for the selected parameter.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
duplicate-session-detection
Configures the default to be NAI-based.
setup-timeout
Default call setup time limit is 60 seconds.
subscriber name
Configures the default subscriber name. name is a string of 1-127 characters.
username mac-address-stripping
Default is to disable stripping the MAC address from the username.
Examples Use the following example to configure the default call setup time limit:
default setup-timeout
duplicate-session-detection
Configures the PDIF to detect duplicate call sessions using old IMSI or NAI addresses and clear old call
information.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
no
Stops duplicate session detection.
default
Configures the default setting, which is NAI-based detection.
imsi-based
Configures the PDIF to detect duplicate call sessions based on the IMSI address.
nai-based
Configures the PDIF to detect duplicate call sessions based on the NAI address. This is the default setting.
Usage Guidelines If an MS leaves the Wi-Fi coverage area and subsequently comes back online, it may initiate a new session
setup procedure. After both the device authentication with HSS and the subscriber authentication with AAA
server are completed, PDIF runs the internal mechanism to see whether there was any other session bound
with the same IMSI. If an old session is detected, PDIF starts clearing this old session by sending a proxy-MIP
Deregistration request to the HA. PDIF resumes new session setup by sending a proxy-MIP registration
request. When the old session is aborted, PDIF sends Diameter STR messages and RADIUS Acct STOP
messages to corresponding AAA servers.
PDIF allows duplicate session detection based on either the NAI or IMSI addresses. When detecting based
on NAI, it is the first-phase (device authentication) NAI that is used.
Examples The following command configures duplicate session detection to use IMSI addressing:
duplicate-session-detection imsi
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
hss
Configures the Home Subscriber Server (HSS) parameters.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
no
Removes a previously configured HSS profile.
default
Resets the defaults for this command.
failure-handling mac-address-validation-failure
Configures how the HSS is to handle errors.
If HSS returns a list of MAC addresses and if PDIF fails to match the subscriber MAC address against the
list, the session is always terminated.
mac-address-validation
If mac-address-validation is enabled, the PDIF queries the HSS for a list of MAC addresses associated with
the Mobile Directory Number (MDN). Default: Disabled
update-profile
Update the HSS with the subscriber profile. Default: Disabled
Usage Guidelines An HSS provides MAC address validation and store part of the subscriber profile. This command enables or
disables validation and profile updates, and configures how the system responds to failures: terminate or
continue a session.
An ims-sh-service and Diameter interface need to be configured to communicate with the HSS.
ims-sh-service
Associates the IMS-Sh-service parameters.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
no
Removes a previously configured IMS-Sh-service.
name
Names the IMS-Sh-service in the pdif-service context.
ip source-violation
Sets the parameters for IP source validation. Source validation is useful if packet spoofing is suspected or for
verifying packet routing and labeling within the network.
Source validation requires that the source address of the received packets matches the IP address assigned to
the subscriber (either statically or dynamically) during the session.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
clear-on-valid-packet
Configures the service to reset the reneg-limit and drop-limit counters after receipt of a properly addressed
packet. Default: disabled
drop-limit num
Sets the number of allowed source violations within a detection period before forcing a call disconnect. If
num is not specified, the value is set to the default.
num is an integer from 1 to 1000000. Default: 10
period secs
Sets the length of time (in seconds) for a source violation detection period to last.
If secs is not specified, the value is set to the default.
secs is an integer from 1 to1000000. Default: 120
Usage Guidelines This function is intended to allow the operator to configure a network to prevent problems such as when a
user gets handed back and forth between two PDIFs a number of times during a handoff scenario.
Examples The following command sets the drop limit to 15 and leaves the other values at their defaults:
ip source-violation drop-limit 15
mobile-ip
Sets the MIP FA context for the specific PDIF service.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
no
Removes previously configured parameters.
fa-service string
Specifies the name of the FA service in the FA context as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Shows in which context the FA is located and names the FA service.
Examples This command configures MIP for the FA context named fa1:
mobile-ip foreign-agent context fa1
setup-timeout
Configures the maximum time allowed to set up a session.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
setup-timeout integer
Specifies the session setup timer (in seconds) as an integer from 2 through 300. Default: 60
default setup-timeout
Defaults the session setup timer to 60 seconds.
Usage Guidelines PDIF clears both user session and tunnels if a call does not initiate successfully before the timer expires.
Examples The following command sets the setup-timeout to the default 30 seconds:
default setup-timeout
username
Configures mac-address-stripping on a username coming in from a mobile station session.
Product PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDIF Service Configuration
configure > context context_name > pdif-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdif-service)#
username mac-address-stripping
Configures mac-address stripping from the Network Access Identifier (NAI).
default
Configures the default parameter which is disabled.
no
Returns the configuration to the default condition.
Usage Guidelines When enabled, PDIF strips the MAC address from a mobile username NAI before sending to the RADIUS
AAA server.
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
aaa attribute
Sets the attributes that the system uses in AAA messages.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
3gpp-negotiated-qos-profile string
Specifies the 3GPP negotiated QoS profile to use in AAA messages during IMS emergency call handling as
an alphanumeric string of 1 through 31 characters.
no aaa attribute
Removes a previously configured AAA attribute.
Usage Guidelines Specifies the 3GPP negotiated QoS profile to use in AAA messages during IMS emergency call handling.
Examples The following command specifies the 3GPP negotiated QoS profile to use during IMS emergency call handling:
aaa attribute 3gpp-negotiated-qos-profile 100
associate sgtp-service
Identifies the SGTP service to be associated with the PDG service to enable TTG functionality on the PDG/TTG.
TTG functionality supports GTP-C (GTP control plane) messaging and GTP-U (GTP user data plane) messaging
between the TTG and the GGSN over the Gn' interface.
Important This command can be used before the associated service instance is created and configured but care should
be used to match the service names.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
no
Removes the service association definition from the configuration.
sgtp-service sgtp_service_name
Specifies which SGTP service configuration, by naming the SGTP service instance, to associate with this
PDG service.
sgtp_service_name is an alphanumeric string of 1 through 63 characters with no spaces.
context sgpt_context_name
Defines the context in which the SGTP service was created. If no context is specified, the current context is
used.
sgtp_context_name is an alphanumeric string of 1 through 63 characters with no spaces.
Usage Guidelines Use this command to associate the SGTP service to be associated with the PDG service to enable TTG
functionality on the PDG/TTG.
Examples The following command associates SGTP service sgtp_service_1 with this PDG service:
associate sgtp-service sgtp_service_1 context sgtp_context_1
certificate-selection
Configures the PDG/TTG to select the trusted certificate (and the private key for calculating the AUTH
payload) to be included in the first IKE_AUTH message from the PDG/TTG based on the APN (Access Point
Name). The selected certificate is associated with the APN included in the IDr payload of the first IKE_AUTH
message from the UE.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
certificate-selection apn-based
Selects a trusted certificate for the first IKE-AUTH message based on the APN.
no certificate-selection
Disables APN-based certificate selection and resumes sending a certificate bound to a crypto template.
default certificate-selection
Sets the default certificate selection method to a certificate bound to a crypto template.
Usage Guidelines Configures the PDG/TTG to select the trusted certificate to be included in the first IKE_AUTH message based
on the APN.
bind
Binds the PDG service IP address to a crypto template and specifies the maximum number of sessions the
PDG service supports.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
Syntax Description [ no ] bind address ipv4_address { crypto-template string } mode { ttg | pdg } [ max-sessions number ]
no
Removes a previously configured binding.
crypto-template string
Specifies the name of the crypto template to be bound to the PDG service. This is the name of the IPSec policy
to be used as a template for PDG/TTG subscriber session IPSec policies. The crypto template includes most
of the IPSec and IKEv2 parameters for keepalive, lifetime, NAT-T, and cryptographic and authentication
algorithms. There must be one crypto template per PDG service.
This is a mandatory parameter.
string is an alphanumeric string of 0 through 127 characters.
• In TTG mode, PDN connectivity is provided through the GGSN. PDG functionality is provided by the
combined TTG and GGSN.
• In PDG mode, PDN connectivity and PDG functionality are provided directly through the PDG service.
Dependencies:
When you configure the PDG service to be in TTG mode, you must also configure the SGTP service using
the associate sgtp-service command, as the TTG needs to connect with the GGSN to complete the PDG
functionality.
The following behaviors occur when the PDG service operates in TTG mode:
• If the SGTP service associated with PDG service is not configured, the PDG service is not started.
• If the SGTP service associated with PDG service is not started, the PDG service is not started.
• If the SGTP service associated with PDG service is stopped, the PDG service is stopped.
• If the SGTP service associated with PDG service is re-started, the PDG service is re-started.
• If the SGTP service is not yet configured, whenever the SGTP service is started, the PDG service is
started.
Note that starting or stopping the PDG service has no impact on the SGTP service.
max-sessions number
Specifies the maximum number of sessions to be supported by the PDG service as an integer from 0 through
1000000. Default: 1000000
If the max-sessions value is changed on an existing system, the new value takes effect immediately if it is
higher than the current value. If the new value is lower than the current value, existing sessions remain
established, but no new sessions are permitted until usage falls below the newly-configured value.
Usage Guidelines Use this command in PDG Service Configuration Mode to bind the IP address used as the connection point
for establishing IKEv2/IPSec sessions to a crypto template. You can also use it to define the maximum number
of sessions the PDG service supports.
Examples The following command binds a PDG service with an IP address of 10.2.3.4 to the crypto template
crypto_template_1, sets the mode to TTG, and sets the maximum number of sessions to 500000:
bind address 10.2.3.4 crypto-template crypto_template_1 mode ttg max-sessions 500000
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
ip gnp-qos-dscp
Configures the quality of service (QoS) differentiated service code point (DSCP) used when sending data
packets over the Gn' interface in the uplink direction.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
Syntax Description [ no ] ip gnp-qos-dscp { background dscp | conversationaldscp | interactive dscp | streaming dscp |
interactive [ traffic-handling-priority traffic_priority ] { allocation-retention-priority
allocation_retention_priority } } +
default ip gnp-qos-dscp
no
Disables the overriding of the ToS (Type of Service) field and enables the pass-through option.
background dscp
Specifies the DSCP marking to be used for packets of sessions subscribed to the 3GPP background class, in
which the data transfer is not time-critical (for example, in e-mail exchanges). This traffic class is the lowest
QoS.
dscp: Sets the DSCP for the specified traffic class. See the dscp section below.
conversational dscp
Specifies the DSCP marking to be used for packets of sessions subscribed to the 3GPP conversational class,
in which there is a constant flow of traffic in both the uplink and downlink direction. This traffic class is the
highest QoS.
dscp: Sets the DSCP for the specified traffic pattern. See the dscp section below.
allocation-retention-priority allocation_retention_priority
Specifies the DSCP for the interactive class if the allocation priority is present in the QoS profile.
allocation-retention-priority can be the integers 1, 2, or 3.
DSCP uses the values in the following table based on the traffic handling priority and allocation/retention
priority if the allocation priority is present in the QOS profile.
Allocation Priority 1 2 3
Traffic Handling Priority
1 ef ef ef
2 af21 af21 af21
3 af21 af21 af21
Important If you only configure DCSP marking for interactive traffic classes without specifying ARP, it may not
properly take effect. The CLI allows this scenario for backward compatibility however, it is recommended
that you configure all three values.
streaming dscp
Specifies the DSCP marking to be used for packets of sessions subscribed to the 3GPP streaming class, in
which there is a constant flow of data in either in the uplink or downlink direction. This traffic class has a
higher QoS than the interactive class, but not as high as the conversational class.
dscp: Set the DSCP for the specified traffic pattern. See the dscp section below.
dscp
Default:
• background: be
• interactive
• Traffic Priority 1: ef
• Traffic Priority 1: af21
• Traffic Priority 1: af21
• streaming: af11
• conversational: ef
Specifies the DSCP for the specified traffic pattern. dscp can be configured to any one of the following:
+
More than one of the above keywords can be entered within a single command.
Usage Guidelines DSCP levels can be assigned to specific traffic patterns in order to ensure that data packets are delivered
according to the precedence with which they're tagged. The diffserv markings are applied to the IP header of
every subscriber data packet transmitted over the Gn' interface(s).
The four traffic patterns have the following order of precedence: background (lowest), interactive, streaming,
and conversational (highest). Data packets falling under the category of each of the traffic patterns are tagged
with a DSCP that further indicate their precedence as shown in the following tables:
2 Class 1
3 Class 2
4 Class 3
5 Class 4
The DSCP level can be configured for multiple traffic patterns within a single instance of this command.
Examples The following command configures the DSCP level for the streaming traffic pattern to be ef:ip gnp-qos-dscp
streaming ef
The following command configures the DSCP levels for the conversational, streaming, interactive and
background traffic patterns to be ef, af22, and af41, respectively:ip gnp-qos-dscp conversational ef streaming
ef interactive af22 background af41
ip qos-dscp
Configures the quality of service (QoS) differentiated service code point (DSCP) used when sending data
packets over the Wu interface in the downlink direction.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
allocation-retention-priority
Specifies the DSCP for interactive class if the allocation priority is present in the QOS profile.
allocation-retention-priority can be the integers 1, 2, or 3.
DSCP values use the following matrix to map based on traffic handling priority and Alloc/Retention priority
if the allocation priority is present in the QOS profile.
The following table shows the DSCP value matrix for allocation-retention-priority.
qci
Configures the QCI attribute of QoS. Here the qci_val is the QCI for which the negotiate limit is being set, it
ranges from 1 to 9.
dscp
Default QCI:
• 1: ef
• 2: ef
• 3: af11
• 4: af11
• 5: ef
• 6: ef
• 7: af21
• 8: af21
• 9: be
Specifies the DSCP for the specified traffic pattern. dscp can be configured to any one of the following:
+
More than one of the above keywords can be entered within a single command.
Usage Guidelines You can assign DSCP to specific traffic patterns to ensure that data packets are delivered according to the
precedence with which they are tagged. The diffserv markings are applied to the outer IP header of every GTP
data packet. The diffserv marking of the inner IP header is not modified.
The traffic patterns are defined by QCI (1 to 9). Data packets falling under the category of each of the traffic
patterns are tagged with a DSCP that further indicate their precedence as shown in the following tables:
1 Class 1
2 Class 2
3 Class 3
4 Class 4
The DSCP level can be configured for multiple traffic patterns within a single instance of this command.
The no ip qos command can be issued to remove a QOS setting and return it to it's default setting.
Examples The following command configures the DSCP level for QCI to be Expedited Forwarding,ef:
ip qos-dscp qci 1 ef
ip source-violation
Sets the parameters for IP source validation. Source validation is useful if packet spoofing is suspected, or
for verifying packet routing and labeling within the network.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
clear-on-valid-packet
Configures the service to reset the drop-limit counters upon receipt of a properly addressed packet. Default:
disabled
drop-limit num
Sets the maximum number of allowed IP source violations within the detection period before dropping a call.
If num is not specified, the value is set to the default value.
num is an integer from 1 to 1000000. Default: 10
period secs
Sets the detection period (in seconds) for IP source violations as an integer from 1 through 1000000. If secs
is not specified, the value is set to the default value. Default: 120
no ip source-violation clear-on-valid-packet
The drop-limit counters are not reset upon receipt of a properly addressed packet.
Usage Guidelines Source validation is useful if packet spoofing is suspected or for verifying packet routing and labeling within
the network.
Source validation requires the source address of received packets to match the IP address assigned to the
subscriber (either statically or dynamically) during the session.
This function operates in the following manner: When a subscriber packet is received with a source IP address
violation, the system increments the IP source violation drop-limit counter and starts the timer for the IP
source violation period. Every subsequent packet received with a bad source address during the IP source
violation period causes the drop-limit counter to increment. For example, if the drop-limit is set to 10, after
10 source violations, the call is dropped. The detection period timer continues to count throughout this process.
Examples The following command sets the drop limit to 15 and leaves the other values at their default values:
ip source-violation drop-limit 15
max-tunnels-per-ue
Specifies the maximum number of IKEv2/IPSec tunnels allowed per UE by the PDG/TTG. This maximum
number is specified per PDG service.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
integer
Specifies the maximum number of IKEv2/IPSec tunnels allowed per UE as an integer from 1 to 11. Default:
11
default max-tunnels-per-ue
Sets the maximum number of IKEv2/IPSec tunnels allowed per UE to its default value, which is 11.
Usage Guidelines Use this command to set the maximum number of IKEv2/IPSec tunnels allowed per UE.
Examples Use the following command to set the maximum number of IKEv2/IPSec tunnels allowed per UE to 2:
max-tunnels-per-ue 2
plmn id
Configures location specific mobile network identifiers used to help translate local emergency and
service-related numbers. Default is disabled.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
mcc mnc_number
Specifies the mobile country code (MCC) portion of the PLMN identifier as an integer from 200 through 999.
mnc mnc_number
Specifies the mobile network code (MNC) portion of the PLMN identifier as a 2- or 2-digit integer from 00
through 999.
Usage Guidelines The PLMN ID is included in the RAI (Routing Area Identity) field of the PDP Create Request messages sent
to the GGSN. Multiple PDG services can be configured with the same PLMN identifier. Up to five PLMN
IDs can be configured for each PDG service.
Examples The following command configures the PLMN identifier with an MCC of 462 and MNC of 2:
plmn id mcc 462 mnc 02
setup-timeout
Specifies the maximum time allowed to set up a session.
Product PDG/TTG
Command Modes Exec > Global Configuration > Context Configuration > PDG Service Configuration
configure > context context_name > pdg-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdg-service)#
setup-timeout integer
Sets the session setup timeout value (in seconds) as an integer from 2 through 300. Default: 60
default setup-timeout
Sets or restores the default session setup timer value to 60 seconds.
Usage Guidelines The PDG/TTG clears both the user session and tunnels if a call does not initiate successfully before the session
setup timer expires.
Examples The following command sets the session setup timeout value to the default value of 60 seconds:
default setup-timeout
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
a11-signalling-packets
Applies DSCP marking for IP header carrying outgoing A11-signalling packets.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the a11-signalling-packets ip-header-dscp option configuration.
default
Sets / Restores default value assigned for specified parameter a11-signalling-packets ip-header-dscp.
hexa_number
a Hexa decimal number between 0x0 and 0x3F.
Usage Guidelines Use this command to configured value of DSCP to be set for all outgoing A11 signaling msg.
By default the CLI is disabled and DSCP is marked as 0 in ip-header.
Examples The following command configures value of DSCP to be set for all outgoing to A11 signaling message 0x3F:
a11-signalling-packets ip-header-dscp 0x3F
aaa 3gpp2-service-option
Specifies the value for the 3gpp2-service option.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the aaa 3gpp2-service option configuration.
default
Sets / Restores default value assigned for specified parameter aaa 3gpp2-service-option.
number
Service option number is integer and should be between 0 to 32767.
Usage Guidelines Allows the configuration of a default service option value to be sent in accounting when service option values
are not received from PCF. The PDSN will default the service option value to the configured value if the value
is not specified by the PCF.
aaa nas-ip-address
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the aaa nas-ip-address option configuration.
default
Sets / Restores default value assigned for specified parameter aaa nas-ip-address by default this is disabled.
ipv4 address
Specifies the IPv4 addresses to be used.
access-flow traffic-validation
If access-flow traffic-validation is enabled for the service and the subscriber then the flows are checked
against the filter rules. If the packets does not match the filter rules, and N violations occur in K seconds, the
rp connection is downgraded to best-effort flow, if it is not already a best-effort flow.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disable traffic validation for the service.
default
Traffic validation configuration for the service is set to the default value.
Usage Guidelines Use this command to enable traffic validation for the current PDSN service.
Examples The following command enables traffic validation for the current PDSN service and sets the limit allowed to
100 violations within 5 seconds:
access-flow traffic-validation threshold violations 100 interval 5
access-network
Configures access network parameters.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the access-network.
accounting identifier
Configures accounting for the access-network. This value must be a string from 1 to 128 characters in length.
realm realm_name
Configures the realm for the access-network. realm_name must be a string from 1 to 128 characters in length.
Usage Guidelines Use this command to configure access-network parameters for accounting and realms.
airlink bad-sequence-number
Configures PDSN behavior for airlink related parameters.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
airlink bad-sequence-number { accept | deny [use-deny-code { poorly-formed-request |
unsupported-vendor-id}]}
[ no | default ] airlinkbad-sequence-number
no
Disables the deny of bad-sequence number and accept it.
default
It is the default behavior.
accept
Accepts the A11 RRQ messages that have an Airlink Sequence number less than or equal to a previously
received sequence number.
It is the default behavior.
deny
Rejects the A11 RRQ messages that have an Airlink Sequence number less than or equal to a previously
received sequence number.
It uses poorly-formed-request option by default to deny a request.
poorly-formed-request will deny the A11 request on the basis of request formation or structure. It is the
default deny code for deny sub-command.
Usage Guidelines This command is used to configure the airlink parameters for A11 RRQs.
When configured it denies the A11 RRQ messages that have an Airlink Sequence number less than or equal
to a previously received sequence number.
Examples The following command would configure the system to deny all A11 RRQ messages having unsupported
vendor Id or bad structure of message, including those having airlink sequence number less than or equal to
a previously received sequence number:
airlinkbad-sequence-number deny
allow alt-ppp
Allows proprietary modified versions of PPP type sessions to connect this PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the allowed alternate PPP feature.
default
Sets the specified parameter to default.
Usage Guidelines This command is used to deviate from standard PPP protocol and use a proprietary modified version of PPP
with a pre-defined non-negotiable PPP parameters.
It is a vendor-specific licensed feature command.
always-on-indication
Enables/disables the inclusion of 3GPP2 Always On Indicators in messages to the PCF.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the sending of 3GPP2 Always On Indication messages.
Usage Guidelines This command is available when the 3GPP2 Always-On RP Extensions feature-use license is installed.
When enabled, this command causes the PDSN service to include the Always On Indicators in the Normal
Vendor Specific Extension (NVSE) part of an A11 Session Update message to the PCF. The indicator will
only be sent for those subscriber sessions in which Always On functionality is enabled as determined after a
successful authentication: the 3GPP2-Always-On attribute is set to a value of 1 (Active) for subscribers
configured on a AAA server, or the always-on parameter is set for locally configured subscribers.
This functionality is enabled by default.
Examples Use the following command to Enables the inclusion of 3GPP2 Always On Indicators in messages to the PCF.
always-on-indication
associate
Associates a PDSN-service with a Quality of Service (QoS) policy.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the configuration to associate PDSN-serivce with qos policy.
qci-qos-mapping string
qci-qos-mapping configures QCI to QoS mapping for this PDSN service.
string a string of size 1 to 63.
Usage Guidelines The following is used for configuration to associate PDSN-serivce with qos policy.
authentication
Configures authentication parameters for specific PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Configures authentication parameters for specific PDSN service.
allow-noauth
Default: Disabled
This option configures the system to provide subscribers with network access even though they have not been
authenticated. This command issued by itself would cause the system to not attempt to authenticate subscribers.
When the allow-noauth option is used in conjunction with commands specifying other authentication protocols
and priorities to use, then if attempts to use those protocols fail, the system will treat the allow-noauth option
as the lowest priority.
If no authentication is allowed, then NAI construct will be implemented in order to provide accounting records
for the subscriber.
chap chap_priority
Default: 1
This option configures the system to attempt to use the Challenge Handshake Authentication Protocol (CHAP)
to authenticate the subscriber.
A chap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
chap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference. CHAP
is enabled by default as the highest preference.
mschap mschap_priority
Default: Disabled
This option configures the system to attempt to use the Microsoft Challenge Handshake Authentication
Protocol (MSCHAP) to authenticate the subscriber.
A mschap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
mschap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference.
pap pap_priority
Default: 2
This option configures the system to attempt to use the Password Authentication Protocol (PAP) to authenticate
the subscriber.
A pap_priority must be specified in conjunction with this option. Priorities specify which authentication
protocol should be attempted first, second, third and so on.
pap_priority must be an integer from 1 through 1000. The lower the integer, the higher the preference. PAP
is enabled by default as the second highest preference.
msid-auth
Default: Disabled
This option configures the system to attempt to authenticate the subscriber based on their Mobile Station
Identity (MSID).
Usage Guidelines Use to specify how the PDSN service should handle authentication and what protocols to use. The flexibility
is given to configure this option to accommodate the fact that not every mobile will implement the same
authentication protocols.
The chassis is shipped from the factory with the authentication options set as follows:
• allow-noauth disabled
• chap enabled with a priority of 1
• mschap disabled
• msid-auth disabled
• pap enabled with a priority of 2
Important At least one of the keywords must be used to complete the command.
Examples The following command would configure the system to allow no authentication for subscribers and would
perform accounting using the default NAI-construct of username@domain:
authentication allow-noauth
The following command would configure the system to attempt subscriber authentication first using CHAP,
then MSCHAP, and finally PAP. If the allow-noauth command was also issued, if all attempts to authenticate
the subscriber using these protocols fail, then the subscriber would be allowed access:
authentication chap 1 mschap 2 pap 3
bcmcs
Sets the BCMCS (Broadcast Multicast Service) group username and password for RADIUS access.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
Syntax Description bcmcs { customptt | encrypted grppasswd group_passwd | flow-id value [flow-id-type { flow-id
| program-id } ] | grppasswd group_password | grpusrnamegroup_name | ptt {
destination-context dest_name | disconnect-dscp-label dscp_label | mtu transmission_unit |
rohc-profile-name rohc_profile_name } }
default bcmcs [ custom ptt | ptt { disconnect-dscp-label | mtu | rohc-profile-name } ]
no bcmcs { custom ptt | flow-id value [flow-id-type { flow-id | program-id } | grppasswd |
grpusrname | ptt { destination-context | disconnect-dscp-label | mtu | rohc-profile-name } }
custom
Customise the BCMCS configuration.
flow-id value
Set the BCMCS flow-id. This value must be a hex string between 0x1000 and 0xFFFFFFFF.
Making this entry opens a new mode: bcmcs-flow-id.
rohc-profile name : Configure ROHC parameters name, name should be string of size 1 to 63.
grpusrname group_name
Sets the BCMCS group name for RADIUS access requests. This value must be a string from 1 to 127 characters
in length.
Usage Guidelines Use this command to set the BCMCS group username and password for RADIUS access requests.
bind
Binds the PDSN service to a logical IP interface serving as the R-P interface. Specifies the maximum number
of subscribers that can access this service over the interface.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Removes a previously configured binding.
address
Specifies the IP address (address) of the interface configured as the R-P interface. address is specified in
dotted decimal notation.
max-subscribers count
Default: 500000
Specifies the maximum number of subscribers that can access this service on this interface.
count can be configured to any integer value between 0 and 2500000.
Important The maximum number of subscribers supported is dependant on the license key and the number of active
PACs/PSCs installed in the system. A fully loaded system with 13 active PACs/PSCs can support 2500000
total subscribers. Refer to the license key command for additional information.
Usage Guidelines Associate or tie the PDSN service to a specific logical IP address. The logical IP address or interface takes
on the characteristics of an R-P interface. Only one interface can be bound to a service. The interface should
be configured prior to issuing this command.
This command also sets a limit as to the number of simultaneous subscribers sessions that can be facilitated
by the service/interface at any given time.
When configuring the max-subscribers option, be sure to consider the following:
• The total number of interfaces that you will configure for use as R-P interfaces
• The maximum number of subscriber sessions that all of the interfaces may handle during peak busy
hours
• The average bandwidth for each of the sessions
• The type of physical port (10/100Base-Tx or 1000Base-T) to which these interfaces will be bound
Taking these factors into account and distributing your subscriber session across all available interfaces will
allow you to configure your interfaces to optimally handle sessions without degraded performance.
Examples The following command would bind the logical IP interface with the address of 192.168.3.1 to the PDSN
service and specifies that a maximum of 600 simultaneous subscriber sessions can be facilitated by the
interface/service at any given time.
bind address 192.168.3.1 max-subscribers 600
The following command disables a binding that was previously configured:
no bind address
data-available-indicator
Enables sending Data Available Indicator extension in R-P Registration Reply.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Default: Disabled
Disable the sending of the Data Available Indicator extension in R-P Registration Reply.
default
Sets / Restores default value assigned for specified parameter for data-available-indicator.
Usage Guidelines Use this command to enable or disable the sending of the Data Available Indicator extension in R-P Registration
Reply
Examples Use the following command to enable sending the Data Available Indicator extension in R-P Registration
Reply:
data-available-indicator
Use the following command to disable sending the Data Available Indicator extension in R-P Registration
Reply:
no data-available-indicator
data-over-signaling
Enables the data-over-signaling marking feature for A10 packets.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets / Restores default value assigned for specified parameter for data-over signaling
no
Default: Enabled
Disable the data-over signaling feature for A10 packets.
Usage Guidelines Use this command to enable or disable the data-over signaling feature for A10 packets.
Examples no data-over-signaling
default subscriber
Specifies the name of a subscriber profile configured within the same context as the PDSN service from which
to base the handling of all other subscriber sessions handled by the PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables the option default subscriber profile_name
profile_name
Specifies the name of the configured subscriber profile. profile_name can be between 1 and 127 alpha and/or
number characters and is case sensitive.
Usage Guidelines Each subscriber profile specifies "rules" such as permissions, PPP settings, and timeout values.
By default, the PDSN service will use the information configured for the subscriber named default within the
same context. This command allows for multiple PDSN services within the same context to apply different
"rules" to sessions they process. Each set of rules can be configured under a different subscriber name which
is pointed to by this command.
Use the no default subscriber profile_name command to delete the configured default subscriber.
Examples To configure the PDSN service to apply the rules configured for a subscriber named user1 to every other
subscriber session it processes, enter the following command:
default subscriber user1
direct-lte-indicator
Enables sending Direct LTE Indicator VSA in Access Request.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets / Restores default value assigned for specified parameter for data-over signaling
no
Default: Enabled
Disables sending Direct LTE Indicator VSA in Access Request.
Usage Guidelines Use this command to enable or disable sending Direct LTE Indicator VSA in Access Request.
Examples no direct-lte-indicator
dormant-transition
Configures the PDSN behavior to terminate A10 session, when the PDSN receives the A11-RRQ (Type 4)
before the session for the original MN is established completely.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Terminates the A10 session, when PDSN receives the A11-RRQ (Type 4) before the original session established
completely.
default
Keeps the A10 session live in case of A11-RRQ (Type 4) is received before the original session is established
completely.
Usage Guidelines When the status of A10 session goes to dormant before the session for the original MN is established completely,
the different MN may possibly send the A11-RRQ (Type 4) to the PDSN and PPP renegotiation may start.
This command is used to terminate the A10 session when the PDSN receives the A11-RRQ (Type 4) before
the session for original MN is established completely.
Examples Following command is used to release the A10 session in case of receiving A11-RRQ (Type 4) before the
original session is established completely:
no dormant-transition initial-session-setup
end
Exits the current configuration mode and returns to the Exec mode.
Product All
enhanced-pcf-redirection
Enables or disables PDSN support for enhanced PCF redirection.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables PDSN support for enhanced PCF redirection.
enhanced-pcf-redirection
Enables PDSN support for enhanced PCF redirection.
Usage Guidelines Use this command to enable or disable PDSN support for enhanced PCF redirection. By default, this feature
is disabled.
Examples The following command will disable PDSN support for enhanced PCF redirection.
no enhanced-pcf-redirection
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
fragment
Enables or disables PPP payload fragmentation.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the fragmentation of ppp data.
default
Default enables ppp data fragmentation.
Usage Guidelines This command is to indicate to the RP module to NOT fragment PPP payloads being sent to the PCF, if the
total packet size (PPP+GRE+IP) exceeds 1500 bytes.
Disabling fragmentation may cause the sessmgr to perform outer IP fragmentation of the outgoing packet, if
the resulting packet exceeds the MED MTU.
gre
Configures Generic Routing Encapsulation (GRE) parameters for the A10 protocol within the PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the specified functionality.
default
Restores the specified parameter to its default setting.
checksum
Default: disabled
Enables the introduction of the checksum field in outgoing GRE packets.
checksum-verify
Default: disabled
Enables verification of the GRE checksum (if present) in incoming GRE packets.
reorder-timeout
Default: 100
Configures max number of milliseconds to wait before processing reordered out-of-sequence GRE packets.
milliseconds must be an integer from 0 through 5000.
segmentation
Default: disabled
Enables GRE Segmentation for the PDSN service.
• A packet arrives that fills a gap in the sequenced numbers stored in the queue and creates a subset of
packets whose sequence numbers are continuous with the current accepted sequence number. This subset
of packets in the queue is sent for processing. The reorder timer continues to run and the accepted
sequence number is updated to the highest number in the subset delivered.
sequence-numbers
Enables insertion of GRE sequence numbers in data that is about to be transmitted over the A10 interface.
Data coming into the system containing sequence numbers but that is out of sequence is not re-sequenced.
threegppp2-ext-headers qos-marking
When threegppp2-ext-headers qos-marking is enabled and the PCF negotiates capability in the A11 RRQ,
the PDSN will include the qos optional data attribute in the GRE 3gpp2 extension header.
The no keyword, enables qos-marking in the gre header based on the tos value in the header.
Usage Guidelines The gre protocol-type command can be used to prevent the PDSN service from servicing PCFs that use a
specific form of encapsulation.
Use the no gre sequence-numbers command to disable the inclusion of GRE sequence numbers in the A10
data path.
The chassis is shipped from the factory with the authentication options set as follows:
• protocol-type any
• sequence-numbers enabled
Examples Use this command to configure the PDSN service to exclude byte stream encapsulated GRE traffic:
gre protocol-type ppp
inter-pdsn-handoff mobility-event-indicator
Configures the PDSN to support the Mobility Event Identifier (MEI) during inter-PDSN handoffs. The presence
of the Mobility Event Indicator (MEI) and Access Network Identifier (ANID) elements in a A11 handoff
request represents an Inter-PDSN handoff.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables support for the MEI during inter-PDSN handoffs.
default
Sets / Restores default value assigned for inter-pdsn-handoff mobility-event-indicator. By default it is
disabled.
Usage Guidelines Use this command to configure support for the MEI during inter-PDSN handoffs.
Examples Use the following command to enable support for the MEI during inter-PDSN handoffs
inter-pdsn-handoff mobility-event-indicator
inter-pdsn-handover
Configures Inter-PDSN handoff related parameters.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables support for the MEI during inter-PDSN handoffs parameters.
default
Sets / Restores default value assigned for inter-pdsn-handoff mobility-event-indicator. By default it is
disabled.
Usage Guidelines Use this command to configure Inter-PDSN handoff related parameters.
Examples Use the following command to econfigure Inter-PDSN handoff related parameters.
inter-pdsn-handover use-canid-panid
ip header-compression rohc
Enters PDSN Service ROHC Configuration Mode and allows you to configure ROHC parameters that the
PDSN conveys to the PCF in the initial A11 RRP message before PPP authentication.
By default, ROHC is disabled for a PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets all PDSN Service ROHC Configuration Mode values back to the defaults and disable ROHC for this
PDSN service.
no
Disable IP header compression for this PDSN Service.
Usage Guidelines Use this command to enter the PDSN Service ROHC Configuration Mode or disable ROHC for the current
PDSN service.
Examples The following command disables ROHC for the current PDSN service and sets all of the values for commands
in PDSN Service ROHC Configuration Mode back to their default settings:
no ip header-compression rohc
ip local-port
Configures the local User Datagram Protocol (UDP) port for the R-P interfaces' IP socket.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
number
Default: 699
Specifies the UDP port number.
number can be any integer value between 1 and 65535.
default
Designates UDP port, default value as 699.
Usage Guidelines Specify the UDP port that should be used for communications between the Packet Control Function (PCF)
and the PDSN.
Important The UDP port setting on the PCF must match the local-port setting for the PDSN service on the system
in order for the two devices to communicate.
Examples Use the following command to specify a UDP port of 3950 for the PDSN service to use to communicate with
the PCF on the R-P interface:
iplocal-port 3950
ip source-violation
Sets the parameters for IP source validation. Source validation is useful if packet spoofing is suspected or for
verifying packet routing and labeling within the network.
Source validation requires the source address of received packets to match the IP address assigned to the
subscriber (either statically or dynamically) during the session.
Product PDSN
PDIF
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
Syntax Description ip source-violation { clear-on-valid-packet | drop-limit num | period secs | reneg-limit num }
no ip source-violation clear-on-valid-packet
default ip source-violation { drop-limit | period | reneg-limit }
no
Enables/Disables ip source-violation clear-on-valid-packet.
default
Configure default settings related to ip source-violation.
clear-on-valid-packet
Default: disabled
Configures the service to reset the reneg-limit and drop-limit counters after receipt of a properly addressed
packet.
drop-limit num
Default: 10
Sets the number of allowed source violations within a detection period before forcing a call disconnect. If
num is not specified, the value is set to the default.
period secs
Default: 120
The length of time, in seconds, for a source violation detection period to last. drop-limit and reneg-limit
counters are decremented each time this value is reached.
The counters are decremented in this manner: reneg-limit counter is reduced by one (1) each time the period
value is reached until the counter is zero (0); drop-limit counter is halved each time the period value is reached
until the counter is zero (0). If secs is not specified, the value is set to the default.
secs can be any integer value from 1 to 1000000.
reneg-limit num
Default: 5
Sets the number of allowed source violations within a detection period before forcing a PPP renegotiation. If
num is not specified, the value is set to the default.
num can be any integer value from 1 to 1000000.
Usage Guidelines This function is intended to allow the operator to configure a network to prevent problems such as when a
user gets handed back and forth between two PDIFs PDSNs a number of times during a handoff scenario.
This function operates in the following manner:
When a subscriber packet is received with a source address violation, the system increments both the IP
source-violation reneg-limit and drop-limit counters and starts the timer for the IP-source violation period.
Every subsequent packet received with a bad source address during the IP-source violation period causes the
reneg-limit and drop-limit counters to increment.
For example, if reneg-limit is set to 5, then the system allows 5 packets with a bad source address (source
violations), but on the 5th packet, it re-negotiates PPP.
If the drop-limit is set to 10, the above process of receiving 5 source violations and renegotiating PPP occurs
only once. After the second 5 source violations, the call is dropped. The period timer continues to count
throughout this process.
If the configured source-violation period is exceeded at any time before the call is dropped, the reneg-limit
counter is checked. If the reneg-limit counter is greater than zero (0), the reneg-limit is decremented by 1. If
the reneg-limit counter equals zero, the drop-limit is decremented by half.
Examples The following command sets the drop limit to 15 and leaves the other values at their defaults:
ip source-violation drop-limit 15
lifetime
Specifies the time that an A10 connection can exist before its registration is considered expired.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no lifetime
Specifies that an A10 connection can exist for an infinite amount of time.
default lifetime
Sets / Restores default value assigned for lifetime as 1800.
time
Default: 1800
Specifies the time that an A10 connection can exist before its registration is considered expired.
time is measured in seconds and can be configured to any integer value between 1 and 65534.
Usage Guidelines Set a limit to the amount of time that a subscriber session can remain up whether or not the session is active
or dormant. If the lifetime timer expires before the subscriber terminates the session, their connection will be
terminated automatically.
Use the no lifetime command to delete a previously configured lifetime setting. If after deleting the lifetime
setting you desire to return the lifetime parameter to its default setting, use the default lifetime command.
Examples The following command specifies a time of 3600 seconds (1 hour) for subscriber sessions on this PDSN
service:
lifetime 3600
max-retransmissions
Configures the maximum number of times the PDSN service will attempt to communicate with a PCF before
it marks it as unreachable.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets / Restores default value assigned for max-retransmissions as 5.
count
Specifies the maximum number of times the PDSN service will attempt to communicate with a PCF before
it marks it as unreachable.
count can be configured to any integer value between 1 and 1,000,000.
Usage Guidelines If the value configured for the max-retransmissions is reached the call will be dropped.
The chassis is shipped from the factory with the Internet maximum number of retransmissions set to 5.
Examples The following command configures the maximum number of retransmissions for the PDSN service to 3:
max-retransmissions 3
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables mobile-ip foreign-agent context
context_name
Specifies the name of the previously configured context that facilitates the FA service(s).
context_name must be between 1 and 79 alpha or numeric characters and is case sensitive.
fa-service name
This optional keyword allows you to link the PDSN service to a particular FA service in the specified context.
name is the name of the FA service to link to. name is a string of size 1 to 63
Usage Guidelines FA services on the system can be configured either in the same or different contexts from those facilitating
PDSN services. When they are configured in separate contexts, this command configured with a PDSN service
instructs the PDSN service to route traffic to the context facilitating the FA service.
Use the no mobile-ip foreign-agent context to delete a previously configured destination context.
Examples The following command instructs the PDSN service to use the context named FA-destination for FA
functionality:
mobile-ip foreign-agent context fa-destination
mobile-ipv6
Configures Mobile IPv6 parameters within specific PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
mobile-access-gateway
Configures Mobile Access Gateway (MAG) parameters within specific PDSN service.
context context_name
Designates name of the context in which the MAG service is configured. Must be followed by context name
of MAG service.
context_name is a string of size 1 to 79.
mag-service name
Designates name of the MAG service in that context. Must be followed by MAG service name.
name is a string of size 1 to 63.
Usage Guidelines This command is used to configure Mobile IPv6 parameters and Mobile Access Gateway (MAG) parameters
within specific PDSN service.
Examples The following command configures Mobile IPv6 parameters and Mobile Access Gateway (MAG) parameters
within specific PDSN service.
mobile-ipv6 mobile-access-gateway context pdsn1 mag-service serv1
msid length
Configures checking the length of the A11 MSID in A11 Session Specific Extn and airlink records.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Specifies the default length of MSID (10 to 15) as per standard. By default msid is disabled.
min min_length
Specifies the minimum length for MSID.
min_length is any Integer value between 10 to 15, but should be less than max_length specified with max.
Default is 10.
max max_length
Specifies the maximum length for MSID.
max_length is any Integer value between 10 to 15, but should be more than min_length specified with min.
Default is 15.
Usage Guidelines MSID length can be configured either in the standard length or different customized length form. This command
is used to specify the allowed length of MSID.
Examples The following command specifies an MSID length between 12 and 15:
msid length min 12 max 15
nai-construction
Specifies a domain alias that will be used to represent the context which the PDSN service should use for
AAA functionality.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
domain alias
Alias represents the "domain" name that you would like to associate with the context in which AAA
functionality is configured. alias can be between 1 and 79 alpha and/or numeric characters and is case-sensitive.
Usage Guidelines Enabling NAI will be constructed for the subscriber in the event that their mobile station (MS) does not
negotiate CHAP, PAP, or MSCHAP. If this option is selected, no further attempts will be made to authenticate
the user. Instead, the constructed NAI will be used for accounting purposes.
The context specified by this command would be used to provide the communication with the RADIUS
accounting server.
Use the no nai-constructed domain command to deleted a configured alias.
Important This command should only be used if the PDSN service is configured to allow no authentication using
the authentication allow-noauth command.
Additionally, the aaa constructed-nai command in the Context Configuration mode can be used to configure
a password for constructed NAIs.
Examples The following command configured a domain alias of aaa_context for the PDSN service to use when an NAI
is constructed for a subscriber session:
nai-construction domain aaa_context
new-call conflict
Enable or disable to send A11-RUPD to current PCF, when system receives the A11-RRQ(Type1) from new
PCF during the session exists.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disable to send A11-RUPD to current PCF, when system receives the A11-RRQ(Type1) from new PCF
during the session exists.
default
Enable to send A11-RUPD to current PCF, when system receives the A11-RRQ(Type1) from new PCF during
the session exists.
Usage Guidelines This configuration supports to enable or disable to send A11-RUPD to current PCF, when the system receives
the A11-RRQ(Type1) from new PCF during the session exists.
If the configuration is no new-call conflict terminate-session-old-pcf system will not send registration update
to old PCF on receiving a new call (A11-RRQ(Type1)) request for an existing active/dormant session. The
default behavior is to send registration updates.
Examples The following command configured a system to send a registration update on receiving an A11-RRQ (Type
1) request for an existing active/dormant session:
new-call conflict terminate-session-old-pcf
pcf-monitor
Enables the monitoring of all the PCFs that have sessions associated with it. The PDSN stops monitoring a
PCF if it is determined to be down. Once a PCF is determined to be down, the PDSN tears down all sessions
that correspond to the PCF and generates AAA Accounting Stop messages. All the PCFs that are connected
to the PDSN service are monitored.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
Syntax Description pcf-monitor[ interval seconds | max-inactivity-time seconds | num-retry num | timeout seconds
]
[ no | default ] pcf-monitor
pcf-monitor
Entering the command with no keywords enables the PCF monitoring function with all parameters set to the
defaults.
no
Disables the pcf monitoring function.
default
Sets / Restores default value assigned for pcf-monitor.
interval seconds
Default: 60 seconds
Sets the amount of time to wait between ping request messages.
seconds must be an integer in the range from 1 through 3600.
max-inactivity-time seconds
Default: 120 seconds
The maximum amount of time (seconds) with no A10 traffic from a PCF before the ICMP-ping mechanism
is triggered.
seconds must be an integer from 1 through 3600.
num-retry num
Default: 5
Sets the number of times that the PDSN retries to ping the PCF. When num-retry for a given PCF has been
exhausted with no response, sessions that correspond to the non-responsive PCF are terminated and Accounting
Stop records for each terminated session are generated.
num must be an integer in the range from 0 through 100.
timeout seconds
Default: 3 seconds
The amount of time to wait for a response before retrying.
seconds must be in the range from 1 through 10.
Usage Guidelines Use this command to enable the PDSN service to monitor the PCFs that have sessions associated with the
PDSN service.
Examples The following command enables PCF monitoring with parameters set to the defaults:
pcf-monitor
The following command enables PCF monitoring and sets the timeout to 10 seconds:
pcf-monitor timeout 10
The following command disables pcf-monitoring:
no pcf-monitor
pcf-session-id-change restart-ppp
Manages current session and PPP renegotiation on GRE-key change without any change in
PCF/PANID/CANID. This command disables or enables the PPP renegotiation restart on receiving an RP
registration request from the current PCF with GRE key (PCF session Id) change. With this command the
PDSN aborts and restarts the call causing PPP renegotiation.
This is enabled by default.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables the pcf-session-id-change restart-ppp function.
With this option PDSN does not restart the PPP renegotiation on GRE key change from current PCF in an
RP registration request, unless it indicates change in PCF/PANID/CANID.
default
Set the pcf-session-id-change function to the default state on enabled.
Usage Guidelines GRE key (PCF session ID) is sued to identify the data packet for a session and is negotiated through the A11
signaling messages between PCF and PDSN. By default PDSN aborts and restart the PPP renegotiation on
receipt of any RP registration request with change in GRE key or PCF session Id.
With use of no pcf-session-id-change restart-ppp command PDSN is configured to disable the restart of call
or PPP renegotiation on receipt of any RP registration request with changed GRE key, unless it has any
PCF/ANID/CANID change. PDSN silently switches the GRE key for the session, retaining the existing PPP
session.
Examples The following command disables the PPP renegotiation restart action on receipt of any RP RRQ with changed
GRE key from same PCF/PANID/CANID.
no pcf-session-id-change restart-ppp
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disables pdsn type0-tft attempt-inner-match.
default
Sets / Restores default value assigned for pdsn type0-tft attempt-inner-match.
Usage Guidelines This CLI is used make PDSN match inner IP packets for an AIMS call. When enabled, the PDSN tries to
match a type-0 tft to match both outer and inner packet, so that MN can use a Type-0 filter for HoA traffic
which are tunneled.
This is disabled by default.
peer-pcf
Configures settings for any PCF that has a connection with this PDSN.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
ip_address | ip_address/mask
ip_address must be specified using the standard IPv4 dotted decimal notation or colon notation for IPv6.
ip_address/mask must be specified using the standard IPv4 dotted decimal notation or colon notation for IPv6,
followed by the mask.
Usage Guidelines Use this command to configure the settings for any PCF that is connected to this PDSN. You can also specify
bcmcs framing settings to use for the connection.
Examples The following command configures the peer-pcf for an IP address of 131.2.3.4:
peer-pcf 131.2.3.4
pma-capability-indicator
Enables sending PMIP Capability Indicator VSA in Access Request.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
3gpp2
Use 3GPP2 defined VSA. Default is to use Custom1 VSA.
no
Enables/Disables sending PMIP Capability Indicator VSA in Access Request.
default
Sets / Restores default value assigned for PMIP Capability Indicator.
Usage Guidelines Use this command to enable sending PMIP Capability Indicator VSA in Access Request.
Examples The following command enables sending PMIP Capability Indicator using 3GPP2 defined VSA in Access
Request.
pma-capability-indicator 3gpp2
policy
Configures PDSN service policies.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
Syntax Description policy msid-match msid_with_wildcards redirect address [ weight weight_num ] [ address2 [
weight weight_num ]... address16 [ weight weight_num ] ] [ weight weight_num ]
no policy msid-match msid_with_wildcards
policy overload { redirect address [ weight weight_num ] [ address2 [ weight weight_num ] ...
address16 [ weight weight_num ] ] | reject [ use-reject-code { admin-prohibited |
insufficient-resources } ] }
no policy overload [ redirect address [ address2 ... address16 ] ]
default policy overload
policy pcf-zone-match zone_number redirect address [ weight weight_num ] [ address2 [ weight
weight_num ] ... address16 [ weight weight_num ] ] | restricted [ redirect address [ weight
weight_num ] [ address2 [ weight weight_num ] ... address16 [ weight weight_num ] ]
no policy pcf-zone-matchzone_number
[ default | no ] policy rrq mei-from-current-pcf suppress-ppp-restart
policy service-option enforce
[ default | no ] policy service-option
policy unknown-cvse enforce
[ default | no ] policy unknown-cvse
no
Enables/Disables the PDSN service policies.
default
Sets / Restores default value assigned for specified PDSN service policies.
policy overload { redirect address [ weight weight_num ] [ address2 [ weight weight_num ] ... address16 [
weight weight_num ] ] | reject [ use-reject-code { admin-prohibited | insufficient-resources } ] }
Specifies how a PDSN service should handle an overload condition.
redirect: This option enables a redirect policy for overloading conditions. When a redirect policy is invoked,
the PDSN service rejects new sessions with an A11 Registration Reply Code of 88H (unknown PDSN address)
and provides the IP address of an alternate PDSN. This command can be issued multiple times.
address: The IP address of an alternate PDSN expressed in IP v4 dotted decimal notation. Up to 16 IP addresses
can be specified either in one command or by issuing the redirect command multiple times. If you try to add
more than 16 IP addresses to the redirect policy the CLI issues an error message. If you specify an IP address
and weight that already exists in the redirect policy the new values override the existing values.
weight weight_num: When multiple addresses are specified, they are selected in a weighted round-robin
scheme. Entries with higher weights are more likely to be chosen. If a weight is not specified the entry is
automatically assigned a weight of 1. weight_num must be an integer from 1 through 10.
reject: This option will cause any overload traffic to be rejected. The PDSN will send an A11 Registration
Reply Code of 82H (insufficient resources).
use-reject-codeadmin-prohibited: When this keyword is specified and traffic is rejected, the error code
admin prohibited is returned instead of the error code insufficient resources. This is the default behavior.
use-reject-codeinsufficient-resources: When this keyword is specified and traffic is rejected, the error code
insufficient resources is returned instead of the error code admin prohibited.
address: The IP address of an alternate PDSN expressed in IP v4 dotted decimal notation. Up to 16 IP addresses
can be specified either in one command or by issuing the redirect command multiple times. If you try to add
more than 16 IP addresses to the redirect policy the CLI issues an error message. If you specify an IP address
and weight that already exists in the redirect policy the new values override the existing values.
weight weight_num: When multiple addresses are specified, they are selected in a weighted round-robin
scheme. Entries with higher weights are more likely to be chosen. If a weight is not specified the entry is
automatically assigned a weight of 1. weight_num must be an integer from 1 through 10.
restricted: This is an optional keyword which means the zone is restricted. Restricted zone is meaningful
only if enhanced PCF redirection feature is enabled, otherwise the zone follows the default behavior.
Usage Guidelines Policies can be implemented to dictate PDSN service behavior for various conditions such as overloading.
The system invokes the overload policy if the number of calls currently being processed exceeds the licensed
limit for the maximum number of sessions supported by the system.
The system automatically invokes the overload policy when an on-line software upgrade is started.
Use the no policy { overload | service-option } command to delete a previously configured policy. If after
deleting the policy setting you desire to return the policy parameter to its default setting, use the default policy
command.
The chassis is shipped from the factory with the policy options set as follows:
• overload disabled
• sequence-numbers enforced enabled
Caution Incorrect configuration of the policy msid-match and policy pcf-zone-match keywords could result in
sessions failing to be established. For example, if PDSN1 is configured to redirect sessions to PDSN2
while PDSN2 is configured to redirect sessions to PDSN1, a loop is created in which all sessions would
fail to be connected. In addition, sessions will not be established if the PDSN to which the sessions are
being redirected is unavailable.
Examples The following command configures the PDSN service to redirect traffic to two different destinations with
weights of 1 and 10 respectively:
policy overload redirect 192.168.1.100 weight 1 192.168.1.200 weight 10
ppp
Sets PPP tunneling parameters for subscribers in the current PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables the PPP tunneling parameters for subscribers in the current PDSN service.
default
Sets / Restores default value assigned for PPP tunneling parameters for subscribers in the current PDSN
service.
tunnel-context context_name
The name of the context that has a LAC service configured to handle all tunnels from this PDSN service.
Important If the context specified by the ppp tunnel-context context_name command does not have a LAC service
configured and tunnel-type is set to l2tp or the call is rejected.
Important If the PPP tunnel context has not been set or has been cleared with the no ppp tunnel-context command
and tunnel-type is set to l2tp, the context where the current PDSN service resides is used. If that context
does not have a LAC service configured the call is rejected.
Usage Guidelines Use this command to enable or disable forced L2TP tunneling for all subscribers using this PDSN service.
Also use this command to define which context defines the L2TP tunneling parameters.
Examples To set the tunnel context to the context named context1 and enable forced L2TP tunneling, use the following
commands;
ppp tunnel-context context1
ppp tunnel-type l2tp
To enable forced L2TP tunneling with IPSEC security, use the following commands;
ppp tunnel-type l2tp-secure
To disable forced tunneling, use the following command;
ppp tunnel-type none
To clear the setting for the tunnel context, use the following command;
no ppp tunnel-context
qos-profile-id-mapping
Creates the customized QoS profile identifier to QoS mapping for IMS authorization support.
Product PDSN
HA
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Configures the specified QoS profile ID for QoS mapping with default values in this PDSN service.
no
Removes the configured QoS profile ID mapping in this PDSN service.
profile-id id_num
Specifies the profile identifier for QoS parameters to be used as the customized profile ID or modifies the
QoS parameters in a profile ID (id_num) coming from RAN.
id_num must be an integer between 0 and 65535.
description desc
Specifies the user defined description for profile identifier.
desc must be an alpha and/or numeric string between 1 and 32 characters.
downlink-bw dl_bw
Default: 32
Specifies the downlink (towards the MN) data traffic bandwidth in kilo-bits per second for this QoS profile.
dl_bw must be an integer value between 0 and 100000.
drop-rate drop_percentage
Default: 0
Specifies the permitted packet drop rate in percentage for traffic flow to this QoS profile.
drop_percentage must be an integer value between 0 and 1000.
latency latency_duration
Default: 1000
Specifies the permitted latency duration in milli-seconds for this QoS profile.
latency_duration must be an integer value between 0 and 1000.
uplink-bw ul_bw
Default: 32
Specifies the uplink (from the MN) data traffic bandwidth in kilo-bits per second for this QoS profile.
ul_bw must be an integer value between 0 and 100000.
+
More than one of the above keywords can be entered within a single command.
Usage Guidelines Use this command to define the values associated with the profile ID on the PDSN. This profile ID is used
during the mapping to and from the authorized QoS to the QoS parameters for the A10 link. This mapping is
required because the PDSN only knows the profile IDs and not the actual configured values for the profile
ID in the RAN. Also this configuration allows the use of custom profile IDs for the subscribers.
If no values are defined with a QoS profile ID, the values from matching QoS profile ID from RAN will be
applicable to the subscriber traffic.
Examples The following command sets the downlink bandwidth to 32 kbps, latency duration as 1000 ms, uplink bandwidth
to 32 kbps, and QoS class to Class-C for the QoS profile ID 11 in a PDSN service:
default qos-profile-id-mapping profile-id 11
qos update
Sets QoS update parameters for policy mismatches or wait timeouts.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables the qos-update [ policy-mismatch | wait-timeout].
default
Sets / Restores default value for qos-update [ policy-mismatch | wait-timeout].
policy-mismatch
PDSN raises a TFT violation if there is a QoS policy mismatch.
Usage Guidelines This command provides a PDSN service level configurable to configure an action, if the PCF ignores the QoS
Update request from PDSN. It sets the amount of time to wait and the action to take, if no RRQ is received
before the timeout. The action can be to drop packets for the flow, disconnect the session or to downgrade to
best effort.
Important This command is customer-specific. Contact your Cisco account representative for more information.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables the RADIUS accounting related configuration for dropped packets.
Usage Guidelines Use this command to enable or disable the RADIUS accounting related configuration for dropped packets.
By default, the feature is disabled.
Important The configuration will be picked up during call-setup and can not be changed dynamically.
Examples The following command enables the RADIUS accounting related configuration for dropped packets for the
PDSN service:
no radius accounting dropped-pkts
registration-accept
Allows the PDSN to accept registration requests when a handoff disconnect is in progress. When the PDSN
is tearing down a session and the MN moves over to a new PCF and initiates a new session, the PDSN by
default does not accept the handoff until it tears down the old session.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disable accepting of registration requests when a handoff disconnect is still in progress.
default
Default is disabled.
Sets / Restores default value assigned for registration-accept handoff session-disconnect-in-progress.
Usage Guidelines Use this command to allow the PDSN service to accept registration requests when a handoff disconnect is
still in progress.
registration-ack-deny terminate-session-on-error
Configures the PDSN service to terminate an A11 session when a Registration ACK received from the PCF
has an error status.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Disable terminating A11 sessions on a Registration ACK error from the PCF.
default
Sets / Restores default value assigned to registration-ack-deny terminate-session-on-error.
Usage Guidelines Use this command to enable the PDSN service to terminate A11 sessions on a Registration ACK error from
the PCF.
Examples Use the following command to enable this functionality in the PDSN:
registration-ack-deny terminate-session-on-error
registration-deny
Configures parameters related to registration rejection.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets / Restores default value for registration-deny.
no
Disables the specified option.
max-deny-reply-limit num
Default: 3
Configures max number of retries of erroneous registration request message from PCF for a session before
PDSN terminates the session. num can be from 1 to 10.
mismatched-coa-source-address
Default: disabled
Denies RP requests which have a care-of-address field that is different from the request source address.
session-already-active
PDSN denies Registration requests for sessions that are already active with the error code "poorly formed
request" .
session-already-closed
PDSN denies RP renew and dereg requests with error code 0x8E for absent R-P sessions.
session-already-dormant
PDSN denies Registration requests for sessions that are already dormant with the error code "poorly formed
request" .
terminate-session-on-error
Default: Disabled.
Configures PDSN to terminate session if erroneous registration request message is received for the session.
use-zero-gre-key
Configures the PDSN to set the GRE key to zero (0) when denying a new R-P session.
Usage Guidelines Use this command to configure parameters relating to the rejection of registration requests.
Examples To reject calls that do not have the airlink setup record in the RRQ, enter the following command:
registration-deny new-call connection-setup-record-absent
To reject calls if the GRE key collides with that of another user, enter the following command:
registration-deny new-call reverse-tunnel-unavailable
To set the GRE key to 0 (zero) when a new R-P session is denied, enter the following command:
registration-deny new-call use-zero-gre-key
registration-discard
Configures the PDSN service to discard any Registration Request message containing multiple information
elements of the same type or a different GRE key for existing IMSI session.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets/Restores default value assigned for registration-discard .
no
Disables the discarding of Registration request messages containing multiple information elements or different
GRE keys.
bad-extension
Default: Disabled
Configures the PDSN to discard Registration Request message containing multiple information elements of
same type.
gre-key-change
Default: Disabled
Configures PDSN to discard Registration Request message containing different GRE key for existing IMSI
session. Default is disable
handoff connection-setup-record-absent
Default: Disabled
When enabled, discards A11 Handoff requests that do not contain the Airlink Setup record.
Usage Guidelines Use this command to configure the PDSN service to discard and Registration Requests that contain multiple
information elements of the same type or discard Registration Requests that contain GRE keys that have
different GRE keys for the existing IMSI session.
Examples To configure the PDSN service to discard of Registration Requests that have multiple information elements
of the same type, enter the following command:
registration-discard bad-extension
To configure the PDSN service to discard registration Requests that contain a GRE key that is different than
the existing one for the existing IMSI session, enter the following command:
registration-discard gre-key-change
registration-update
Configures registration update related parameters for the PDSN.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
If this option is used with the pdsn-code-nvse keyword, then pdsn-code-nvse configuration is disabled.
If this option is used with the wait-timeout keyword, a separate A11 timer is not used. The PDSN waits for
the ppp retransmit-timeout and then sends the A11 Update. If a value is provided, then the "ppp
retransmit-timeout" is ignored and a separate A11 timeout is started immediately upon sending the LCP
Term-Ack. The A11 Update is then sent when the timer expires.
A value of 0 sends the A11 Update immediately after sending the LCP Term-Ack.
default
Sets/Restores default value assigned for registration-update { pdsn-code-nvse | wait-timeout }
pdsn-code-nvse
Adds the PDSN code NVSE in all A11 registration update messages.
secs
The number of seconds to wait. secs must be an integer in the range from 0 through 16.
wait-timeout
After the Mobile Node terminates a PPP session between the PDSN and the Mobile Node, the PDSN service
waits for the specified time period to receive an A11 RRQ from the PCF before it sends out a
Registration-Update to clear the Session from the PCF.
Examples Use the following command to set the registration wait-timeout to 16 seconds:
registration-update wait-timeout 16
retransmission-timeout
Configures the maximum allowable time for the PDSN service to wait for a response from the PCF before it:
Attempts to communicate with the PCF again (if the system is configured to retry the PCF)
OR
Marks the PCF as unreachable.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables the retransmission-timeout.
default
Sets / Restores default value assigned for retransmission-timeout.
time
Specifies the maximum allowable time for the PDSN service to wait for a response from the PCF before it a)
attempts to communicate with the PCF again (if the system is configured to retry the PCF) or b) marks the
PCF as unreachable.
time is measured in seconds and can be configured to any integer value between 1 and 1,000,000.
Usage Guidelines Use the retransmission timeout command in conjunction with the max-retransmissions command in order
to configure the PDSN services behavior when it does not receive a response from a particular PCF.
Use the no retransmission-timeout command to delete a previously configured timeout value. If after deleting
the lifetime setting you desire to return the lifetime parameter to its default setting, use the default
retransmission-timeout command.
The chassis is shipped from the factory with the retransmission timeout set to 3 seconds.
service-option
If the service option policy is enabled, this command specifies the service options supported by the PDSN
service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Enables/Disables the service-option number
default
Sets / Restores default value assigned for service-option.
number
Default: 7, 15, 22, 23, 24, 25, 33, 59, 67
Specifies a specific Service Option (SO) number that this PDSN service is allowed to support.
number can be configured to any integer value between 1 and 1000.
Usage Guidelines Use the service option command in conjunction with the policy service option enforce command to configure
specific SO numbers that are supported. If a particular SO number is not configured, then any subscriber
session received with that SO number will be rejected and an A11 Registration Reply Code of 86 (poorly
formed request) will be sent.
By default, PDSN services are configured to support the following service option numbers:
• 7: PCF specific
Important Option 67 is used for auxiliary connections for Rev-A calls. PPP encapsulation of data packets does not
flow over this service option connection. ROHC can be performed without PPP for this service option.
Use the no service-option number command to delete a previously configured service option. If after deleting
the service option setting you desire to return the service option parameter to its default setting, use the default
service-option command.
setup-timeout
Specifies the maximum amount of time allowed for session setup.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
default
Sets/Restores default value assigned for setup-timeout.
seconds
Default: 60 seconds
The maximum amount of time, in seconds, to allow for setup of a session. seconds must be an integer from
1 through 1000000
Usage Guidelines Use this command to set the maximum amount of time allowed for setting up a session.
Examples Use the following command to set the maximum time allowed for setting up a session to 300 seconds:
setup-timeout 300
simple-ip allow
Enables or disables Simple-IP sessions from making a connection before authorization takes place.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
When a session attempts PPP authentication, it is assumed that it is a Simple-IP session and it is disconnected
before the user is authenticated (RADIUS or local authentication). Also, if allow-noauth is enabled and PPP
authentication is not performed, after IPCP the session is disconnected if it is discovered that it is a Simple-IP
session.
default
Reset this command to allow Simple-IP sessions to connect.
Usage Guidelines Use this command to prevent Simple-IP sessions from connecting to a PDSN service.
Examples The following command configures the PDSN service so that it will reject any Simple-IP sessions:
no simple-ipallow
The following command configures the PDSN service to allow Simple-IP sessions:
simple-ip allow
spi
Configures the security parameter index (SPI) between the PDSN service and the PCF. This command also
configures the redirection of call based on PCF zone.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
spi-number number
Specifies the SPI (number) which indicates a security context between the PCF and the PDSN in accordance
with IOS 4.1 and RFC 2002.
number can be configured to any integer value between 256 and 4294967295.
encrypted secret enc_secret: Specifies the encrypted shared key (enc_secret) between the PCF and the PDSN
service. enc_secret must be between 1 and 254 alpha and/or numeric characters and is case sensitive.
secret secret: Specifies the shared key (secret) between the PCF and the PDSN services. secret must be
between 1 and 127 alpha and/or numeric characters and is case sensitive.
The encrypted keyword is intended only for use by the chassis while saving configuration scripts. The system
displays the encrypted keyword in the configuration file as a flag that the variable following the secret
keyword is the encrypted version of the plain text secret key. Only the encrypted secret key is saved as part
of the configuration file.
description string
This is a description for the SPI. string must be an alpha and or numeric string of from 1 through 31 characters.
timestamp-tolerance tolerance
Default: 60
Specifies the allowable difference (tolerance) in timestamps that is acceptable. If the difference is exceeded,
then the session will be rejected. If this is set to 0, then time stamp tolerance checking is disabled at the
receiving end.
tolerance is measured in seconds and can be configured to any integer value between 0 and 65535.
zone zone_id
Specifies the different PCF zones to configure in PDSN service. Mapping of a zone-number to a set of PDSNs
can be done per PDSN service basis.
zone_id must be an integer value between 1 and 32. A maximum of 32 PCF zones can be configured for a
PDSN service.
Usage Guidelines An SPI is a security mechanism configured and shared by the PCF and the PDSN service. Please refer to IOS
4.1 and RFC 2002 for additional information.
Multiple SPIs can be configured if the PDSN service is communicating with multiple PCFs.
Important The SPI configuration on the PCF must match the SPI configuration for the PDSN service on the system
in order for the two devices to communicate properly.
Examples The following command configures the PDSN service to use an SPI of 256 when communicating with a PCF
with the IP address 192.168.0.2. The key that would be shared between the PCF and the PDSN service is
q397F65.
spi remote-address 192.168.0.2 spi-number 256 secret q397F65
The following command deletes the configured SPI of 400 for an PCF with an IP address of 172.100.3.200:
no spi remote-address 172.100.3.200 spi-number 400
The following command creates the configured SPI of 400 for an PCF with an IP address of 172.100.3.200
and zone id as 11:
spi remote-address 172.100.3.200 spi-number 400 zone 11
tft-validation wait-timeout
Configures the TFT validation wait timeout value for QoS changes. The QoS update timer triggers automatic
QoS updates based on dynamic policies.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Removes the wait-timeout timer.
default
Sets / Restores default value assigned for tft-validation wait-timeout.
Usage Guidelines Configures the TFT validation wait time value for A11 RRQ for QoS changes. seconds must be an integer
from 1 through 65535.
Examples Use the following command to set the TFT validation wait-timeout to 5 seconds:
tft-validation wait-timeout 5
threshold a11-ppp-send-discard
Sets an alarm or alert for the PDSN service based on the number of packets that the PPP protocol processing
layer internally discarded on transmit for any reason.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Deletes the alert or alarm.
high_thresh
Default: 0
The high threshold number of discarded PPP send packets that must be met or exceeded within the polling
interval to generate an alert or alarm. It can be configured to any integer value between 0 and 100000.
clear low_thresh
Default:0
The low threshold number of discarded PPP send packets that must be met or exceeded within the polling
interval to clear an alert or alarm. It can be configured to any integer value between 0 and 100000.
Important This value is ignored for the Alert model. In addition, if this value is not configured for the Alarm model,
the system assumes it is identical to the high threshold.
Usage Guidelines Use this command to set an alert or an alarm when the number of discarded PPP send packets is equal to or
greater than a specified number.
Alerts or alarms are triggered for the number of discarded PPP send packets is based on the following rules:
• Enter condition: Actual number of discarded PPP send packets > High Threshold
• Clear condition: Actual number of discarded PPP send packets £ Low Threshold
Examples The following command configures a number of discarded PPP send packets threshold of 1000 and a low
threshold of 500 for a system using the Alarm thresholding model:
threshold a11-ppp-send-discard 1000 clear 500
threshold a11-rac-msg-discard
Sets an alarm or alert based on the number of Discarded A11 Registration Acknowledgements for the PDSN
service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Deletes the alert or alarm.
high_thresh
Default: 0
The high threshold number of Discarded A11 Registration Acknowledgements that must be met or exceeded
within the polling interval to generate an alert or alarm. It can be configured to any integer value between 0
and 100000.
clear low_thresh
Default:0
The low threshold number of Discarded A11 Registration Acknowledgements that must be met or exceeded
within the polling interval to clear an alert or alarm. It can be configured to any integer value between 0 and
100000.
Important This value is ignored for the Alert model. In addition, if this value is not configured for the Alarm model,
the system assumes it is identical to the high threshold.
Usage Guidelines Use this command to set an alert or an alarm when the number of Discarded A11 Registration
Acknowledgements is equal to or greater than a specified number.
Alerts or alarms are triggered for the number of Discarded A11 Registration Acknowledgements based on
the following rules:
• Enter condition: Actual number of Discarded A11 Registration Acknowledgements > High Threshold
• Clear condition: Actual number of Discarded A11 Registration Acknowledgements £ Low Threshold
Examples The following command configures a number of Discarded A11 Registration Acknowledgements threshold
of 1000 and a low threshold of 500 for a system using the Alarm thresholding model:
threshold a11-rac-msg-discard 1000 clear 500
threshold a11-rrp-failure
Sets an alarm or alert based on the number of A11 Registration Response failures for the PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Deletes the alert or alarm.
high_thresh
Default: 0
The high threshold number of A11 Registration Response failures that must be met or exceeded within the
polling interval to generate an alert or alarm. It can be configured to any integer value between 0 and 100000.
clear low_thresh
Default:0
The low threshold number of A11 Registration Response failures that must be met or exceeded within the
polling interval to clear an alert or alarm. It can be configured to any integer value between 0 and 100000.
Important This value is ignored for the Alert model. In addition, if this value is not configured for the Alarm model,
the system assumes it is identical to the high threshold.
Usage Guidelines Use this command to set an alert or an alarm when the number of A11 Registration Response failures is equal
to or greater than a specified number.
Alerts or alarms are triggered for the number of A11 Registration Response failures based on the following
rules:
• Enter condition: Actual number of A11 Registration Response failures > High Threshold
• Clear condition: Actual number of A11 Registration Response failures £ Low Threshold
Examples The following command configures a number of A11 Registration Response failures threshold of 1000 and
a low threshold of 500 for a system using the Alarm thresholding model:
threshold a11-rrp-failure 1000 clear 500
threshold a11-rrq-msg-discard
Sets an alarm or alert based on the number of Discarded A11 Registration Requests for the PDSN service.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Deletes the alert or alarm.
high_thresh
Default: 0
The high threshold number of Discarded A11 Registration Requests that must be met or exceeded within the
polling interval to generate an alert or alarm. It can be configured to any integer value between 0 and 100000.
clear low_thresh
Default:0
The low threshold number of Discarded A11 Registration Requests that must be met or exceeded within the
polling interval to clear an alert or alarm. It can be configured to any integer value between 0 and 100000.
Important This value is ignored for the Alert model. In addition, if this value is not configured for the Alarm model,
the system assumes it is identical to the high threshold.
Usage Guidelines Use this command to set an alert or an alarm when the number of Discarded A11 Registration Requests is
equal to or greater than a specified number.
Alerts or alarms are triggered for the number of Discarded A11 Registration Requests based on the following
rules:
• Enter condition: Actual number of Discarded A11 Registration Requests > High Threshold
• Clear condition: Actual number of Discarded A11 Registration Requests £ Low Threshold
Examples The following command configures a number of Discarded A11 Registration Requests threshold of 1000 and
a low threshold of 500 for a system using the Alarm thresholding model:
threshold a11-rrq-msg-discard 1000 clear 500
threshold init-rrq-rcvd-rate
Sets an alarm or alert based on the average number of calls setup per second for the context.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration
configure > context context_name > pdsn-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pdsn-service)#
no
Deletes the alert or alarm.
high_thresh
Default: 0
The high threshold average number of calls setup per second must be met or exceeded within the polling
interval to generate an alert or alarm. It can be configured to any integer value between 0 and 1000000.
clear low_thresh
Default:0
The low threshold average number of calls setup per second that must be met or exceeded within the polling
interval to clear an alert or alarm. It can be configured to any integer value between 0 and 1000000.
Important This value is ignored for the Alert model. In addition, if this value is not configured for the Alarm model,
the system assumes it is identical to the high threshold.
Usage Guidelines Use this command to set an alert or an alarm when the average number of calls setup per second is equal to
or greater than a specified number of calls per second.
Alerts or alarms are triggered for the number of calls setup per second based on the following rules:
• Enter condition: Actual number of calls setup per second > High Threshold
• Clear condition: Actual number of calls setup per second £ Low Threshold
Examples The following command configures a number of calls setup per second threshold of 1000 and a low threshold
of 500 for a system using the Alarm thresholding model:
threshold init-rrq-rcvd-rate 1000 clear 500
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration > PDSN Service
ROHC
configure > context context_name > pdsn-service service_name > ip header-compression rohc
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ip-header-compression-rohc)#
Important The commands, keywords and variables in this mode are available dependent on platform type, product
version, and installed license(s).
cid-mode
Enters the RoHC Profile Compression Options Configuration mode.and configures options that apply during
RoHC compression for the current RoHC profile.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration > PDSN Service ROHC
configure > context context_name > pdsn-service service_name > ip header-compression rohc
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ip-header-compression-rohc)#
default
Reset all options in the RoHC Profile Compression Configuration mode to their default values.
large
Use large packets with optional information for RoHC
small
This is the default packet size.
Use small RoHC packets.
max-cid integer
Default: 15
The highest context ID number to be used by the compressor. integer must be an integer from 0 through 15
when small packet size is selected and must be an integer from 0 through 31 when large packet size is selected.
Usage Guidelines Use this command to set the RoHC packet size and define the maximum
Examples The following command sets large RoHC packet size and sets the maximum CID to 100:
cid-mode large max-cid 100
The following command sets the cid-mode to the default settings of small packets and max-cid 0:
default cid-mode
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
mrru
Sets the size of the largest reconstructed reception unit, in octets, that the decompressor is expected to reassemble
from segments. The size includes the CRC. If MRRU is negotiated to be 0, no segment headers are allowed
on the channel.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration > PDSN Service ROHC
configure > context context_name > pdsn-service service_name > ip header-compression rohc
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ip-header-compression-rohc)#
default
reset the value of this command to its default setting
num_octets
Default: 0
This is the number of octets for the maximum size of the largest reconstructed reception unit allowed.
num_octets must be an integer from 0 through 65535.
Usage Guidelines Use this command to set the size, in octets, of the largest reconstructed reception unit, in octets, that the
decompressor is expected to reassemble from segments.
Examples The following command sets the largest reconstructed reception unit to 1024 octets:
mrru 1024
The following command resets the mrru size to its default of 0 octets:
default mrru
profile
Specifies the header compression profiles to use. A header compression profile is a specification of how to
compress the headers of a specific kind of packet stream over a specific kind of link. At least one profile must
be specified.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > PDSN Service Configuration > PDSN Service ROHC
configure > context context_name > pdsn-service service_name > ip header-compression rohc
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-ip-header-compression-rohc)#
default
Default: esp-ip rtp-udp udp-ip uncompressed-ip
This command sets the RoHC profile configuration back to its default setting.
esp-ip
This enables RoHC Profile 0x0003 which is for ESP/IP compression, compression of the header chain up to
and including the first ESP header, but not subsequent subheaders.
rtp-udp
This enables RoHCProfile 0x0001 which is for RTP/UDP/IP compression
udp-ip
This enables RoHC Profile 0x0002 which is for UDP/IP compression, compression of the first 12 octets of
the UDP payload is not attempted.
uncompressed-ip
This enables RoHC Profile 0x0000 which is for sending uncompressed IP packets.
Usage Guidelines Use this command to specify the RoHC header compression profiles to use.
Examples The following command sets the profiles to use as esp-ip and rtp-udp:
profile esp-ip rtp-udp
The Peer List Configuration Mode is used to add or remove IP address to an SecGW crypto peer list..
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
address
Adds or deletes an IPv4 or IPv6 address to a crypto peer list.
no
Removes the specified IP address from the crypto peer list.
peer_address
Specifies an IP address in either IPv4 dotted-decimal (#.#.#.#/##) or IPv6 colon-separated-hexadecimal
(####:####:####:####:####:####:####:####/###) notation with CIDR (required). The only notation allowed
will be the one specified when the crypto peer list was created.
Usage Guidelines Use this command to add or delete an IPv4 or IPv6 address to a crypto peer list.
Repeat this command to add up to 1,000 peer IP addresses to the crypto peer list. The IP addresses in the list
can only be entered in either IPv4 or IPv6 notation, depending on the address type specified when the list was
created.
The following restrictions apply:
• A maximum of 1,000 peer IP addresses can be added to the peer list via the Peer List Configuration
mode address command.
• WSG service address binding is not allowed if a peer list is configured and both address types do not
match. An error message is generated if they do not match.
• An IPv4 or IPv6 peer list cannot be modified if peer-list peer_list_name is enabled under the WSG
service.
Examples The following command adds IPv4 address 10.1.1.1 to the crypto peer list:
address 10.1.1.1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
arp-mapping
Configures UMTS ARP to Gx ARP mapping for the specific peer profile.
Product GGSN
P-GW
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
default
Sets default values for the peer profile
Usage Guidelines Use this command to configure UMTS ARP to Gx ARP mapping for GGSN peer profile configured through
this mode.
Examples The following command sets the high priority level 4 and low priority level 9 for UMTS to Gx ARP mapping
for a GGSN peer profile:
arp-mapping priority-level high 4 medium 9
description
Sets a relevant descriptive string for the specific peer profile. By default it is blank.
Product GGSN
P-GW
SAEGW
S-GW
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
no
Removes the set description for GGSN, P-GW, or S-GW service peer profile configured through this mode.
desc_string
Indicates the description for GGSN, P-GW, or S-GW service peer profile configured through this mode; must
be an alphanumeric string from 1 through 64 characters.
Usage Guidelines Use this command to set a relevant description for GGSN, P-GW, or S-GW peer profile configured through
this mode.
Examples The following command sets the description ggsn_gtpc_SGSN_profile1 for a GGSN peer profile:
description ggsn_gtpc_SGSN_profile1
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
gtpc
Configure the GTP-C parameters for this peer profile.
Product GGSN
P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
default
Resets the specified parameter to its default value.
no
Disables or removes the configured GTP-C echo settings.
Configures the echo retransmission timeout, in seconds, for the this peer profile.
echo_retrans_dur must be an integer ranging from 1 to 20.
max-retransmissions retrans_num
retransmission-timeout retrans_dur
Note In 17.3 and later releases, this option has been deprecated. Use retransmission-timeout-ms .
retransmission-timeout-ms retrans_dur
Usage Guidelines Use this command to configure GTP-C parameters for GGSN, P-GW, or S-GW peer profile.
Examples The following command sets the GTP-C echo parameters to default values:
default gtpc echo
The following command sets the GTP-C retransmission timeout parameters to 4 seconds:
default gtpc retransmission-timeout-ms
lawful-intercept
Refer to the Cisco ASR 5x00 Lawful Intercept Configuration Guide for a description of this command.
no-qos-negotiation
Configures overriding of No-Qos-Negotiation flag in common flag IE received from peer node.
Product GGSN
P-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
no
Disables or removes the configured overriding of No-Qos-Negotiation flag in common flag IE received from
peer node.
set-flag
Sets flag value to 1 in common flag IE.
unset-flag
Sets flag value to 0 in common flag IE.
Usage Guidelines Use this command to configure the overriding of no-qos-negotiation flag value in Common Flags IE received
from the peer.
Examples The following command sets the flag value to true, i.e. 1, in Common Flags IE:
no-qos-negotiation set-flag
upgrade-qos-supported
Configures overriding of upgrade-Qos-supported flag in common flag IE received from peer node.
Product GGSN
P-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Peer Profile Configuration
configure > peer-profile service-type <service-type> {default | name peer_profile_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-peer-profile-ggsn/pgw/sgw-access/nw)#
no
Disables or removes the configured overriding of upgrade-Qos-supported flag in common flag IE received
from peer node.
set-flag
Sets flag value to 1 in common flag IE.
unset-flag
Sets flag value to 0 in common flag IE.
Usage Guidelines Use this command to configure the overriding of upgrade-Qos-supported flag value in Common Flags IE
received from the peer.
Examples The following command sets the flag value to false, i.e. 0, in Common Flags IE:
upgrade-Qos-supported unset-flag
The Peer-Server configuration mode provides the commands to define and manage the peer server
configuration part of the SS7 routing on an SGSN.
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the configuration mode and returns to the Exec mode.
Product All
exit
Exits the configuration mode and returns to the Global configuration mode.
Product All
mode
Configures the operational mode of the peer-server.
Important In Release 20.0, HNBGW is not supported. This command must not be used for HNBGW in Release 20.0.
For more information, contact your Cisco account representative.
Product SGSN
HNB-GW
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
loadshare
Sets the peer-server to load share. This is the default.
standby
Sets the peer-server to be in standby mode.
name
Defines the unique identification - the name - of the peer-server in the SS7 routing domain.
Important In Release 20.0, HNBGW is not supported. This command must not be used for HNBGW in Release 20.0.
For more information, contact your Cisco account representative.
Product SGSN
HNB-GW
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
no
Removes the peer server's name from this configuration instance.
name
name: Must be a string of 1 to 64 alphanumeric characters to define a unique identification for the peer-server
within the specific SS7 routing domain. Double quotes must be used to create a name that includes spaces.
Usage Guidelines Create peer server names that are easy to remember and uniquely identify the PSP.
Examples Use this command to create an easily remembered alphanumeric name for the peer-server:
name "Berlin West"
psp
Creates the peer-server-process (PSP) instance and enters the PSP configuration mode. See the PSP
Configuration Mode chapter in this guide for information on the configuration commands.
Important In Release 20.0, HNBGW is not supported. This command must not be used for HNBGW in Release 20.0.
For more information, contact your Cisco account representative.
Important This command configures a mandatory parameter in the configuration of the peer server.
Product SGSN
HNB-GW
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
no
Removes the PSP instance from the peer server configuration.
id
id Uniquely identifies the specific peer-server-process configuration. The Id must be an integer from 1 to 4
or for SGSN with Release 15.0 or higher, the value range is 1 to12.
Usage Guidelines Use this command to define the peer-server-process (PSP) instance ID number for the SGSN configuration.
Examples Use this command to create instance #3 for the PSP configuration:
psp instance 3
routing-context
Defines the ID of the routing context for the peer-server to use.
Important In Release 20.0, HNBGW is not supported. This command must not be used for HNBGW in Release 20.0.
For more information, contact your Cisco account representative.
Important This command configures a mandatory parameter in the configuration of the peer server.
Product SGSN
HNB-GW
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
id
id Uniquely identifies a specific routing context for the peer-server-process to use. The Id must be an integer
from 1 to 65535.
From release 17.0 onwards, the SGSN supports an integer of 0 to 4294967295 as a valid value for the
routing-context ID in M3UA messages.
no
Removes the routing-context definition from the peer server configuration.
Usage Guidelines Use this command to define routing contexts for the peer server.
self-point-code
This command defines the point-code to identify the SGSN as a peer server.
Important In Release 20.0, HNBGW is not supported. This command must not be used for HNBGW in Release 20.0.
For more information, contact your Cisco account representative.
Product SGSN
HNB-GW
Command Modes Exec > Global Configuration > SS7 Routing Domain Configuration > Peer-Server Configuration
configure > ss7-routing-domain rd_id variant variant_type > peer-server id server_id
Entering the above command sequence results in the following prompt:
[local]host_name(config-ss7-rd-ss7rd_id-ps-id-peer-server_id)#
point-code
Point-code is an SS7-type address for an element in the SS7 network. Point-codes must be defined in
dotted-decimal format in a string of 1 to 11 digits. Options include:
• 0.0.1 to 7.255.7 for point-code in the ITU range.
• 0.0.1 to 255.255.255 for point-code in the ANSI range.
• 0.0.1 to 15.31.255 for point-code in the TTC Range.
• a string of 1 to 11 digits in dotted-decimal to represent a point-code in a different range.
no
Removes the self-point-code configuration for this linkset in the peer server.
Important Removing the self-point-code will result in the termination of all traffic on this link.
Usage Guidelines Use this command to define the point-code to identify the SGSN.
Examples Use the following command to remove the self-point-code definition from the peer-server configuration:
no self-point-code
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
associate
Associates the P-GW service with specific pre-configured services and/or policies configured in the same
context.
Product P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
no
Removes the selected association from this service.
gtpc-load-control-profile name
Specifies that a GTPC Load Control Profile is to be associated with an existing P-GW service in this context.
name must be an alphanumeric string from 1 to 64 characters in length.
gtpc-overload-control-profile name
Specifies that a GTPC Overload Control Profile is to be associated with an existing P-GW service in this
context.
name must be an alphanumeric string from 1 to 64 characters in length.
ggsn-service name
Specifies that the P-GW service is to be associated with an existing GGSN service within this context.
name must be an alphanumeric string of 1 through 63 characters and be an existing GGSN service.
peer-map map_name
Specifies that the P-GW service is to be associated with an existing peer map within this context.
map_name must be an alphanumeric string of 1 through 63 characters and be an existing peer map.
Refer to the LTE Policy Configuration Mode Commands chapter for more information on peer map creation.
qci-qos-mapping name
Specifies that the P-GW service is to be associated with an existing QCI-QoS mapping configuration within
this context.
name must be an alphanumeric string of 1 through 63 characters and be an existing QCI-QoS mapping
configuration.
QCI-Qos mapping is typically configured in a AAA context. Refer to the QCI-QoS Mapping Configuration
Mode Commands chapter for more information.
Important If a GGSN service is associated with a P-GW service, then the GGSN service will use the QCI-QoS
mapping tables specified in the qci-qos-mapping command and assigned to its associated P-GW service.
Usage Guidelines Use this command to associate the P-GW service with other pre-configured services and/or policies configured
in the same context.
Examples The following command associates this service with an eGTP service called egtp1:
associate egtp-service egtp1
authorize-with-hss
This command enables or disables subscriber session authorization via a Home Subscriber Server (HSS) over
an S6b Diameter interface. This feature is required to support the interworking of GGSN with P-GW and HA.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
Syntax Description authorize-with-hss [ egtp [ s2b ] [ s5-s8 ] [ report-ipv6-addr ] | lma [ s6b-aaa-group aaa-group-name |
report-ipv6-addr ] | report-ipv6-addr ]
{ default | no } authorize-with-hss
default
Disables the default authorization of subscriber over S6b interface. Resets the command to the default setting
of "authorize locally" from an internal APN authorization configuration.
no
Disables the default authorization of subscriber over S6b interface. Resets the command to the default setting
of "authorize locally" from an internal APN authorization configuration.
egtp
Enables S6b authorization for eGTP only.
s2b
Enables S6b authorization for eGTP S2b.
s5-s8
Enables S6b authorization for eGTP S5S8.
lma
Enables S6b authorization for LMA only.
report-ipv6-addr
Enables the IPv6 address reporting through Authorization-Authentication-Request (AAR) towards the S6b
interface.
Usage Guidelines Use this command to enable/disable the authorization support for subscriber over S6b interface, which is used
between P-GW and the 3GPP AAA to exchange the information related to charging, GGSN discovery, etc.
dns-client
Specifies the DNS client context to use for sending DNS queries.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Returns the command to the default setting of targeting the DNS client in the context where the P-GW service
resides.
no
Disables DNS queries.
context name
Specifies the name of the context where the DNS client is used for the resolution of PCSCF-FQDN received
from S6b interface.
name must be an existing context expressed as an alphanumeric string of 1 through 79 characters.
Usage Guidelines Use this command to specify the context where the DNS client resides to perform P-CSCF-FQDN resolution
from the S6b interface.
Examples The following command identifies the egress1 context as the context where the DNS client resides:
dns-client context egress1
egtp
Configures handling of eGTP related procedures.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
Syntax Description egtp { bitrates-rounded-down-kbps | cause-code temp-fail timeout sec retry retries |
gngp-modify-bearer-rsp-with-apn-ambr | modify-bearer-cmd-negotiate-qos | overcharge-protection [
drop-all | transmit-all ] | sgw-restoration session-hold timeout seconds| suppress-ubr no-bitrate-change
}
default egtp { cause-code temp-fail | modify-bearer-cmd-negotiate-qos |
gngp-modify-bearer-rsp-with-apn-ambr | overcharge-protection | sgw-restoration session-hold }
no egtp { bitrates-rounded-down-kbps | cause-code temp-fail | gngp-modify-bearer-rsp-with-apn-ambr
| modify-bearer-cmd-negotiate-qos | overcharge-protection | sgw-restoration session-hold| suppress-ubr
no-bitrate-change }
default
Resets the command to the default setting.
no
Disables the configuration statement.
bitrates-rounded-down-kbps
Bit rate granularity provided by different interfaces was not originally aligned in 3GPP specifications. For
example, the PCRF provided bits per second on the Gx and the GTP utilized kilobits per second. Due to the
conversion of bps to kbps, there were scenarios where the rounding off could have resulted in the incorrect
allocation of MBR/GBR values.
When this keyword is disabled, a bitrate value sent on GTP interface will be rounded up if the conversion
from bps (received from Gx) to kbps results in a fractional value. However, the enforcement of bitrate value
(AMBR, MBR, GBR) values will remain the same. Once the value (in kbps) that is sent towards the Access
side, it needs to be rounded up. Also, show subscribers pgw-only full all will show the APN-AMBR in terms
of bps.
When enabled, the previous behavior of rounded-down kpbs bitrate (AMBR, MBR, BGR) values being sent
towards the Access side is enforced. In addition, show subscribers pgw-only full all displays in terms of
kpbs.
By default, this command is configured to use rounded-up bitrate values.
gngp-modify-bearer-rsp-with-apn-ambr
Sends Modify Bearer Response with APN-AMBR only for GnGp Handoff. By default, this option is disabled.
modify-bearer-cmd-negotiate-qos
This configuration only impacts the PGW QoS negotiation behavior when PCRF is unreachable or event
trigger is not registered while handling Modify Bearer Command. By default, this configuration is disabled.
When enabled, P-GW will always enforce old QoS values authorized by the PCRF. When disabled, the P-GW
will always accept new QoS values (APN-AMBR/Def-EPS-Bearer-QoS) received in Modify Bearer Command.
On S-GW failure indication, P-GW shall check if S-GW Restoration feature is enabled or not. If enabled,
P-GW shall maintain all the affected sessions for session-hold timeout. After session-hold timeout, P-GW
shall clear all the sessions which are not recovered yet.
suppress-ubr no-bitrate-change
Enables the P-GW to suppress the Update Bearer Request (UBR) message UBR if the bit rate is the same
after the round-off.
As the bit rate is expressed in bps on Gx and kbps on GTP, the P-GW does a round-off to convert a Gx request
into a GTP request. When the P-GW receives a RAR from the PCRF with minimal bit rate changes (in bps),
a UBR is sent, even if the same QoS (in kbps) is already set for the bearer. The UBR suppression feature
enables the P-GW to suppress such a UBR where there is no update for any of the bearer parameters.
When the UBR has multiple bearer contexts, the bearer context for which the bit rate change is less than 1
kbps after round-off is suppressed. If other parameters, such as QCI, ARP, and TFT, that might trigger an
UBR are changed and there is no change in bit rates after round-off, then UBR is not suppressed. Suppression
of UBR is applicable for UBR triggered by CCA-I, RAR, and Modify Bearer Command.
Default: disabled. This means that the UBReq should be triggered even if the Gx and GTP bit-rates in kbps
are same after round-off.
If the no option is used, it will disable this feature. That is, the UBReq should be triggered even if the Gx and
GTP bit-rates in kbps are same after round-off.
There is no separate default keyword for this feature. Use the no option to revert to the default behavior.
Important The UBR Suppression Feature is a licensed-controlled feature. Contact your Cisco account or service
representative for detailed licensing requirements.
Usage Guidelines Use this command to configure the behavior of the P-GW/SAEGW for eGTP procedures.
Examples The following command sets the temporary failure timer to 30 seconds and 2 retries:
egtp cause-code temp-fail timeout 30 retry 2
The following command configures the P-GW to accept new QoS values from the modify bearer command
while the PCRF is not reachable:
egtp modify-bearer-cmd-negotiate-qos
The following command enables S-GW restoration functionality and configures session hold timeout on a
P-GW service:
sgw-restoration session-hold timeout seconds
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
fqdn
Configures a Fully Qualified Domain Name for this P-GW service used in messages between the P-GW and
a 3GPP AAA server over the S6b interface.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Returns the command to the default setting of "null".
no
Removes the configured FQDN from this service configuration.
host domain_name
Specifies the domain name of the P-GW service.
domain_name must be an alphanumeric string of 1 through 255 characters.
realm realm_name
Specifies the realm name of the P-GW service.
realm_name must be an alphanumeric string of 1 through 255 characters.
Usage Guidelines Use this command to identify the P-GW service using an FQDN required when sending messages over the
S6b interface to a 3GPP AAA server.
Important In order to properly interact with other nodes in the network, the FQDN should be less than or equal to
96 alphanumeric characters.
Examples The following command configures the FQDN for this P-GW service as 123abc.all.com with a realm name
of all.com:
fqdn host 123abc.all.com realm all.com
The following command configures this P-GW service with an FQDN that enables topology matching:
fqdn host topon.interface_name.pgw01.bos.ma.node.epc .mnc<value>.mcc<value>.3gppnetwork.org
realm node.epc.mnc.mcc.3gppnetwork.org
Important The associated HSGW service must have a corresponding FQDN similar to the following:
topon.interface_name.hsgw01.bos.ma.node.epc.mncvalue.mccvalue.3gppnetwork.org
Product P-GW
Command Modes Exec > Global Configuration > Context Configuration >P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
no
Disables collision handling between SGSN initiated UPC and NRUPC request.
default
Sets default collision handling behavior between SGSN initiated UPC and NRUPC request. By default,
collision handling is enabled.
Usage Guidelines This command is used to enable or disable collision handling between SGSN initiated UPC and NRUPC
request.
Examples The following example disables collision handling between SGSN initiated UPC and NRUPC request.
no gtpc handle-collision upc nrupc
gx-li
Refer to the Lawful Intercept Configuration Guide for a description of this command.
map-initial-setup-auth-fail-to-gtp-cause-user-auth-fail
Maps Gx cause code (5xxx) to access side GTP cause code Auth-failure(92) in Create Session Response
message.
Product P-GW
SAEGW
Privilege P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Maps Gx cause code (5xxx) to access side GTP cause code No-Resource(73) in Create Session Response
message.
no
Maps Gx cause code (5xxx) to access side GTP cause code No-Resource(73) in Create Session Response
message.
Usage Guidelines When Create Session Request message arrives at P-GW, CCR-I is sent to PCRF and PCRF rejects calls with
5xxx cause code in CCA-I. In this case, Create Session Response is sent with failure indicated by GTP cause
code. Use this command to control which GTP cause code is sent, "No Resources Available" or "User
Authentication Failed", in Create Session Response message for this scenario. By default, "No Resources
Available" is sent for this case; however, enabling this command sends "User Authentication Failed" cause
code in Create Session Response.
Examples The following command maps Gx cause code (5xxx) to access side GTP cause code Auth-failure(92) in Create
Session Response message:
map-initial-setup-auth-fail-to-gtp-cause-user-auth-fail
message-timestamp-drift
Allows drift time configuration to take care of NTP drift issues.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Sets drift time to 180 seconds.
If P-GW detects drift less than 180 seconds, it will check for condition "MWT + ReceivedTimeStamp (time
from MME) > CurrentTimeStampAtPGW", and based on that P-GW will reject the call. If this condition is
not met, it will transparently forward MWT and timestamp to AAA/Gx/Gy interfaces.
no
Disables message timestamp drift. MWT and received timestamp will not be passed on to all AAA/Gx/Gy
interfaces.
message-timestamp-drift time_in_seconds
Configures the drift time from the message timestamp, in seconds, up to which P-GW will consider processing
the message timestamp and max-wait-time (MWT) IEs.
If the create-time from MME is off from the currenttime by configured-drift-duration, then this could lead to
a high NTP drift and session uniqueness falls back to using currenttime toward Diameter servers.
If the timestamp received in CSReq is significantly off (more than configured drift), then P-GW will not take
action based on MWT and received timestamp and will transparently pass it to all AAA/Gx/Gy interfaces.
When received drift is less than configured limit, P-GW will reject the call if "MWT + ReceivedTimeStamp
> CurrentTimeStampAtPGW" condition is met. Otherwise, P-GW will forward the timestamp and MWT to
AAA/Gx/Gy interfaces.
time_in_seconds must be an integer from 0 to 1000.
Default: 180
Usage Guidelines When the MME is reselected by the UE or when the MME reselects a different P-GW during timeout scenarios,
it is possible that the old PDN connection request is still being processed in the network and the session created
by the new PDN connection request is overwritten by the stale procedure.
IEs TimeStamp and MWT (MaxWaitTime) have been added in CSReq and forwarded on S6b/Gx/Gy interfaces
in order to maintain session uniqueness at P-GW.
Important Drift time configuration under P-GW service shall be used by the associated LMA service.
Example Scenario
In the following scenario, stale session won't be present on P-GW.
The P-GW is still processing the session creation but the S-GW times out due to timer configurations and
notifies the MME with Create Session Failure (Cause #100: Remote Peer Not Responding). MME reselects
an alternate P-GW in this case, but the original P-GW still continues to process the session. In certain scenarios,
the original P-GW can overwrite the Gx session on the PCRF that is created by the newly selected P-GW. In
this case, the new P-GW session is the valid session and original P-GW session is invalid as far as the UE,
MME, and S-GW are concerned. The same can occur with the AAA session as well based on timing. This
results in PCRF having invalid session information and the user plane works fine anchored on the second
P-GW, but the Rx and Gx signaling fails as this terminates via original P-GW.
This results in VoLTE calls failing after SIP signaling between UE and P-CSCF.
To solve the problem, TimeStamp and MWT IE have been incorporated to be transmitted from MME and
shared across the network nodes.
newcall
Configures the P-GW to accept or reject requests for a static IP address if the address is already in use by
another session.
Product P-GW
SAEGW
SaMOG
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
no
Returns the command to the default setting of "reject".
duplicate-subscriber-requested-address
Configures how duplicate sessions with same IPv4 address request are handled.
duplicate-subscriber-requested-v6-address
Configures how duplicate sessions with same IPv6 address request are handled.
accept | reject
Default: reject
accept: Specifies that the old session with the requested address will be ended to accept the new session with
the same address.
reject: Specifies that the new session requesting the same address will be rejected.
Usage Guidelines Use this command to configure the behavior of the P-GW service when receiving requests for static IP or
IPv6 address already in use by other sessions.
Important This command is only applicable to sessions using services supporting duplicate address abort. These
services include HA, GGSN, and P-GW.
Examples The following command allows for the acceptance of requests for static IP addresses already in use by other
sessions:
newcall duplicate-subscriber-requested-address accept
pcscf-restoration
Configures the mechanism to support P-CSCF restoration when a failure is detected.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
hss-solution
Enables the Release 12-based HSS solution for P-CSCF restoration.
custom-hss-solution
Enables private extension-based HSS solution for P-CSCF restoration.
This is the default setting.
default
Returns pcscf-restoration to the default setting (custom-hss-solution).
Usage Guidelines Use this command to enable/disable the standards-based mechanism for P-CSCF failure detection. This
command enables operators to ensure a failed P-CSCF address is not provided to the IMS client. Prior to
StarOS release18.2, P-CSCF restoration was supported by using the Private Extn IE. In StarOS releases 18.2
and later, the failure detection mechanism can be configured as standards-based. By default this feature is
disabled; in which case the Private Extn mechanism will be used for P-CSCF restoration.
plmn id
Configures Public Land Mobile Network (PLMN) identifiers used to determine if a mobile station is visiting,
roaming, or belongs to a network. Up to 512 PLMN IDs can be configured for each P-GW service.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
no
Removes a previously configured PLMN identifier for the P-GW service.
mcc mcc_value
Specifies the mobile country code (MCC) portion of the PLMN identifier.
mcc_value is the PLMN MCC identifier and must be an integer from 100 through 999.
mnc mnc_value
Specifies the mobile network code (MNC) portion of the PLMN identifier.
mnc_value is the PLMN MNC identifier and can be configured to a 2- or 3-digit integer from 00 through 999.
primary
When multiple PLMN IDs are configured, the primary keyword can be used to designate one of the PLMN
IDs to be used for the AAA attribute.
Usage Guidelines The PLMN identifier is used to aid the P-GW service in the determination of whether or not a mobile station
is visiting, roaming, or home. Multiple P-GW services can be configured with the same PLMN identifier. Up
to 512 PLMN IDs can be configured for each P-GW Service.
Important The number of supported PLMN IDs was increased from 5 to 512 in StarOS Release 17.1. In addition,
the MNC portion of the PLMN ID became optional.
If the MNC portion of a PLMN ID is not specified, home PLMN qualification will be done based solely on
the MCC value and the MNC portion will be ignored for these particular MCCs.
Examples The following command configures the PLMN identifier with an MCC of 462 and MNC of 02:
plmn id mcc 462 mnc 02
session-delete-delay
Configures a delay in terminating a session.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Resets the command to the default setting of 10000 milliseconds.
no
Disables the feature.
timeout msec
Default: 10000
Specifies the time to retain the session (in milliseconds) before terminating it.
msec must be an integer from 1000 to 60000.
Usage Guidelines Use this command to set a delay to provide session continuity in break-before-make scenarios.
Examples The following command sets the session delete delay to the default setting of 10,000 milliseconds:
session-delete-delay timeout
setup-timeout
Configures the maximum amount of time the P-GW service takes for creating a session.
Product P-GW
Command Modes Exec > Global Configuration > Context Configuration > P-GW Service Configuration
configure > context context_name > pgw-service service_name
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-pgw-service)#
default
Configures the default guard timer value for session creation.
setup-time
Default: 60
Specifies the maximum amount of time taken by P-GW for service creation.
setup_time is measured in seconds and can be configured to an integer from 1 through 120.
Usage Guidelines Use this command to limit the amount of time allowed for creating a session. If a "Create Session Request"
is received and the setup-timeout is configured, the timer starts with the configured value. If the setup timeout
is not configured, the timer starts with the default value of 60 seconds.
Examples The following command allows a maximum of 120 seconds for creating a session:
setup-timeout 120
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
apn-name-to-be-included
This command configures the APN name to be included in CCR Gx messages.
Product GGSN
IPSG
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
gn | virtual
Specifies which APN name must be sent in the Gx messages.
gn: Specifies to send the real APN name.
virtual: Specifies to send the virtual APN name if present, else to send the real APN name.
default
Applies the default setting for this command.
Default: gn
Usage Guidelines This feature is developed to implement a single global APN for the Enterprise services with the ability to have
separate virtual APNs per single Enterprise, group of Enterprises sharing the same service group or per
department.
To implement this feature, a configurable option is introduced per interface Rf, Gx, Gy and per APN. That
is, a service specific CLI "apn-name-to-be-included" is configured for interfaces Rf, Gx, Gy separately. It
can take values 'gn' or 'virtual'. Based on the value configured for this command, the Called-Station-Id AVP
is populated.
This command is used to configure the APN name to be included in the CCR Gx messages to the PCRF —
the real APN name or the virtual APN name.
The name of the virtual APN and the IP pool are signaled during the UE attach to the Enterprise PDN from
the 3GPP AAA server over S6b interface with a new vendor-specific AVP "Virtual-APN-Name". The RADIUS
Start, Gy CCR to OFCS and Rf ACR to OCS messages contain the Virtual APN name instead of the Enterprise
APN.
This feature provides customers the desired granularity per enterprise and per department. This also allows
bundling of number of small enterprises under the umbrella of single APN and logically separating them by
virtual APN.
Examples The following command configures sending the real APN name in Gx messages:
apn-name-to-be-included gn
arp-priority-level
This command enables mapping of the ARP priority-level value received from PCRF to inter-user-priority
value and be sent in A11 session update.
Product PDSN
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
Configures the default setting for this command.
Default: arp-priority-level to inter-user-priority mapping not applicable
no
Disables arp-priority-level to inter-user-priority mapping.
Usage Guidelines
Important This command is for a customer-specific implementation to support IP-CAN policy control via Gx interface
in PDSN, wherein the PCRF informs the subscriber's subscription level (such as gold, silver, bronze) to
PDSN/PCEF via Priority-Level AVP, then PDSN maps the subscriber's subscription level to
inter-user-priority and transmits it to PCF via A11 session update message. For more information on the
use of this command contact your Cisco account representative.
associate
This command associates/disassociates a failure handling templateor a local policy template with the IMS
authorization service.
Product GGSN
HA
HSGW
IPSG
PDSN
P-GW
SAEGW
S-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
no
Disassociates a failure handling templateor local policy template with the IMS authorization service.
failure-handling-template template_name
Associates a previously created failure handling template with the IMS authorization service. template_name
specifies the name for a pre-configured failure handling template. template_name must be an alphanumeric
string of 1 through 63 characters.
For more information on failure handling templates, refer to the failure-handling-template command in the
Global Configuration Mode Commands chapter.
Usage Guidelines Use this command to associate a configured failure handling template or local policy service with the IMS
authorization service.
The failure handling template defines the action to be taken when the Diameter application encounters a failure
supposing a result-code failure, tx-expiry or response-timeout. The application will take the action given by
the template. For more information on failure handling template, refer to the Failure Handling Template
Configuration Mode Commands chapter.
Important Only one failure handling template can be associated with the IMS authorization service. The failure
handling template should be configured prior to issuing this command.
If the association is not made to the template then failure handling behavior configured in the application with
the failure-handling command will take effect.
To support fallback to local policy in case of failure at PCRF for CCFH continue, the local policy service
should be associated with an IMS authorization service. In case of any failures, the local policy template
associated with the ims-auth service will be chosen for fallback.
Examples The following command associates a pre-configured failure handling template called fht1 to the IMS
authorization service:
associate failure-handling-template fht1
cc-profile
This command configures the value of the Offline AVP sent to the PCRF based on the Charging Characteristics
(CC) profile received from the SGSN.
Product GGSN
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
Configures the default setting for this command.
Default: Deletes all previously configured mappings.
no
Deletes all previously configured mappings.
cc_profile_number
Specifies the CC profile number to map.
For example, 1 for Hot Billing.
cc_profile_number must be an integer from 0 through 15.
cc_profile_number_range_end
Specifies, for a range of CC profile numbers to map, the end number. That is, from cc_profile_number through
cc_profile_number_range_end.
cc_profile_number_range_end must be an integer from 1 through 15.
map-to offline-avp { 0 | 1 }
Specifies to map the CC profile number(s) to the Offline AVP value sent to the PCRF.
• 0: Corresponds to the value DISABLE_OFFLINE (0).
• 1: Corresponds to the value ENABLE_OFFLINE (1).
Usage Guidelines Use this command to configure the CC Profile to Offline AVP value mapping. The Offline AVP's value
(DISABLE_OFFLINE (0), ENABLE_OFFLINE (1)) is derived based on the CC profile received from the
SGSN as specified by this mapping.
The following example shows how this command can be configured multiple times:
cc-profile 1 to 2 map-to offline-avp 1
cc-profile 4 map-to offline-avp 0
cc-profile 8 map-to offline-avp 1
On configuring the above set of commands, the Offline AVP value is sent as 1 (Offline enabled) for the CC
profiles 1 (Hot Billing), 2 (Flat Rate), and 8 (Post-Paid). And, as 0 (Offline disabled) for the CC profile 4
(Pre-paid).
When configuring this command, overlapping of CC profile numbers is not permitted. In the following
example, after configuring the first command, which specifies to send the Offline AVP's value as 1 (Offline
enabled) for the CC profiles 1 through 15, the second command, which specifies to map CC profile 7, is not
permitted:
cc-profile 1 to 15 map-to offline-avp 1
cc-profile 7 map-to offline-avp 0
Examples The following command specifies to send Offline AVP value as 1 (Offline enabled) for the CC profile 1 (Hot
Billing):
cc-profile 1 map-to offline-avp 1
The following command specifies to delete all previously configured mappings:
no cc-profile
custom-reauth-trigger
This command enables custom reauthorization event triggers.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
Configures the default setting for this command. The default setting is to enable all the event triggers.
none
Disables all custom event triggers.
apn-ambr-mod-failure
Enables APN AMBR Modification Failure event trigger.
default-bearer-qos-change
Enables Default EPS bearer QoS change event trigger.
default-bearer-qos-mod-failure
Enables Default EPS Bearer QOS Modification Failure event trigger.
qos-change
Enables QoS change trigger.
resource-modification-request
Enables Resource modification trigger.
ue-ip-addr-allocate
Enables UE IP address allocate trigger.
ue-ip-addr-release
Enables UE IP address release trigger.
preservation-changed
Enables preservation-changed event trigger.
Important This keyword is for use with a customer-specific implementation, and will be available only if a valid
license is installed.
reactivation-changed
Enables reactivation-changed event trigger.
Important This keyword is for use with a customer-specific implementation, and will be available only if a valid
license is installed.
Usage Guidelines Use this command to enable/disable custom reauth event triggers.
It is recommended that the preservation-changed and reactivation-changed triggers both be enabled. As, when
the bearer goes into preservation mode with the preservation-changed trigger, the reactivation-changed trigger
must also be enabled for the bearer to get reactivated subsequently.
In 16.0 and later releases, this CLI command overwrites the previously configured triggers with the new event
triggers. For example, if the following triggers are configured – QoS change, UE IP address allocation, UE
IP address release, preservation-changed, reactivation-changed, then the APN-AMBR modification failure
and Resource modification request triggers should be configured. This operation will overwrite all previously
configured triggers and will configure only new APN-AMBR modification failure and Resource modification
request triggers. By default, these event triggers are enabled.
diameter 3gpp-r9-flow-direction
This command controls PCEF from sending Flow-Description, TFT-Filter, andPacket-Filter-Content AVPs
in Rel. 9 format.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
3gpp-r9-flow-direction
Encodes Flow-Direction, Flow-Description, TFT-Filter, and Packet-Filter-Content AVPs based on 3GPP Rel.
9 specification.
no
Encodes Flow-Direction, Flow-Description, TFT-Filter, and Packet-Filter-Content AVPs in Rel. 8 format.
This is the default configuration.
Usage Guidelines Use this command to enable Rel.9 changes for Flow-Description, TFT-Filter, and Packet-Filter-Content AVPs
sent by PCEF in CCR-U. This CLI command works in conjunction with diameter update-dictionary-avps
{ 3gpp-r9 | 3gpp-r10 }. When diameter 3gpp-r9-flow-direction is configured and negotiated supported
feature is 3gpp-r9 or above, PCEF will send Flow-Description, TFT-Filter, and Packet-Filter-Content AVPs
in 3GPP Rel. 9 format.
Per the 3GPP Rel. 8 standards, the IPFilterRule in Flow-Description, TFT-Filter, and Packet-Filter-Content
AVPs is sent as "permit in" for UPLINK and "permit out" for DOWNLINK direction. From 3GPP Rel. 9
onwards, the Flow-Description AVP within the Flow-Information AVP will have only "permit out" and the
traffic flow direction is indicated through Flow-Direction AVP. In 3GPP Rel. 9 format, both UPLINK and
DOWNLINK are always sent as "permit out" and hence the usage of "permit in" is deprecated.
Backward compatibility is maintained, i.e. both Rel. 8 (permit in/out) and Rel. 9 (permit out with flow-direction)
formats are accepted by PCEF.
This CLI command must be used only after the PCRF is upgraded to Rel. 9. For more information on this
feature, see the 3GPP Rel.9 Compliance for IPFilterRule section in the Gx Interface Support chapter in the
administration guide for the product you are deploying.
Examples The following command enables Rel.9 changes for Flow-Description, TFT-Filter, and Packet-Filter-Content
AVPs:
diameter 3gpp-r9-flow-direction
diameter dictionary
This command specifies the Diameter Policy Control Application dictionary to be used by the IMS
Authorization Service for the policy control application.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
dpca-custom1
Custom-defined Diameter dictionary for the Gx interface.
dpca-custom2
Custom-defined Diameter dictionary for Rel. 7 Gx interface.
dpca-custom3
Custom-defined Diameter dictionary for the Gx interface in conjunction with IP Services Gateway (IPSG).
dpca-custom4
Standard Diameter dictionary for 3GPP Rel. 7 Gx interface.
dpca-custom5
Custom-defined Diameter dictionary for Rel. 7 Gx interface.
dynamic-load
Configures the dynamically loaded Diameter dictionary. The dictionary name must be an alphanumeric string
of 1 through 15 characters.
For more information on dynamic loading of Diameter dictionaries, see the diameter dynamic-dictionary
in the Global Configuration Mode Commands chapter of this guide.
gx-wimax-standard
Gx WiMAX standard dictionary.
gxa-3gpp2-standard
Gxa 3GPP2 standard dictionary.
gxc-standard
Gxc standard dictionary.
pdsn-ty
This keyword is restricted.
r8-gx-standard
R8 Gx standard dictionary.
Standard
Standard Diameter dictionary for the 3GPP Rel. 6 Gx interface.
Default: Enabled for Gx support in 3GPP networks.
std-pdsn-ty
This keyword is restricted.
ty-plus
This keyword is restricted.
ty-standard
This keyword is restricted.
default
Sets the default Diameter dictionary.
Default: Standard
Usage Guidelines Use this command to specify the Diameter dictionary for IMS Authorization Service.
Examples The following command sets the Standard dictionary for Diameter Policy Control functions in 3GPP network:
diameter dictionary Standard
diameter encode-event-avps
This command enables encoding of all the event-related information AVPs in CCR-U messages.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
Applies the default setting for this command.
Default: Sends AVPs relevant to the Event-Trigger subscribed by the PCRF.
always
This keyword option always sends the event-related AVPs in all CCR messages.
local-fallback
This keyword option sends the event-related AVPs in CCR-U messages in the event of local fallback scenario.
Usage Guidelines Use this command to facilitate sending of all the event-related information AVPs in CCR-U messages.
In releases prior to 14.0, per the 3GPP standards for Gx, AVPs relevant to the Event-Trigger subscribed by
the PCRF were always sent in the CCR messages. This release onwards, sending of event-related AVPs for
all update (both access side and internal) and terminate requests is CLI controlled.
Note that the QoS-Info AVP will be encoded in all CCR-U messages if the CLI command "diameter
encode-event-avps always" is enabled. This implementation impacts only the dpca-custom15 dictionary.
Examples The following command enables to always send the event-related AVPs in all CCR messages:
diameter encode-event-avps always
diameter encode-supported-features
This command enables/disables encoding and sending of Supported-Features AVP.
Product GGSN
P-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
adc-rules
This keyword enables configuration of Application Detection and Control (ADC) rules over Gx interface.
For ADC 6th bit of supported feature will be set. By default, this supported feature will be disabled.
Important ADC Rule support is a licensed-controlled feature. Contact your Cisco account representative for detailed
information on specific licensing requirements.
This keyword "adc-rules" will be available only when the feature-specific license is configured.
In release 18, the gateway node will use ADC functionality over Gx as defined in the Release 11 specification
of 3GPP standard. ADC extension over Gx provides the functionality to notify PCRF about the start and stop
of a specific protocol or a group of protocols, and provide the possibility to PCRF that with the knowledge
of this information, change the QoS of the user when the usage of application is started and until it is finished.
The provision of ADC information is done through the ADC rule, the action initiated by PCRF is done through
the PCC rule.
ADC rules are certain extensions to dynamic and predefined PCC rules in order to support specification,
detection and reporting of an application flow. These rules are installed (modified/removed) by PCRF via
CCA-I/CCA-U/RAR events. ADC rules can be either dynamic PCC or predefined PCC rules, and the existing
attributes of dynamic and predefined rules will be applicable.
Dynamic PCC rule contains either traffic flow filters or Application ID. When Application ID is present, the
rule is treated as ADC rule. Application ID is the name of the ruledef which is pre-defined in the boxer
configuration. This ruledef contains application filters that define the application supported by P2P protocols.
PCEF will process and install ADC rules that are received from PCRF interface, and will detect the specified
applications and report detection of application traffic to the PCRF. PCRF in turn controls the reporting of
application traffic.
PCEF monitors the specified applications that are enabled by PCRF and generates Start/Stop events along
with the Application ID. Such application detection is performed independent of the bearer on which the ADC
PCC rule is bound to. For instance, if ADC rule is installed on a dedicated bearer whereas the ADC traffic is
received on default bearer, application detection unit still reports the start event to PCRF.
netloc
Enables the NetLoc feature. The NetLoc feature indicates the support for reporting of the Access Network
Information.
Important Network Provided Location Information (NPLI) feature is a license-controlled feature. A valid feature
license must be installed prior to configuring this feature. Contact your Cisco account representative for
more information.
A new feature "netloc" (feature bit 10) has been added as part of the Supported-Features AVP to implement
the Network provided Location Info (NPLI) feature for IMS. NPLI is used to support variety of applications
like emergency call, Lawful intercept, charging, etc.
The netloc feature bit will be sent to PCRF on demand via CCR-I message. A new event trigger
"ACCESS_NETWORK_INFO_REPORT (45)" and a new Diameter AVP "Required-Access-Info" have been
added to support the NPLI enhancement.
The gateway node provides the required access network information (e.g. user location and/or user time zone
information) to the PCRF within the 3GPP-User-Location-Info AVP, User-Location-Info-Time AVP (if
available), and/or 3GPP-MS-TimeZone AVP as requested by the PCRF. The gateway also provides the
ACCESS_NETWORK_INFO_REPORT event trigger within Event-Trigger AVP.
netloc-ran-nas-cause
Enables the Netloc-RAN-NAS-Cause feature. By default, this supported feature will be disabled.
This feature is used to send detailed RAN and/or NAS release cause code information from the access network
to PCRF. This feature is added to be in compliance with Release 12 specification of 3GPP TS 29.212. It
requires that the NetLoc feature is also supported.
A new feature "netloc-ran-nas-cause" (feature bit 22) has been added as part of the Supported-Features AVP
to support the 3GPP RAN/NAS Release Cause Code Information Element (IE) on Gx interface.
Important This feature can be enabled only when the NetLoc feature license is installed.
If the supported features "netloc-ran-nas-code" and "netloc" are enabled, then netloc-ran-nas-cause code will
be sent to PCRF via CCR-T message. A new Diameter AVP "RAN-NAS-Release-Cause" has been added to
support this feature. This AVP will be included in the Charging-Rule-Report AVP and in CCR-T for bearer
and session deletion events respectively.
pending-transactions
Configures the Pending Transactions feature as part of supported features. This keyword addition is to handle
race conditions on Gx i.e. process the Diameter messages in the order they are received.
Gx-based applications are vulnerable to certain race conditions (e.g. concurrent RAR/CCR). Enhancements
are done on the Diameter protocol to deterministically handle the race conditions on Gx.
In a scenario wherein RAR is received while waiting for CCA-U, Gx application rejects RAR with
Experimental-Result-Code AVP set to DIAMETER_PENDING_TRANSACTION. This should be done only
if PCRF supports this functionality otherwise Gx client should continue with the current implementation.
If race conditions are not processed properly, it can lead to unpredictable behavior from each node, resulting
in subscriber disconnection. With this feature, the outcome in such situation is deterministic and operator has
the ability to influence the node behavior aligned with their policy.
Important Currently only one pending transaction is supported. So, all other transactions (like handoffs, etc) while
one is pending will be rejected.
In 17.0 and later releases, in order to comply with 4G Network Upgrade 3GPP Standard, the following changes
are implemented:
• Support for Negotiation of PT in initial session establishment.
• Support for receiving/sending 4144 with 3GPP Vendor ID in CCA/RAA.
• Retry of CCR-U when 4144 is received from PCRF.
• No Support for 4198 with Proprietary Vendor ID.
• Recovery of negotiated Supported features.
session-recovery
Enables the Session Recovery feature. This functionality helps ensure that the PCRF and P-GW can be in
sync on session information and recover any lost Gx sessions. By default, session recovery and session sync
features are not enabled.
Gx sessions typically tend to be long-lived. In case of session loss in PCRF (e.g. due to software failure), or
a message loss in PCRF (e.g. Gx:RAA is dropped due to overload control), there is no existing mechanism
to allow the PCRF and P-GW to sync-up on session state like Rules Status, APN-AMBR, QoS, Event Triggers,
etc. In this release, the Gx interface between P-GW and PCRF has been enhanced to allow the PCRF and
P-GW to sync-up. This is currently not part of 3GPP 29.212.
Important In this release, the Session Recovery and Sync will be supported only for the IMS APN.
This keyword is used to achieve the session recovery. When this feature is enabled, P-GW and PCRF will
exchange session information and P-GW provides the complete subscriber session information to enable
PCRF to build the session state.
session-sync
Enables the Session Synchronization feature. This functionality helps ensure that the PCRF and P-GW can
be in sync on session information and recover any lost Gx sessions. By default, Session Recovery and Session
Sync features will not be enabled.
Gx sessions typically tend to be long-lived. In case of session loss in PCRF (e.g. due to software failure), or
a message loss in PCRF (e.g. Gx:RAA is dropped due to overload control), there is no existing mechanism
to allow the PCRF and P-GW to sync-up on session state like Rules Status, APN-AMBR, QoS, Event Triggers,
etc. The Gx interface between P-GW and PCRF is enhanced to allow the PCRF and P-GW to sync-up. This
is currently not part of 3GPP 29.212.
Important In this release, the Session Recovery and Sync will be supported only for the IMS APN.
This keyword is used to achieve the session sync-up. When this feature is enabled, P-GW and PCRF will
exchange session information and P-GW provides the complete subscriber session information to enable
PCRF to build the session state.
sgw-restoration
This keyword enables configuration of S-GW Restoration feature.
P-GW is configured to support S-GW Restoration feature. P-GW sends S-GW Restoration feature in
Supported-Features AVP through the CCR-I message during session creation. If P-GW receives S-GW
Restoration feature in Supported-Features AVP in CCA-I message, then P-GW enables S-GW Restoration
feature.
If P-GW and PCRF support S-GW Restoration feature, then the P-GW accepts CCA and RAR during S-GW
restoration. Only Rule removal or RAR with session release cause is processed. Any rule install or modify is
dropped. P-GW triggers CCR-U with PCC rule failure report and AN_GW_STATUS AVP to inform PCRF
that S-GW is down. After receiving the SGW_Restoration indication, PCRF does not initiate any rule install
or modification towards the P-GW. The P-GW informs the PCRF when the S-GW has recovered using the
Event-Trigger AVP set to AN_GW_CHANGE and including the AN-GW-Address AVP related to the restored
or new S-GW. If S-GW restoration is reported to PCRF, then the P-GW sends CCR-U with AN_GW_CHANGE
trigger.
If S-GW Restoration feature is not negotiated through the Supported-Features AVP, then P-GW falls back to
the old behavior as follows:
• Drops all internal updates towards PCRF
• Rejects CCA and RAR during S-GW Restoration
• Does not include AN_GW_STATUS as AN_GW_FAILED (0) AVP in CCR-U
• Sends an RAA command with the Experimental-Result-Code set to UNABLE_TO_COMPLY (5012)
upon receiving RAR command
After configuring the S-GW Restoration feature on Gx interface, the failure is sent to PCRF with
Rule-Failure-Code as AN_GW_FAILED in both failure and restoration scenarios.
sponsored-connectivity
Enables the Sponsored (data) Connectivity feature.
With sponsored data connectivity, the sponsor has a business relationship with the operator and the sponsor
reimburses the operator for the user's data connectivity in order to allow the user access to an associated
Application Service Provider's (ASP) services. Alternatively, the user pays for the connectivity with a transaction
which is separate from the subscriber's charging. It is assumed the user already has a subscription with the
operator.
The purpose of this feature is to identify the data consumption for a certain set of flows differently and charge
it to sponsor. To support this, a new reporting level "SPONSORED_CONNECTIVITY_LEVEL" is added
for reporting at Sponsor Connection level and two new AVPs "Sponsor-Identity" and
"Application-Service-Provider-Identity" have been introduced at the rule level.
This CLI command "diameter encode-supported-features" has been added in Policy Control Configuration
mode to send Supported-Features AVP with Sponsor Identity.
Sponsored Connectivity feature will be supported only when both P-GW and PCRF support 3GPP Rel. 10.
P-GW advertises release as a part of supported features in CCR-I to PCRF. If P-GW supports Release 10 and
also Sponsored Connectivity but PCRF does not support it (as a part of supported features in CCA-I), this
feature is turned off.
This feature implementation impacts only the Gx dictionary "dpca-custom15".
virtual-apn
This keyword enables configuration of Gx-based Virtual APN (VAPN) feature. For VAPN 4th bit of supported
feature will be set. By default, this supported feature will be disabled.
Important Gx-based VAPN is a licensed-controlled feature. Contact your Cisco account representative for detailed
information on specific licensing requirements.
This keyword "virtual-apn" will be available only when the feature-specific license is configured.
In releases prior to 19, VAPN selection was possible through RADIUS or local configuration. In Release 19,
ASR5K uses PCRF and Gx interface for Virtual APN selection to achieve signaling reduction.
This keyword enables Gx based Virtual APN Selection feature for a given IMS authorization service. When
this configuration is enabled at P-GW/GGSN, then P-GW/GGSN advertises this feature to PCRF through the
Supported-Features AVP in CCR-I. When the VAPN is selected, then the PCRF rejects the CCR-I message
with the Experimental-Result-Code AVP set to 5999 (DIAMETER_GX_APN_CHANGE), and sends a new
APN through the Called-Station-Id AVP in CCA-I message. The existing call is then disconnected and
established with the new virtual APN. Note that the Experimental Result Code 5999 will have the Cisco
Vendor ID.
Important Enabling this feature might have CPU impact (depending on the number of calls using this feature).
Limitations:
• Virtual APN supported feature negotiation, Experimental Result Code (5999), Called-Station-Id AVP
should be received to establish the call with new virtual APN. When any one of conditions is not met
then the call will be terminated.
• Failure-handling will not be taken into account for 5999 result-code when received in the CCA-I message.
• When the Experimental Result Code 5999 is received in the CCA-U then failure-handling action will
be taken.
• If the Called-Station-Id AVP is received in CCA-U or CCA-T, then the AVP will be ignored.
• If virtual-apn is received in local-policy initiated initial message then the call will be terminated.
• When PCRF repeatedly sends the same virtual-apn, then the call will be terminated.
default | no
This keyword removes the previously configured supported features.
Usage Guidelines This command is used to enable encoding and sending of Supported-Features AVP.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
subscriber-limit subs_limit
Specifies the limit of subscribers to switch or reselect the PCRF for subscribers not more than subs_limit in
time duration of duration second(s).
subs_limit must be an integer from 1 through 10000000.
time-interval duration
Specifies the time duration, in seconds, to reselect PCRF for subscribers not more than subs_limit in time
duration of duration second(s).
duration must be an integer from 1 through 3600.
default
Applies the default setting for this command.
Sets the PCRF reselection or switching to default state.
no
Removes the configured PCRF reselection method and disables the reselection or switching of PCRF.
Usage Guidelines Use this command to specify the pacing of reselection or switching of the PCRF in an IMS authorization
service..
In case IMS authorization session have been opened on certain PCRF on the basis of the current selection
table, and the current active table configuration is changed, the IMSA starts selection procedure for the PCRF.
Existing sessions on current PCRF from earlier table is required to close and reopened on the selected PCRF
from the new table. This reselection periodicity is controlled by this command and it indicates the number of
subscriber sessions subs_limit to be reselected or moved in duration seconds.
For example, if this command is configured with 100 subscribers and 2 seconds, then the system reselects the
PCRF for no more than 100 subscribers per 2 seconds.
Examples The following command sets the system to reselect the new PCRF for no more than 1000 subscriber in 15
seconds:
diameter host-select reselect subscriber-limit 1000 time-interval 15
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
Syntax Description diameter host-select row-precedence precedence_value table { { { 1 | 2 } host host_name [ realm realm_id
] [ secondary host host_name [ realm realm_id ] ] } | { prefix-table { 1 | 2 } msisdn-prefix-from
msisdn_prefix_from msisdn-prefix-to msisdn_prefix_to host host_name [ realm realm_id ] [ secondary host
sec_host_name [ realm sec_realm_id ] algorithm { active-standby | round-robin } ] } } [ -noconfirm ]
no diameter host-select row-precedence precedence_value table { { 1 | 2 } | prefix-table { 1 | 2 } }
Important In 8.1 and later releases, precedence_value must be an integer from 1 through 128. In 8.0 and previous
releases, precedence_value must be an integer from 1 through 100.
table { 1 | 2 }: Specifies the Diameter host table to add/append the primary and secondary Diameter host
addresses.
host host_name: Specifies the primary host name. host_name must be an alphanumeric string of 1 through
127 characters in length.
realm realm_id: Specifies the primary realm ID. realm_id must be an alphanumeric string of 1 through 127
characters in length.
secondary host sec_host_name [ realm sec_realm_id ]: Specifies the secondary host name and realm ID:
host sec_host_name: Specifies the secondary host name. host_name must be an alphanumeric string of 1
through 127 characters in length.
realm sec_realm_id: Specifies the secondary realm ID. realm_name must be an alphanumeric string of 1
through 127 characters in length.
Important In 8.1 and later releases, precedence_value must be an integer from 1 through 128. In 8.0 and previous
releases, precedence_value must be an integer from 1 through 100.
prefix-table { 1 | 2 }: Specifies the MSISDN prefix range table to add the primary and/or secondary Diameter
host addresses.
msisdn-prefix-from msisdn_prefix_from: For a range of MSISDNs, specifies the starting MSISDN.
msisdn-prefix-to msisdn_prefix_to: For a range of MSISDNs, specifies the ending MSISDN.
Important To enable the Gx interface to connect to a specific PCRF for a range of MSISDNs/subscribers configure
msisdn_prefix_from and msisdn_prefix_to with the starting and ending MSISDNs respectively. The
MSISDN ranges must not overlap between rows. To enable the Gx interface to connect to a specific PCRF
for a specific MSISDN/subscriber, configure both msisdn_prefix_from and msisdn_prefix_to with the
same MSISDN.
host host_name: Specifies the primary host name. host_name must be an alphanumeric string of 1 through
127 characters in length.
realm realm_id: Specifies the primary realm ID. realm_id must be an alphanumeric string of 1 through 127
characters in length.
secondary host sec_host_name [ realm sec_realm_id ]: Specifies the secondary host name and realm ID:
host sec_host_name: Specifies the secondary host name. sec_host_name must be an alphanumeric string of
1 through 127 characters in length.
realm sec_realm_id: Specifies the secondary realm ID. sec_realm_id must be an alphanumeric string of 1
through 127 characters in length.
algorithm { active-standby | round-robin }: Specifies the algorithm for selection between primary and
secondary servers in the MSISDN prefix range table.
Default: active-standby
active-standby: Specifies selection of servers in the Active-Standby fashion.
round-robin: Specifies selection of servers in the Round-Robin fashion.
Important The Round Robin algorithm for PCRF selection is effective only over a large number of PCRF selections,
and not at a granular level.
[-noconfirm]
Specifies that the command is to execute without any additional prompt and confirmation from the user.
Usage Guidelines Use this command to add, update, or delete rows specified with a precedence from a Diameter host table or
MSISDN prefix range table.
In the Rel. 7 Gx implementation, when the Gateway interworks with multiple PCRFs, the Gateway can
configure the primary and secondary server based on the MSISDN-prefix range in the MSISDN prefix range
table. Using this command, you can add a new prefix row into the MSISDN prefix table.
If a row with the precedence that you add already exists in a table, the existing prefix row is removed and the
new row is inserted with the same precedence.
Examples The following command adds a row with precedence 12 in table 2 with primary host name as star_ims1 and
secondary host name as star_ims2 to Diameter host table.
diameter host-select row-precedence 12 table 2 host star_ims1 secondary host star_ims2
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
• prefer-ipv4: Specifies that IPv4 addresses are to be used, if an IPv4v6 call is received, for selecting the
rows in the host table.
• prefer-ipv6: Specifies that IPv6 addresses are to be used, if an IPv4v6 call is received, for selecting the
rows in the host table.
• msisdn-modulus: This algorithm divides the MSISDN value in binary without the leading "+" of the
subscriber by the number of rows in the table, and the remainder is used as an index in the specific table
to select the row.
• round-robin: This algorithm rotates all rows in the active table for selection of the row in round-robin
fashion. If no algorithm is specified this is the default behavior.
Important The Round Robin algorithm is effective only over a large number of selections, and not at a granular level.
default
Applies the default setting for this command.
no
Removes previous configuration.
When no table is selected, the system will not communicate with any PCRF for new sessions.
Usage Guidelines Use this command to configure the Diameter host table and row selection methods to select host name or
realm for PCRF.
When this command is used to change which table the system should be using, user must re-determine which
E-PDF the system should be using for each subscriber. If a different E-PDF results from the configuration
change in the table, the system will wait for all of the IMS sessions for the subscriber to be no longer active
and then the system either closes/opens Gx sessions with the old/new PDFs respectively, or the system
deactivates the PDP contexts of the subscriber.
Here is an example of how row selection is configured for three hosts that the system will use for load-balancing.
Operator can configure six rows in a table, as follows.
1 1 3
2 2 1
3 2 3
4 3 1
5 3 2
In the above table, the three hosts are named 1, 2, and 3. When all hosts are working, the load will be distributed
among all the three hosts. If host 1 fails, then the load will be distributed between the remaining two hosts.
In this scenario, the modulo 6 results of 2 and 4 will return rows that have primary hosts but no working
back-up host.
In the Rel. 7 Gx implementation, the GGSN/PCEF is required to discover and select an appropriate PCRF to
establish control relationship at primary PDP context activation. The ip-address-modulus, msisdn-modulus,
and round-robin algorithms are supported by the GGSN/PCEF for PCRF discovery. In addition, the
active/standby and round-robin algorithms are used for selection between primary and secondary servers
based on the MSISDN Prefix Range Table.
Examples The following command specifies table 1 with round-robin algorithm to select the rows with host name for
E-PDF in Diameter host table.
diameter host-select table 1 algorithm round-robin
diameter host-select-template
This command specifies the Diameter host server template to be associated with this IMS Authorization
service. The service uses the specified template (and associated host-select table) to select a Diameter peer
server. It then uses the returned host name(s) to contact the PCRF and establish the call.
Product GGSN
HA
HSGW
IPSG
PDSN
P-GW
SAEGW
S-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
no
Removes the binding of the Diameter host select template with the IMS Authorization service.
tmplt_name
Specifies the name of an existing Diameter host server template (configured in Global Configuration mode)
to bind with the IMS Authorization service. It is an alphanumeric string of 1 through 255 characters.
Usage Guidelines Use this command to bind a configured Diameter host select template to the IMS Authorization service for
DPCA. This IMS authorization service searches the associated host select table to select a Diameter peer
server. For additional information refer to the Diameter Host Select Configuration Mode Commands chapter
and the description of the diameter-host-template command in the Global Configuration Mode Commands
chapter.
Important Prior to issuing this command, the Diameter host select template should be configured using the
diameter-host-template command in the Global Configuration mode.
Important If no association is made to the template then the diameter peer-select command configured at the
application level will be used for peer selection.
Examples The following command binds a configured Diameter host select template named diamtemplate to the IMS
authorization service:
diameter host-select-template diamtemplate
diameter map
This command enables selecting the value to which the USAGE_REPORT and APN_AMBR_MOD_FAILURE
Event-Trigger should be mapped to.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
usage-report { 29 | [ 26 | 33 ] [ 26 | 33 ] }
Maps the USAGE_REPORT of Event-Trigger AVP to one or a combination of these values.
• 26 – Event-Trigger 26 will mapped to USAGE_REPORT. Note this will not affect any other
Event-Trigger.
• 29 – Event-Trigger 29 will mapped to USAGE_REPORT, and 33 to APN_AMBR_MOD_FAILURE.
• 33 – Event-Trigger 33 will mapped to USAGE_REPORT, and 29 to APN_AMBR_MOD_FAILURE.
default
The default behavior is to configure the Event-Trigger USAGE_REPORT to be mapped to 26.
Usage Guidelines The Event-Trigger AVP's USAGE_REPORT has been given different values in the 3GPP TS 29.212 standard
spec. As a result of that, the releases of TS 29.212 are not backward compatible. To address this, this CLI
command has been introduced in Policy Control configuration mode to map the USAGE_REPORT to either
26/29/33 or a combination of these values in order to be flexible enough to interoperate with various operators.
• TS 29.212 v9.5.0 - USAGE_REPORT (26)
• TS 29.212 v9.6.0 - USAGE_REPORT (29)
If this CLI command diameter map usage-report 29 is configured in the chassis and PCRF sends 29
event-trigger then on volume threshold breach CCR-U with volume-report and event-trigger 29 will be sent
to the PCRF. Same is the case with the values 26 and 33.
In 17.1 and later releases, to be able to gracefully handle the change when moving between 3GPP releases
supporting the different values for the Usage Report, the existing CLI command diameter map usage-report
is modified to support configuration of multiple values of usage report mapping. While migrating from older
versions to current version, all of the sessions created before the migration will continue to use 26 as usage
report event trigger value. The new session will use usage-report value based on PCRF value or default value.
In releases prior to 17.1, when diameter map usage-report is mapped to 26, then APN AMBR modification
failure event trigger is not supported. In 17.1 and later releases, APN AMBR modification failure event trigger
is supported for all usage report trigger values (26, 33, 29).
Examples The following command maps the Event-Trigger USAGE_REPORT to 29 and APN_AMBR_MOD_FAILURE
to 33:
diameter map usage-report 29
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
endpoint endpoint_name
endpoint_name is the Diameter endpoint configured in Context Configuration Mode to bind with IMS
authorization service, and must be an alpha/numeric string of 1 through 63 characters in length.
no
Removes the binding of Diameter origin endpoint with IMS Authorization service.
Usage Guidelines Use this command to bind a configured Diameter origin endpoint to the IMS Authorization service for DPCA.
This IMS authorization service searches all system contexts until it finds one with a matching Diameter origin
endpoint name specified.
Examples The following command binds a configured endpoint named test to the IMS authorization service:
diameter origin endpoint test
diameter request-timeout
This command configures the request-timeout setting for Diameter-IMSA Gx interface.
Product GGSN
P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
Syntax Description diameter request-timeout timeoutdeciseconds msg-type { any | ccr-initial | ccr-terminate | ccr-update }
default diameter request-timeout
timeout
Specifies the timeout duration (in deciseconds). The value must be an integer from 1 through 3000.
Default: 10 seconds
default
Applies the default setting for this command.
Usage Guidelines Use this command to configure the request-timeout setting for Diameter-IMSA Gx interface. At the
request-timeout value, DPCA will apply failure-handling to the subscriber. Action will be taken based on the
failure-handling configuration (terminate/retry-terminate/continue).
Examples The following command configures the Diameter request-timeout setting to 20 seconds:
diameter request-timeout 20
diameter sgsn-change-reporting
This command enables reporting of SGSN_CHANGE event trigger and SGSN-Address AVP for 2G and 3G
calls on GnGp P-GW.
Product GGSN
P-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
sgsn-change-reporting
This keyword specifies to detect SGSN change and send SGSN-Address AVP and SGSN_CHANGE event
trigger for a subscriber in 2G/3G on Gx interface during GnGp scenario.
no
This variant specifies to send AN-GW-Address AVP during the call setup, when SGSN change happens, or
during the handoff from 4G to 3G. This is the default setting.
Usage Guidelines The current implementation does not send SGSN_CHANGE event trigger and SGSN- Address AVP. Instead
it sends AN-GW-Address AVP and AN_GW_CHANGE event trigger for GnGp case. This behavior is not
compliant to 3GPP standard TS 29.212 specification. Hence, in release 18, this CLI command "diameter
sgsn-change-reporting" has been introduced to control this behavior.
This release provides, the GnGp P-GW users, the flexibility to configure detection of SGSN_CHANGE event
trigger and to send SGSN-Address AVP for a subscriber in 2G/3G on Gx interface, so that PCRF can use this
information to apply appropriate policies.
In releases prior to 18, AN-GW-Address AVP was sent in CCR-I message on GnGp scenario.
AN_GW_CHANGE event trigger and AN-GW-Address AVP were sent when the inter-sgsn handoff or 4G
to 2G/3G GnGp handoff happens.
When this CLI command is configured, SGSN-Address AVP will be sent in the CCR-I message for 2G/3G
GnGp P-GW subscribers. SGSN_CHANGE event trigger and SGSN-Address AVP will be sent when the
inter-sgsn handoff or 4G to 2G/3G GnGp handoff happens.
Important This feature is applicable only for SGSN IPv4 address. For SGSN IPv6 address, the SGSN-Address AVP
will not be sent.
By default, AN-GW-Address AVP will be sent during the call setup, when SGSN change happens, or during
the handoff from 4G to 3G.
Examples The following command configures to detect SGSN change and send SGSN-Address AVP in CCR-I :
diameter sgsn-change-reporting
diameter update-dictionary-avps
This command enables dictionary control of the AVPs that need to be added based on the version of the
specification to which the PCEF is compliant with.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default | no
Configures this command with the default setting.
The default behavior is that R9 support will not be indicated as part of Supported-Features AVP in a R7/R8
dictionary and R8 support will not be indicated as part of Supported-Features AVP in a R7 dictionary.
3gpp-r8
Specifies to select the 3GPP Rel. 8 AVPs for encoding.
3gpp-r9
Specifies to select the 3GPP Rel. 9 AVPs for encoding.
3gpp-r10
Specifies to select the 3GPP Rel. 10 AVPs for encoding.
Usage Guidelines
Important This command is applicable only to Diameter dictionaries that support standard based volume reporting
over Gx feature.
Use this command to encode the AVPs in the dictionary based on the release version of the specification to
which the PCEF is compliant with.
Release 12.0 onwards, if a 3GPP Rel. 7 based dictionary is already configured with diameter dictionary
dpca-custom4 command, and then if the diameter update-dictionary-avps 3gpp-r9 command is applied,
the Supported-Features AVP with feature bit 1 being set will be sent in the CCR-I to indicate that 3GPP Rel.
9 AVPs are also supported.
Both default and no command have the same behavior, as if the CLI command is not configured. Hence, in
the output of show configuration verbose command, the default and no command is shown as no diameter
update-dictionary-avps.
This CLI command when configured results in behavioral changes as indicated in the following table.
3GPP Rel. 7 based dictionary upgraded In the CCR-I, Supported-Features AVP will be encoded with value
to 3GPP Rel. 8 1 for the Feature-List AVP.
For example: [V] [M] Supported-Features:
diameter dictionary dpca-custom4 [M] Vendor-Id: 10415
diameter update-dictionary-avps [V] [M] Feature-List-ID: 1
3gpp-r8 [V] [M] Feature-List: 1
The Feature-List AVP value suggest that it is 3GPP Rel. 8
compliant. But, it is not fully complaint to 3GPP Rel. 8.
In the current release, for this upgrade scenario (3GPP Rel. 7 to
3GPP Rel. 8), none of the features mentioned in 3GPP Rel. 8 will
be supported.
3GPP Rel. 9 based dictionary upgraded In the CCR-I, value for the Feature-List AVP in the
to 3GPP Rel. 10 Supported-Features AVP will be 8.
For example: [V] [M] Supported-Features:
diameter dictionary r8-gx-standard [M] Vendor-Id: 10415
diameter update-dictionary-avps [V] [M] Feature-List-ID: 1
3gpp-r10 [V] [M] Feature-List: 8
The Feature-List AVP value suggest that it is 3GPP Rel. 10
compliant. But, it is not fully complaint to 3GPP Rel. 10.
In 14.1 and later releases, Supported-Features AVP is extended to support 3GPP Rel. 10 in EPS 3.0 in addition
to 3GPP Rel. 8 and Rel. 9. If the diameter update-dictionary-avps 3gpp-r10 command is applied, the
Supported-Features AVP with feature bit 1 being set will be sent in the CCR-I / CCA to indicate that 3GPP
Rel. 10 AVPs are also supported. The 'M' bit setting for the Feature-List AVP and Feature-List-ID AVP must
be the same as defined in 3GPP TS 29.229 and must not be affected by the 'M' bit setting of the
Supported-Features AVP.
Examples The following command enables encoding of AVPs in the dictionary based on 3GPP Rel. 9:
diameter update-dictionary-avps 3gpp-r9
end
Exits the current configuration mode and returns to the Exec mode.
Product All
endpoint-peer-select
This command enables Diabase to select the Diameter peers in all failure scenarios.
Product GGSN
PGW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
on-host-select-failure
Specifies to perform server selection at Diabase when the hosts could not be selected by IMS Authorization
application.
on-inactive-host
Specifies to perform server selection at diabase when the hosts selected by application are inactive.
default | no
Default/no behavior is to terminate the call when the hosts could not be selected by application or when the
hosts selected by application are inactive.
Usage Guidelines Use this command to perform server selection at Diabase when the hosts could not be selected by application
or when the hosts selected by the IMS Authorization application is inactive. For example, host table is not
configured in IMSA service, host table is configured but not activated, none of the rows in prefix table match
the subscriber, host template is not associated with IMSA service, host template could not select the hosts.
This CLI command is added in policy control configuration mode to maintain backward compatibility with
the old behavior of terminating the call when server selection fails at application.
Examples The following command enables Diabase to select peers when the hosts selected by application are inactive.
endpoint-peer-select on-inactive-host
event-report-indication
This command enables event report indication.
Product P-GW
SAEGW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default | no
Disables event report indication.
Usage Guidelines Use this command to determine what type of event changes are reported from the PCRF.
Examples The following command enables event report indication for all triggers.
event-report-indication all
event-update
This command configures sending usage monitoring information in event updates either for all event triggers
or for a specific event trigger.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
Configures the default setting for this command.
Default: Usage report is not sent in event update.
no
Disables sending usage report in event update.
reset-usage
Resets the usage at PCEF after reporting in event update.
Usage Guidelines Use this command to send volume usage information when an event change is reported to the PCRF in a
CCR-U message.
To send customized usage information based on specific event triggers, the event should be accordingly
configured with the event-update send-usage-report events command. For example, if the usage report is
required whenever RAT change occurs, this can be accomplished using the event-update send-usage-report
events rat-change command.
Examples The following command specifies to send volume usage report in event updates to the PCRF for all event
triggers:
event-update send-usage-report reset-usage
The following command specifies to send volume usage report in event updates to the PCRF for RAT change
scenarios:
event-update send-usage-report reset-usage events rat-change
The following command specifies to send volume usage report in event updates to the PCRF if either RAT
change or QOS change occurs:
event-update send-usage-report reset-usage events rat-change qos-change
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
failure-handling
This command configures Diameter failure handling behavior.
Product All
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
no
Disables previous failure-handling configuration.
retry-and-terminate
Specifies that in the event of a failure the user session continues for the duration of one retry attempt with the
server. If this retry attempt also fails, the session is terminated.
terminate
Specifies that in the event of a failure the user session be terminated.
Important This keyword is valid only for update-request though it is allowed to configure for all
the requests. The failure-handling command configuration will throw an error/warning
message if it is configured for any request other than the update request.
Important Failure handling action "continue retry-server-on-event" will be taken only if failure
happens to CCR-U message, not for CCR-I messages.
Important This keyword is valid only for update-request though it is allowed to configure for all
the requests. The show configuration errors command will throw an error/warning
message if it is configured for any request other than the update request.
• retry-and-terminate: In the event of a failure the user session continues for the duration of one retry
attempt with the server. If this retry attempt also fails, the session is terminated.
• terminate: In the event of a failure the user session is terminated.
• continue: In the event of a failure the user session continues. DPCA/Diameter will make periodic request
and/or connection retry attempts and/or will attempt to communicate with a secondary peer depending
on the peer config and session-binding setting.
• retry-and-terminate: In the event of a failure the user session continues for the duration of one retry
attempt with the server. If this retry attempt also fails, the session is terminated.
• terminate: In the event of a failure the user session is terminated.
cc-request-type
As in 8.0 release:
This optional keyword defines the type of credit control request with failure result code and credit control
failure handling action for a session.
• any-request: Specifies the request type as any request for a new session.
• initial-request: Specifies the request type as initial request for a new session.
• terminate-request: Specifies the request type as terminate request for a session.
• update-request: Specifies the request type as update request for an active session.
Usage Guidelines Use this command to configure the Diameter Policy Control Application (DPCA) failure handling behavior.
When an unknown rulebase comes in CCA, changing of rulebase and failure handling is managed in the
following manner:
• If the new and existing rulebases have the same CCA policy, then switch to the new rulebase is successful.
• If the new rulebase is valid and has CCA-enabled, in CCA-Initial/Update request, switch to the new
rulebase is successful.
• If the new rulebase is valid and does NOT have CCA enabled, whereas the existing rulebase has credit
enabled, or vice versa, in CCA-Initial/Update request:
◦CCFH-Continue: Goes offline immediately after sending the CCR-T with termination cause as
BAD_ANSWER.
◦CCFH-RETRY and TERMINATE: Goes offline immediately after sending the CCR-T with
termination cause as BAD_ANSWER.
◦CCFH-TERMINATE: Goes offline immediately after sending the CCR-T with termination cause
as BAD_ANSWER.
Important Unreported usage will be lost, if CCR-U message delivery fails for last rule removal or usage reporting
for monitoring stop indication from PCRF. Also, note that preserving unreported usage monitoring
information is currently not supported for dpca-custom9 dictionary.
Examples The following command sets the DPCA failure handling to retry-and-terminate and return a result code of
3456 for credit control request type initial-request:
As in 8.0 release:
failure-handling diameter-result-code 3456 ccfh retry-and-terminate cc-request-type initial-request
As in 8.1 and later releases:
failure-handling cc-request-type initial-request diameter-result-code 3456 retry-and-terminate
li-secret
Refer to the Cisco ASR 5000 Lawful Intercept Configuration Guide for a description of this command.
max-outstanding-ccr-u
This command enables or disables the gateway to send multiple back-to-back CCR-Us to PCRF.
Product GGSN
HA
PDSN
P-GW
S-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
default
This keyword sets the default value as 1 for the maximum number of outstanding CCR-U messages to be sent
to PCRF.
value
This keyword configures a value for the maximum number of outstanding CCR-U messages to be sent to
PCRF.
value must be an integer value from 1 through 12.
Usage Guidelines This command enables the gateway to send multiple outstanding CCR-Us per session to PCRF.
In releases prior to 17.0, ASR5K node supports only one pending CCR-U message per session over Gx
interface. Any request to trigger CCR-U (for access side updates/internal updates) were ignored/dropped,
when there was already an outstanding message pending at the node. PCEF and PCRF were out of synch if
CCR-U for critical update (like RAT change/ULI change) was dropped.
In 17.1 and later releases, this CLI command "max-outstanding-ccr-u" under IMS Authorization Service
configuration mode allows multiple CCR-Us towards PCRF. That is, this CLI will allow the user to configure
a value of up to 12 as the maximum number of CCR-U messages per session.
The CLI-based implementation allows sending request messages as and when they are triggered and processing
the response when they are received. The gateway does re-ordering if the response messages are received out
of sequence.
Examples The following command configures the maximum number of outstanding CCR-U messages as 2.
max-outstanding-ccr-u 2
subscription-id service-type
This command enables required subscription-id types for various services. The Subscription-ID AVP will be
encoded based on the configured subscription-ID type.
Product GGSN
HA
IPSG
PDSN
P-GW
Command Modes Exec > Global Configuration > Context Configuration > IMS Authorization Configuration > Policy Control
Configuration
configure > context context_name > ims-auth-service service_name > policy-control
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-imsa-dpca)#
Syntax Description subscription-id service-type { closed_rp | ggsn | ha | ipsg | l2tplns | mipv6ha | pdsn | pgw } { e164 | imsi
| nai } +
{ default | no } subscription-id service-type { closed_rp | ggsn | ha | ipsg | l2tplns | mipv6ha | pdsn | pgw
}
default | no
Configures this command with the default setting.
The default behavior is that Subscription-ID AVP will be encoded based on service-type and Diameter
dictionary.
{ closed_rp | ggsn | ha | ipsg | l2tplns | mipv6ha | pdsn | pgw } { e164 | imsi | nai }
Controls the encoding of Subscription-ID AVP based on the following service-types associated with services
such as GGSN, HA, IPSG, PDSN, etc.
• E164
• IMSI
• NAI
+
Indicates that more than one of the keywords can be entered in a single command.
Usage Guidelines In releases prior to 15.0, Subscription-ID AVP is encoded based on service-type and Diameter dictionary.
In 15.0 and later releases, when IMS Authorization service encodes the Subscription-ID AVP, IMSA will
first check whether or not this CLI command subscription-id service-type is configured. If the CLI is
configured for the current service, then IMSA will encode the Subscription-ID AVP based on the configured
subscription-ID type. This CLI command takes more precedence than the default behavior.
If the CLI configuration does not encode any Subscription-ID AVP, then IMSA will encode this AVP based
on the default behavior. For example, in GGSN/IPSG service, NAI support is not available. If this CLI
command is configured for GGSN/IPSG service with NAI type, then based on CLI IMSA cannot encode any
Subscription-ID AVP. By this time default behavior (old behavior based on service-type and dictionary) will
add the subscription-ID.
Examples The following command enables encoding of the Subscription-ID AVP based on IMSI parameter for GGSN
service:
subscription-id service-type ggsn imsi
This chapter describes the commands available in the Plugin Configuration Mode. This mode is associated
with the Dynamic Software Upgrade (DSU) process described in the System Administration Guide.
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
attribute
This command is not supported in this release.
Product All
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
module priority
Configures the priority in the Version Priority List (VPL) for a specified version of a plugin module.
Product ADC
Syntax Description module priority number version plugin_version attribute attribute_name attribute_value
no module priority number
no
Removes the configured priority in the module priority list.
priority number
Specifies the priority of the plugin module as an integer from 1 through 100. Priority "1" has the highest
priority.
version plugin_version
Specifies the version number of the plugin module. The version number is derived from the filename of the
downloaded plugin.
Usage Guidelines Assign a priority number to a specific version of patch for the corresponding plugin. The priority number in
the module priority list determines which version will be loaded when the update module command is used.
The show plugin command displays the VPL configuration status of this plugin module.
Examples The following command sets the priority of p2p plugin module version 1.17.4340 to 2 where the filename
was libp2p-1.2.0.so.tgz:
module priority 2 version 1.2.0
The Permanent Virtual Connection (PVC) configuration mode commands bind IP interfaces or SS7-Frame
Relay links a PVC as well as configure PVC operational parameters for a specific port.
Command Modes Exec > Global Configuration > ATM Port Configuration > PVC Configuration
configure > port atm slot_number/port_number > pvc vpi vpi_number vci vci_number
Entering the above command sequence results in the following prompt:
[local]host_name(config-port-slot_number/port_number-pvc-pvc_number/vci_number)#
Important The commands or keywords/variables that are available are dependent on platform type, version, and
installed license(s).
bind
This command binds an IP interface or an SS7 link to the PVC.
Important Prior to attempting the binding, the interface and context or the SS7 routing information and link must
have been configured.
Product SGSN
Command Modes Exec > Global Configuration > ATM Port Configuration > PVC Configuration
configure > port atm slot_number/port_number > pvc vpi vpi_number vci vci_number
Entering the above command sequence results in the following prompt:
[local]host_name(config-port-slot_number/port_number-pvc-pvc_number/vci_number)#
Syntax Description [ no ] bind { interface interface_name context_name | link ss7-routing-domain rd_id linkset-id id link-id
id }
no
Removes the binding from the configuration.
interface_name
Defines the name of the virtual interface to be bound to the PVC. interface_name: Must be a unique string
consisting of 1 to 79 alphanumeric characters.
context_name
Specifies the name of the context to be bound to the virtual interface. context_name: Must be a unique string
consisting of 1 to 79 alphanumeric characters.
ss7-routing-domain rd_id
Identifies a specific SS7 routing domain. rd_id must be an integer from 1 to 12
linkset-id id
Identifies a specific linkset within the routing domain. id: must be an integer from 1 to 33
link-id id
Identifies a specific link within the linkset. id: must be an integer value 1 - 16
Usage Guidelines Use this command to bind the PVC to an interface or a specific link.
Examples Use a command similar to the following to bind a PVC to a link ID #2:
bind ss7-routing-domain 1 linkset-id 23 link-id 2
encapsulation aal5
Specify the data encapsulation type for the ATM adaptation layer 5 (AAL5) frames for the PVC.
Product SGSN
Command Modes Exec > Global Configuration > ATM Port Configuration > PVC Configuration
configure > port atm slot_number/port_number > pvc vpi vpi_number vci vci_number
Entering the above command sequence results in the following prompt:
[local]host_name(config-port-slot_number/port_number-pvc-pvc_number/vci_number)#
llc-snap
Frames protocol is identified in the AAL5 using logical link control (LLC) encapsulation.
vc-mux
Frames are not encapsulated and use virtual circuit multiplexing (VC-MUX) to identify the protocols used
for the AAL5 frames.
Usage Guidelines Use this command to identify the protocol type for the circuit.
end
Exits the PVC configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the PVC configuration mode and returns to the ATM port configuration mode.
Product SGSN
shaping
Specify the type of traffic shaping (rates) for this PVC.
Product SGSN
Command Modes Exec > Global Configuration > ATM Port Configuration > PVC Configuration
configure > port atm slot_number/port_number > pvc vpi vpi_number vci vci_number
Entering the above command sequence results in the following prompt:
[local]host_name(config-port-slot_number/port_number-pvc-pvc_number/vci_number)#
Syntax Description shaping { cbr pcr prc_num | ubr pcr prc_num | ubr+ pcr prc_num mrc mrc_num | vbr pcr prc_num scr
src_num mbs mbs_num }
cbr
Constant bit rate
pcr - peak cell rate = cells per second
prc_num: Must be an integer from 75 to 1412830
ubr
Unspecified Bit Rate
pcr - peak cell rate = cells per second
prc_num: Must be an integer from 75 to 1412830
ubr+
Unspecified Bit Rate with Minimum Cell Rate.
The PCR and MCR values should be set to maintain the following relationship: PCR >= (MCR + minRate),
where the current recomment minRate is 75.
pcr - peak cell rate = cells per second
prc_num: Must be an integer from 75 to 1412830
mcr - minimum cell rate
mrc_num: Must be an integer from 75 to 1412830
vbr
Variable Bit Rate, NRT (not real time) type.
The PCR and MCR values should be set to maintain the following relationship: PCR >= (MCR + minRate),
where the current recomment minRate is 75.
pcr - peak cell rate = cells per second
prc_num must be an integer from 75 to 1412830
scr - sustained cell rate
src_num must be an integer from 75 to 1412830
mbs - maximum burst size
mbs_num must be an integer from 75 to 1412830
Usage Guidelines Use this command to configure the shaping for egress traffic on this PVC.
shutdown
Disables/enables traffic over the current VLAN.
Product SGSN
Command Modes Exec > Global Configuration > ATM Port Configuration > PVC Configuration
configure > port atm slot_number/port_number > pvc vpi vpi_number vci vci_number
Entering the above command sequence results in the following prompt:
[local]host_name(config-port-slot_number/port_number-pvc-pvc_number/vci_number)#
no
Enables the VLAN. When omitted the VLAN is non-functional.
Examples To disable a VLAN from sending or receiving network traffic use the following command:
shutdown
To enable a VLAN use the following command:
no shutdown
The PVC (permanent virtual connection) Interface configuration mode is used to create and manage the IP
parameters for PVC interface(s) associated with an OLC (ATM-type) for a specific context.
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
description
Defines descriptive text to provide useful information about the current interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Erases the port's description from the configuration file.
text
text: Must be a string of 1 to 79 alphanumeric characters with no spaces or a string within double quotes that
includes printable characters. The description is case-sensitive.
Usage Guidelines Set the description to provide helpful information, for example the port's primary function, services, end users.
Define any information, the only limit is the number of characters, 79.
end
Exits the PVC interface configuration mode and returns to the Exec mode.
Product SGSN
exit
Exits the interface configuration mode and returns to the context configuration mode.
Product SGSN
ip
The commands in this section are used to configure the IP parameters for the PVC interface.
Important Before configuring the OSPF parameters in this section, you need to enable OSPF using the router command
and OSPF configuration sub-mode commands accessed in the Context configuration mode and documented
in the Context Configuration Mode chapter of this Command Line Interface Reference.
ip access-group
This command identities the access control list (ACL to be associated with this PVC interface in this context.
Important Prior to using this command, the access list must be created for this context with the ip access-list command
in the Context configuration mode and then the ACL must be configured using the commands described
in CLI chapter ACL Configuration Mode.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Indicates the specified access group to be removed from the access list.
name
Specifies the access control list (ACL) rule to be added or removed from the group.
name: Must be a string of 1 to 79 alphanumeric characters with no spaces.
Important Up to 8 ACLs can be applied to a group provided that the number of rules configured within the ACL(s)
does not exceed the 128 rule limit for the interface.
in | out
in: Specifies list is for in-bound access control.
out: Specifies the list is for out-bound access control.
Important Even though "in" or "out" can be specified, context-level ACL rules are automatically applied to both
directions.
Usage Guidelines Use this command to add IP access lists configured for the same context to an IP access-group. The list can
be configured to apply to all inbound and/or outbound traffic.
Examples The following adds ACL access-list-1 to the IP access-group associated with this PVC for this context.
ip access-group access-list-1 in
ip address
Defines the primary IP address and the network mask to be associated with this PVC interface for this context.
This command can also be used to configure the secondary IP address.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Removes the IP address information for this PVC from the configuration. It is not necessary to include the
subnet mask with the command.
The command must first be issued with the secondary IP address if one exists and then re-issued with the
primary IP address.
secondary
secondary: Including this keyword indicates the IP address and subnet mask being defined are to be used as
the secondary IP address for this PVC interface. This is referred to as multi-homing of the interface.
Usage Guidelines Configures or deletes the IPv4 or IPv6 addresses and subnet mask to be associated with this PVC.
Examples The following configures the secondary IP address to associate with the interface.
ip address 131.2.3.4 255.255.255.0 secondary
The following set of commands removes the primary IP address from the PVC interface configuration for
this context.
no ip address secondary address
no ip address primary address
ip mtu
Configures the maximum transmission unit (MTU) to be supported on this interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Disables and/or restores the option to the system default.
mtu value
Configures the maximum transmission unit in octets.
value: Enter an integer between 576 and 1600. Default is 1500.
ip ospf authentication-key
This command configures the password or key to be used for OSPF (Open Shortest Path First) authentication
with neighboring routers.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the authentication key.
encrypted
Enter this keyword if you are pasting a previously encrypted authentication key into the password auth_key
for this command.
passwordauth_key
auth_key is a string variable, from 1 through 16 alphanumeric characters, that denotes the authentication key
(password). This variable is entered in clear text format.
Usage Guidelines Use this command to set the authentication key used when authenticating with neighboring routers.
Examples To set the authentication key to 123abc, use the following command;
ip ospf authentication-key password 123abc
Use the following command to delete the authentication key;
no ip ospf authentication-key
ip ospf authentication-type
This command configures the OSPF authentication method to be used with OSPF neighbors over the logical
interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Disable this function.
message-digest
Set the OSPF authentication type to use the message digest (MD) authentication method.
null
Set the OSPF authentication type to use no authentication, thus disabling either MD or clear text methods.
text
Set the OSPF authentication type to use the clear text authentication method.
Usage Guidelines Use this command to set the type of authentication to use when authenticating with neighboring routers.
Examples To set the authentication type to use clear text, enter the following command;
ip ospf authentication-type text
ip ospf cost
This command configures the cost associated with sending a packet over this logical interface.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Disable this function.
value
Default: 10
The cost to assign to OSPF packets. This must be an integer from 1 through 65535.
Usage Guidelines Use this command to set the cost associated with routes from the interface.
ip ospf dead-interval
This command configures the dead-interval and the delay time in seconds, for OSPF communications.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the value set and returns the value to its default.
value
The interval, in seconds, that the router should wait. During this interval, if no packets are received then the
system considers the neighboring router to be off-line. This interval is typically 4 times the duration of the
hello-interval.
value must be an integer from 1 through 65535. Default: 40
Usage Guidelines Use this command to set the dead-intervals or delays for OSPF communications.
ip ospf hello-interval
This command configures the delay time in seconds, for OSPF hello interval.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the value set and returns the value to its default.
value
The interval, in seconds, between sending hello packets. This value is typically set to be 1/4 of the value of
the dead-interval.
value must be an integer from 1 through 65535. Default: 10
Usage Guidelines Use this command to set the delays for the hello-interval.
ip ospf message-digest-key
This command enables the use of MD5-based OSPF authentication.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the key.
message-digest-key key_id
Specifies the key identifier number. key_id must be an integer from 1 through 255.
encrypted
Use this if you are pasting a previously encrypted authentication key into the CLI command.
password authentication_key
The password to use for authentication. authentication_key is a string variable, from 1 through 16 alphanumeric
characters, that denotes the authentication password. This variable is entered in clear text format.
Usage Guidelines Use this command to create an authentication key that uses MD5-based OSPF authentication.
Examples To create a key with the ID of 25 and a password of 123abc, use the following command;
ip ospf message-digest-key 25 md5 password 123abc
To delete the same key, enter the following command;
no ip ospf message-digest-key 25
ip ospf network
Configures the OSPF network type.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Disable this function.
broadcast
Sets the network type to broadcast.
non-broadcast
Sets the network type to non-broadcast multi access (NBMA).
point-to-multipoint
Sets the network type to point-to-multipoint.
point-to-point
Sets the network type to point-to-point.
Usage Guidelines Use this command to specify the OSPF network type.
Examples To set the OSPF network type to broadcast, enter the following command;
ip ospf network broadcast
ip ospf priority
This command designates the OSPF router priority.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Disable this function.
value
The priority value to assign. This must be an integer from 0 through 255.
Usage Guidelines Use this command to set the OSPF router priority.
ip ospf retransmit-interval
This command configures the retransmit-interval and the delay time in seconds, for OSPF communications.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the value set and returns the value to its default.
value
The interval, in seconds, between LSA (Link State Advertisement) retransmissions.
value must be an integer from 1 through 65535. Default: 5
Usage Guidelines Use this command to set the retransmit-intervals or delays for OSPF communications.
ip ospf transmit-delay
This command configures the transmit-delay the OSPF communications parameters.
Product SGSN
Command Modes Exec > Global Configuration > Context Configuration > PVC Interface Configuration
configure > context context_name > interface interface_name point-to-point
Entering the above command sequence results in the following prompt:
[context_name]host_name(config-if-pvc)#
no
Deletes the value set and returns the value to its default.
transmit-delay value
The interval, in seconds, that the router should wait before transmitting a packet.
value must be an integer from 1 through 65535. Default: 1
Examples To set the transmit delay to 5 seconds, use the following command;
ip ospf transmit-delay 5
To delete the setting for the transmit-delay or reset the transmit-delay value to its default of 1, use the following
command'
no ip ospf transmit-delay
Command Modes Exec > Global Configuration > QCI-QoS Mapping Configuration
configure > qci-qos-mapping name
Entering the above command sequence results in the following prompt:
[local]host_name(config-qci-qos-mapping)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current configuration mode and returns to the Exec mode.
Product All
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
operator-defined-qci
Creates and maps non-standard QCI values to enforceable QoS parameters.
Product P-GW
SAEGW
Privilege Administrator
Command Modes Exec > Global Configuration > QCI-QoS Mapping Configuration
configure > qci-qos-mapping name
Entering the above command sequence results in the following prompt:
[local]host_name(config-qci-qos-mapping)#
Syntax Description operator-defined-qci num { gbr | non-gbr } [ { downlink | uplkink } [ encaps-header { copy-inner |
copy-outer | dscp-marking dscp-marking-value } [ internal-qos priority priority ] | internal-qos priority
priority | user-datagram dscp-marking dscp-marking-value [ encaps-header { copy-inner | copy-outer |
dscp-marking dscp-marking-value } [ internal-qos priority priority ] ] | pre-rel8-qos-mapping num ]
no operator-defined-qci num
no
Disables the selected non-standard QCI value.
num
Specifies the non-standard, operator-defined QCI value to be enabled. num must be an integer from 128
through 254.
Important Standards-based QCI values 1 through 9 are configured through the qci command.
gbr
Specifies that this QCI type is Guaranteed Bit Rate (GBR).
non-gbr
Specifies that this QCI type is non-Guaranteed Bit Rate (non-GBR).
downlink
Configures parameters for downlink traffic.
uplink
Configures parameters for uplink traffic.
pre-rel8-qos-mapping num
Maps non-standard QCI to a standard QCI that has the characteristics (TC, THP, SI, TD, SSD) similar to
desired pre-rel8 standard QoS values during 3G call or GnGp handover.
num must be an integer from:
• 1 through 4 for GBR
• 5 through 9 for non-GBR
Important If the wrong value is chosen, one of the following configuration errors will appear: "Failure: Only QCI
range 1 - 4 are allowed for GBR QCI" or "Failure: Only QCI range 5 - 9 are allowed for Non-GBR QCI".
QCI values 1 through 9 are defined in 3GPP Specification TS 23.203 "Policy and charging control architecture".
Usage Guidelines Use this command to create and map non-standard QCI values to enforceable QoS parameters in P-GW so
that calls can be accepted when non-standard QCI values are received from UE or PCRF.
Important Use of non-standard QCIs require that a valid license key be installed. Contact your Cisco Account or
Support representative for information on how to obtain a license.
3G GGSN Call
If the pre-rel8-qos-mapping field is not configured for the non-standard QCI under P-GW which is associated
with a GGSN, then the 3G call would be rejected.
GnGp Handoff
1 If the pre-rel8-qos-mapping field is not configured for the non-standard QCI for default bearer, then the
handoff would be rejected.
2 If the pre-rel8-qos-mapping field is not configured for the non-standard QCI for dedicated bearer, then
only that bearer would be rejected during handoff.
3 In the following scenario:
• default bearer with standard QCI or non-standard QCI (with pre-rel8-qos-mapping configured)
• more than one dedicated bearer (some with standard QCI, some with non-standard QCI with
pre-rel8-qos-mapping configured, and some with non-standard QCI with no mapping)
Examples The following command creates an operator-defined GBR QCI value of 129 and maps it to a pre-rel8 standard
QoS value of 2:
operator-defined-qci 129 gbr pre-rel8-qos-mapping 2
qci
Creates and maps standard QCI values to enforceable QoS parameters.
Product HSGW
P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > QCI-QoS Mapping Configuration
configure > qci-qos-mapping name
Entering the above command sequence results in the following prompt:
[local]host_name(config-qci-qos-mapping)#
Syntax Description qci num [ arp-priority-level arp_value ] [ downlink [ encaps-header { copy-inner | dscp-marking
dscp-marking-value } ] [ internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value
] ] [ uplink [ downlink] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [
internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ]
qci num [ delay-class delay-class-value precedence-class precedence-class-value reliability-class
reliability-class-value [ downlink [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ]
[ internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] [ uplink [
downlink ] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [ internal-qos priority
priority ] [ user-datagram dscp-marking dscp-marking-value ] ] ]
qci num [ downlink [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [ internal-qos
priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ]
qci num [ gbr [ delay-class delay-class-value precedence-class precedence-class-value reliability-class
reliability-class-value ] [ downlink [ encaps-header { copy-inner | dscp-marking dscp-marking-value
} ] [ internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] [
max-packet-delay max-packet-delay-value max-error-rate max-error-rate ] [ traffic-policing interval
value ] [ uplink [ downlink ] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [
internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] ]
qci num [ max-packet-delay max-packet-delay-value max-error-rate max-error-rate [ downlink [
encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [ internal-qos priority priority ] [
user-datagram dscp-marking dscp-marking-value ] ] [ uplink [ downlink ] [ encaps-header { copy-inner
| dscp-marking dscp-marking-value } ] [ internal-qos priority priority ] [ user-datagram dscp-marking
dscp-marking-value ] ] ]
qci num [ non-gbr [ delay-class delay-class-value precedence-class precedence-class-value reliability-class
reliability-class-value ] [ downlink [ encaps-header { copy-inner | dscp-marking dscp-marking-value
} ] [ internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] [
max-packet-delay max-packet-delay-value max-error-rate max-error-rate ] [ traffic-policing interval
value ] [ uplink [ downlink] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [
internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] ]
qci num [ traffic-policing interval interval [ delay-class delay-class-value precedence-class
precedence-class-value reliability-class reliability-class-value ] [ downlink [ encaps-header { copy-inner
| dscp-marking dscp-marking-value } ] [ internal-qos priority priority ] [ user-datagram dscp-marking
dscp-marking-value ] ] [ max-packet-delay max-packet-delay-value max-error-rate max-error-rate ]
[ uplink [ downlink] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [ internal-qos
priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ] ]
qci num [ uplink [ downlink] [ encaps-header { copy-inner | dscp-marking dscp-marking-value } ] [
internal-qos priority priority ] [ user-datagram dscp-marking dscp-marking-value ] ]
[ default | no ] qci num
Important The optional keywords associated with each of the initial optional keywords are abbreviated in the syntax
examples above for clarity. Refer to the definitions below for the full keyword paths and associated
descriptions for each keyword string in this command.
default
Resets the default values for the select QCI value.
no
Disables the selected QCI value.
num
Specifies the QCI value to be enabled; must be an integer between 1-9 or 128-254 .
QCI values 1 through 9 are defined in 3GPP Specification TS 23.203 "Policy and charging control architecture".
In release 21 and forward, QCI options 65 and 66 are available for guaranteed bit rate (GBR) network initiated
QCI values only.
In release 21 and forward, QCI options 69 and 70 are available for non-GBR network initiated QCI values
only.
downlink
Configures parameters for downlink traffic.
gbr
Specifies that this QCI type is Guaranteed Bit Rate (GBR).
Important Defaults for standards-based QCI values are defined in 3GPP Specification TS 23.203 "Policy and charging
control architecture".
non-gbr
Specifies that this QCI type is non-Guaranteed Bit Rate (non-GBR).
uplink
Configures parameters for uplink traffic.
Syntax Description qci num [ delay-class num precedence-class num reliability-class num [ downlink | uplink ] ]
qci num [ downlink [ 802.1p-value priority | encaps-header { copy-inner [ 802.1p-value priority ] |
dscp-marking hex [ 802.1p-value priority ] } | user-datagram dscp-marking hex [ 802.1p-value priority ]
| encaps-header { copy-inner [ 802.1p-value priority ] | dscp-marking hex | copy-outer [ 802.1p-value
priority ] } ] ]
qci num [ gbr [ delay-class | downlink | max-packet-delay | traffic-policing | uplink ] ]
qci num [ max-packet-delay num max-error-rate num [ downlink | uplink ] ]
qci num [ non-gbr [ delay-class | downlink | max-packet-delay | traffic-policing | uplink ] ]
qci num [ traffic-policing interval interval [ delay-class | downlink | max-packet-delay | uplink ] ]
qci num [ uplink [ 802.1p-value priority | encaps-header { copy-inner [ 802.1p-value priority ] |
dscp-marking hex | copy-outer[ 802.1p-value priority ] } | mpls-exp-value value [ downlink { 802.1p-value
priority | encaps-header { copy-inner [ 802.1p-value priority ] | dscp-marking hex| copy-outer } |
user-datagram dscp-marking hex [ 802.1p-value priority | encaps-header { copy-inner [ 802.1p-value
priority ] | dscp-marking hex [ 802.1p-value priority ] } ] ] ]| user-datagram dscp-marking hex [ 802.1p-value
priority ] | encaps-header { copy-inner [ 802.1p-value priority ] | dscp-marking | copy-outer hex [
802.1p-value priority ] } ] ]
[ default | no ] qci num
Important The optional keywords associated with each of the initial optional keywords are abbreviated in the syntax
examples above for clarity. Refer to the definitions below for the full keyword paths and associated
descriptions for each keyword string in this command.
default
Resets the default values for the select standards-based QCI value.
no
Disables the selected standards-based QCI value.
num
Specifies the standards-based QCI value to be enabled.
num must be an integer from 1 through 256.
QCI values 1 through 9 are defined in 3GPP Specification TS 23.203 "Policy and charging control architecture".
gbr
Specifies that this QCI type is Guaranteed Bit Rate (GBR).
Important Defaults for standards-based QCI values are defined in 3GPP Specification TS 23.203 "Policy and charging
control architecture".
non-gbr
Specifies that this QCI type is non-Guaranteed Bit Rate (non-GBR).
Usage Guidelines Use this command to create and map QCI values to enforceable QoS parameters.
Important Non-standard QCI values are only supported with the license-enabled operator-defined-qci command.
Examples The following command creates a QCI value of 8 and defines the uplink encapsulation header as using the
DSCP marking from the encapsulated UDP header:
qci 8 uplink encaps-header copy-inner
Command Modes Exec > Global Configuration > QCI - RAN ID Mapping Configuration
configure > profile-id-qci-mapping-table name
Entering the above command sequence results in the following prompt:
[local]host_name(config-profile-id-qci-mapping-table)#
Important The commands or keywords/variables that are available are dependent on platform type, product version,
and installed license(s).
end
Exits the current mode and returns to the Exec Mode.
Product All
exit
Exits the current mode and returns to the previous mode.
Product All
profile-id
Maps a QCI ID to a RAN profile ID and modifies data flow bit rate ranges.
Product HSGW
Privilege Administrator
Command Modes Exec > Global Configuration > QCI - RAN ID Mapping Configuration
configure > profile-id-qci-mapping-table name
Entering the above command sequence results in the following prompt:
[local]host_name(config-profile-id-qci-mapping-table)#
Syntax Description profile-id id qci num [ uplink { gbr rate [ mbr rate ] | mbr rate [ gbr rate ] } downlink { gbr rate [ mbr
rate ] | mbr rate [ gbr rate ] ]
no profile-id id
no
Removes the specified profile ID entry from this map.
id
Specifies the profile ID to which a QCI ID will be mapped. id must be an integer value from 1 to 65535.
qci num
Specifies the QCI number to which the profile ID will be mapped. num must be an integer value from 1 to
255.
uplink
Specifies that the guaranteed bit rate (GBR) and/or maximum bite rate (MBR) setting that follow this keyword
will be applied to the uplink data flow.
downlink
Specifies that the guaranteed bit rate (GBR) and/or maximum bite rate (MBR) settings that follow this keyword
will be applied to the downlink data flow.
gbr rate
Specifies the guaranteed bit rate for the uplink or downlink data flow. rate must be an integer value from 0
to 4294967295.
mbr rate
Specifies the maximum bit rate for the uplink or downlink data flow. rate must be an integer value from 0 to
4294967295.
Usage Guidelines Use this command to map a QCI ID to a RAN profile ID and, optionally, modify data flow bit rate ranges.
Examples The following command maps a QCI ID (1) to a profile ID (10) and sets the uplink guaranteed bite rate to
10000 and the downlink guaranteed bit rate to 20000:
profile-id 10 qci 1 uplink gbr 10000 downlink gbr 20000
Command Modes Exec > Global Configuration > QoS L2 Mapping Configuration
configure > qos-l2-mapping
Enter the above command sequence results in the following prompt:
[local] host_name (config-qos-l2-mapping)#
The commands or keywords/variables that are available are dependent on platform type, product version
and installed license(s).
internal-priority
Maps internal QoS priority with Class of Service (COS) values
Product ePDG
HSGW
P-GW
SAEGW
S-GW
Privilege Administrator
Command Modes Exec > Global Configuration > Qos L2 Mapping Configuration
configure > qos l2-mapping-table { name map_table_name | system-default }
Entering the above command sequence results in the following prompt:
[local] host_name (config-qos-l2-mapping)#
Syntax Description internal-priority cos class_of_service_value color color_value [ 802.1p-value 802.1p_value ] [ mpls-tc
mpls_tc_value ]
default internal-priority cos cos_value color color_value
default
Restores default value assigned for specified parameter.
cos class_of_service_value
Maps to the internal QoS priority/COS.
class_of_service_value must be a Hexadecimal number between 0x0 and 0x7.
color color_value
Controls drop precedence of service to map to.
color_value must be a Hexadecimal number between 0x0 and 0x3.
802.1p-value 802.1p_value
Map to a 802.1p value. This also includes both P-bits and DEI/CFI. DEI is the lsb bit.
Caution Setting an odd value (DEI/CFI to 1) makes some switches drop packets.
mpls-tc mpls_tc_value
Map to an MPLS traffic class.
mpls_tc_value must be a Hexadecimal number between 0x0 and 0x7.
Usage Guidelines This command is used to map internal QoS priority with COS values.
Important The internal-priority CLI command also offers the ability to configure both 802.1p priority and setting
of DEI/CFI bit. This flexibility installation will treat the bit as DEI (drop eligibility indicator). However,
for installations that treat the bit as CFI (canonical format indicator), this should be set to 0. Otherwise,
the packet will be dropped.
Examples This command is used to map internal QoS priority with COS values:
internal-priority cos 0x2 color 0x1
The QoS Profile Configuration mode is used to create and configure a QoS Profile.
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
apn-ambr
Configures the APN-AMBR (aggregate maximum bit rate) that will be stored in the Home Subscriber Server
(HSS).
Product MME
SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
Syntax Description apn-ambr max-ul mbr-up max-dl mbr-dwn { pgw-upgrade | prefer-as-cap } { local | minimum |
rej-if-exceed }
remove apn-ambr
remove
Removes the APN-AMBR changes from the configuration for this APN profile.
prefer-as-cap
This keyword configures the capping that is applied on the subscription value received from the HSS or the
value received from the peer-node (MME/S4-SGSN) during inbound relocation. One of the following actions
must be configured under prefer-as-cap -- Note that the resulting value is used for the QoS parameter and
sent in the Create Session Request or the Modify Bearer Command (in case of HSS-initiated QoS/APN-AMBR
modification) message:
• local - The configured local value will be used.
• minimum - The minimum (lowest) value of the configured local value or the HSS-provided value will
be used.
• reject-if-exceed - The request/procedure is rejected if the HSS-provided value exceeds the configured
local value.
pgw-upgrade
MME only.
This keyword configures the QoS capping to be applied on the values received from the PGW during Attach
/ PDN-connectivity / Bearer-creation / Bearer-modification procedures. One of the following actions must be
configured under pgw-upgrade -- Note that the resulting value is used for the QoS parameter and sent to the
UE:
• local - The configured local value will be used.
• minimum - The minimum (lowest) value of the configured local value or the PGW-provided value will
be used.
• reject-if-exceed - The request/procedure is rejected if the PGW-provided value exceeds the configured
local value.
Usage Guidelines Use this command to define the MBR that will be enforced by the P-GW for both uplink and downlink traffic
shaping.
For the MME, use the apn-ambr command to set local values QoS capping type to be applied for the
APN-AMBR received from HSS/PGW/peer-node. One or both prefer-as-cap and/or pgw-upgrade must be
configured to override the default behavior, which is to accept the received value from the HSS/peer-node/PGW.
Examples A command similar to the following sets the APN-AMBR maximum uplink and maximum downlink bit rates
for the QoS profile:
qos apn-ambr max-ul 24234222 max-dl 23423423
A command similar to the following sets the prefer-as-cap minimumcapping action to be taken when the
SGSN or MME receives outside of APN-AMBR maximum uplink and maximum downlink bit rates :
qos apn-ambr max-ul 24234222 max-dl 23423423 prefer-as-cap minimum
associate
This command associates a specific bearer control profile with this QoS profile.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Add this command prefix to the command to delete the association between the QoS profile and the identified
bearer-control-profile.
bc_profile_name
Identifies the name of the bearer control profile being associated with the QoS profile. Enter 1 to 64
alphanumeric characters.
Usage Guidelines Use the associate command in Quality of Service Profile configuration mode to associate the bearer control
profile with the QoS profile and map a specific QCI or a range of QCI to the bearer control profile being
associated with the QoS profile.
A specific QCI cannot be associated to more than one bearer control profile. The QCI of the bearer is used to
identify the applicable bearer control profile.
• For dedicated bearers, the QCI of bearer is initially determined by the QCI value received from PGW
during dedicated bearer activation or the value received from peer MME/S4-SGSN.
• For default bearers, the QCI of bearer is initially determined by the subscription from HSS or the value
received from peer MME/S4-SGSN during inbound relocation.
Examples The following sample command associates the BCprof1 bearer control profile with the QoS profile and maps
QCI 7 to this bearer control profile:
associate bearer-control-profile BCprof1 qci 7
class
Configures local values for the traffic class (TC) parameters for the quality of service (QoS) configured for
this QoS profile.
Important To enable any of the values/features configured with this command, the prefer-as-cap configuration (also
in the QoS profile configuration mode) must be set to either local or both-hlr-and-local.
Product SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes previously defined values for the specified option or for an entire class if a qualifying option is not
included in the command.
background
Selects the background traffic class. This 'best-effort' class manages traffic that is handled as a background
function, like email, where time to delivery is not a key factor. The selection of background traffic class can
be refined with the addition of one of the following qualifying options:
• all-values
• arp
• evolved-arp
• mbr-down
• mbr-map-down
• mbr-map-up
• mbr-up
• residual-bit-error-rate
• sdu
conversational
Selects the 'real-time' conversational traffic class of service, which has the most stringent time requirements
of the four classes and is typically reserved for voice traffic. The section of the conversational traffic class
can be refined with the addition of one of the following qualifying options:
• all-values
• arp
• evolved-arp
• gbr-down
• gbr-up
• mbr-down
• mbr-map-down
• mbr-map-up
• mbr-up
• min-transfer-delay
• residual-bit-error-rate
• sdu
interactive
Selects interactive traffic class of service. This class is characterized by a request/response pattern (someone
sends data and then waits for a response) which requires the preservation of the data but delivers on a 'best-effort'
model. The section of the interactive traffic class can be refined with the addition of one of the following
qualifying options:
• all-values
• arp
• evolved-arp
• mbr-down
• mbr-map-down
• mbr-map-up
• mbr-up
• residual-bit-error-rate
• sdu
• thp
streaming
Selects the streaming traffic class of service, which handles one-way, real-time data transmission - such as
streaming video or audio. The section of the interactive traffic class can be refined with the addition of one
of the following qualifying options:
• all-values
• arp
• evolved-arp
• gbr-down
• gbr-up
• mbr-down
• mbr-map-down
• mbr-map-up
• mbr-up
• min-transfer-delay
• residual-bit-error-rate
• sdu
qualif_option
Qualifying options are the QoS parameters and they include:
• all-values - This option will change the configuration to predefined values for all the relevant QoS
parameters for the class. This keyword is not used if other options are to be defined. The predefined
values are:
Allocation/Retention Priority 3
• gbr-down - Guaranteed Kbps rate for the downlink direction. Enter an integer from the range 1 to
256000.
• gbr-up - Guaranteed Kbps rate for the uplink direction. Enter an integer from 1 to 256000.
• mbr-down - Maximum Kbps rate for the downlink direction. Enter an integer from the range 1 to 256000.
• mbr-map-down from from_kbps to to_kbps - Map received HLR MBR (from value) to a locally
configured downlink MBR value (to value):
◦from_kbps - Enter an integer from 1 to 25600.
◦to_kbps - Enter an integer from 1 to 25600.
• mbr-map-up from from_kbps to to_kbps - Map received HLR MBR (from value) to a locally configured
uplink MBR value (to value):
◦from_kbps - Enter an integer from 1 to 25600.
◦to_kbps - Enter an integer from 1 to 25600.
• mbr-up - Maximum Kbps rate for the uplink direction. Enter an integer from 1 to 256000.
◦Conversational TC residual-bit-error-rate range is from 5*10^-2 to 10^-6. Enter one of the following
integers, where:
◦1: represents 5*10^-2
◦2: represents 10^-2
◦3: represents 5*10^-3
◦5: represents 10^-3
◦6: represents 10^-4
◦7: represents 10^-5
◦8: represents 10^-6
◦Interactive TC residual-bit-error-rate range is from 4*10^-4 to 6*10^-8. Enter one of the following
integers, where:
◦4: represents 4*10^-3
◦7: represents 10^-5
◦9: represents 6*10^-8
◦Streaming TC residual-bit-error-rate range is from 5*10^-2 to 10^-6. Enter one of the following
integers, where:
◦1: represents 5*10-2
◦2: represents 10^-2
◦3: represents 5*10^-3
◦5: represents 10^-3
◦6: represents 10^-4
◦7: represents 10^-5
◦8: represents 10^-6
• sdu - Signalling data unit keyword, must include one of the following options:
• delivery-order- Enter one of the two following options:
• error-ratio- The SDU error-ratio range is from 10^-3 to 10^-6. Enter an integer from 1 to 6, where:
◦3- Represents 10^-3
◦4- Represents 10^-4
◦6- Represents 10^-6
• max-size- Defines the maximum number of octets (size) of the SDU. Enter an integer from 10 to
1502.
Usage Guidelines This command defines the qualifying options (parameters) for each QoS traffic class defined for this QoS
profile.
Repeat the command as often as needed with different options to define all required QoS criteria. For example,
to configure the maximum bit rate (MBR) for the downlink and uplink directions for a traffic class, this
command must be used twice, specifying mbr-down once and mbr-up once.
Advantage for local mapping of MBR: some HLRs cannot be configured with high MBR values. Using the
mbr-map-up and the mbr-map-down parameters allows the SGSN to be configured to treat a specific HLR
value as meaning the desired high MBR value. In a case where the HLR does not support HSPA+ bit rates,
but the handsets and network do, this feature allows the operator to overcome limitations on the HLR and
provide HSPA+ bit rates by overwriting the provisioned HLR-QoS MBR values with SGSN-configured
values. When MBR mapping is configured, if QoS is preferred as the HLR value, then the subscription QoS
MBR received from the HLR is compared with the "from" value in the table. If it matches, then it is converted
to the value specified by the "to" value in the table. QoS negotiation happens based on the converted value.
Advantage for QoS capping with THP and ARP: Controlling THP and ARP via Operator Policy: This
functionality can differentiate home vs. roaming subscribers, and prevent visiting subscribers from receiving
a high-tiered service. For example, a service provider could offer service differentiation using
Ultra/Super/Standard service levels based upon QoS; this could justify charging a corporate customer more
to use the Internet APN than would be charged to a consumer. This could be accomplished by controlling the
traffic handling priority (THP) over the air interface, i.e. THP 1 = Ultra, THP 2 = Super and THP 3 = Standard.
Examples Use the following command to configure the entire conversational traffic class with predefined QoS options:
class conversational all-values
description
Defines a descriptive string relevant to the specific QoS profile.
Product SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes the configured description from this QoS profile.
description
Specifies a description for this QoS profile as an alphanumeric string of 1 through 100 characters. The string
may include spaces, punctuation, and case-sensitive letters if the string is enclosed in double quotation marks
(").
Usage Guidelines Define information that identifies this particular QoS profile.
Examples Indicate that QoS profile qosprof1 is to be used for customers in India and that the profile was created on
April 10th of 2014:
description "qosprof1 defines QoS for customers in India (4/10/14)."
end
Exits the current configuration mode and returns to the Exec mode.
Product All
epc-qos-params-in-gtpv1
This command enables or disables the SGSN to send EPC QoS parameters to the GGSN.
Product SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes previous configuration changes and resets the default.
eps-subscription
If the keyword eps-subscription is configured, the EPC QoS parameters from EPS subscription are sent to
the GGSN. (Note: This option is not supported in this release).
gprs-subscription
If the keyword gprs-subscription is configured, E-ARP and APN-AMBR from the GPRS subscription are
sent. The UE-AMBR value is read from the user (local capping).
Usage Guidelines This command is disabled by default. On enabling this command E-ARP and APN-AMBR parameters are
included in the GTPV1 SM messages towards the GGSN.
Examples The following command enables the SGSN to send EPC QoS parameters to the GGSN. The E-ARP and
APN-AMBR values are picked from the GPRS subscription and the UE-AMBR value is read from the user
(local capping).
epc-qos-params-in-gtpv1 gprs-subscription
exit
Exits the current mode and returns to the parent configuration mode.
Product All
Usage Guidelines Use this command to return to the parent configuration mode.
operator-defined-qci
This command enables Operator Specific QCI in MME. If this command is enabled, MME accepts the QCI
range 128 - 254 from HSS and P-GW.
Product MME
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes the operator defined QCI configuration from the QoS profile.
Usage Guidelines The non-standard QCIs provides Operator Specific QoS for M2M and other mission critical communications.
In order to use operator specific QCIs, the QoS parameters need to be configured using the
pre-rel8-qos-mapping command. On configuring this command, the operator defined QCI values are mapped
to the PreRelease8QoS parameters during to UTRAN/GERAN.
prefer-as-cap
This command instructs the SGSN to choose the QoS configuration as the "qos parameters" for session
establishment.
Product SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
both-subscription-and-local
This keyword instructs the SGSN to use, as the capping value during session establishment, the lower of either
the locally configured QoS bit rate or the subscription received from HLR/HSS.
subscription
Instructs the SGSN to take QoS parameters from the subscription received from HLR (or HSS) and use the
same as the capping value for session establishment.
Default for SGSN.
local
Instructs the SGSN to take QoS parameters from the local configuration and use it for session establishment.
Usage Guidelines Use this command to instruct the SGSN to choose the QoS configuration for sessionestablishment.
Examples The following command instructs the SGSN to cap the bit rate with the lower rate of the two configurations,
subscription or local:
prefer-as-cap both-subscription-and-local
prefer-tc
Use this command to instruct which traffic class to use. This command overrides the traffic class received
from subscription.
Product SGSN
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes previous configuration changes and resets the default.
background
Use this keyword to use the background traffic class.
conversational
Use this keyword to use the conversational traffic class.
streaming
Use this keyword to use the streaming traffic class.
interactive
Use this keyword to use the interactive traffic class.
Usage Guidelines Use this command to instruct which traffic class to use. This command is applicable only if following is
configured, or the configuration will be ignored during call processing:
• The prefer-as-cap is set to local or both-subscription-and-local.
• The Traffic class configured in prefer-tc should be configured. For example, if prefer-tc is configured
as background then background class under QoS should also be configured.
Examples The following command is used to choose the background traffic class as the preferred traffic class:
prefer-tc background
qci-when-missing-in-subscription
This command is used to assign a default QCI value when a QCI value is not received from the subscription.
Product MME
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove
Removes the default QCI value configuration from the QoS profile.
qci_value
The qci_value variable in this configuration is considered as a default QCI value. The QCI value accepted is
either a Standard QCI value or Operator Specific value. The Standard QCI values range from 1 to 9, and new
Standard QCI values - 65, 66, 69 and 70. The Operator Specific values range from 128 to 254. The configuration
does not accept any other value apart from the ones mentioned above.
Usage Guidelines Use this command to configure a default QCI to avoid rejection during handovers to UTRAN/GERAN by
MME when a QCI value is not received from the Subscription.
qci-reject
Use this command to identify a specific QCI or a range of QCI for which the MME must reject bearer
establishment or modification.
Product MME
Privilege Administrator
Command Modes Exec > Global Configuration > Quality of Service Profile Configuration
configure > quality-of-service-profile
Entering the above command sequence results in the following prompt:
[local]host_name(qos-of-service-profile)#
remove qci-reject
Deletes this configuration from the QoS profile configuration.
dedicated-bearer
Identifies either a specific QoS class identifier (QCI) or a range of QCI for the dedicated-bearer.
default-bearer
Identifies either a specific QoS class identifier (QCI) or a range of QCI for the default-bearer.
qci_value
Identifies a specific QCI value.
• For dedicated-bearers, enter an integer from 1 through 9.
• For default-bearers, enter an integer from 5 through 9.
to end_qci_value
Type "to" and then enter an integer for the QCI value to end the range.
• For dedicated-bearers, enter an integer from 2 through 9 that is greater than the QCI value entered for
the beginning of the range.
• For default-bearers, enter an integer from 6 through 9 that is greater than the QCI value entered for the
beginning of the range.
Usage Guidelines The MME can reject default-bearers and dedicated-bearers based on QCI received from the subscription or
the peer-MME/S4-SGSN during inbound relocation or the Create Session Response/Update Bearer
Request/Create Bearer Request procedure.
Examples The following is a sample command that illustrates the configuration for the MME to reject bearer establishment
for the dedicated-bearer channel if QCI 7 is received from the P-GW:
qci-reject
dedicated-bearer 7