HMC Docs
HMC Docs
HMC Docs
What Is an HMC?
You will use the HMC to manage the configuration and operation of partitions in a system, as well
as add and remove hardware without interrupting system operation. With an HMC, you can
control Capacity on Demand resources. One HMC is capable of controlling multiple servers. The
systems that are monitored by the HMC are called managed systems.
The HMC provides a graphical user interface (GUI) for configuring and operating single or
multiple managed systems. There is also a command line interface (CLI).
The HMC may connect locally to the systems it manages via a private network or connect
remotely via an open network. Using the Web-based System Manager Remote Client, you may
connect to the HMC from your PC.
In this course, we are concerned with the HMC as it relates to the IBM POWER5 Servers. There
are five HMC models available for use with POWER5 servers. They are the 7310-C03, 7310-C04,
and 7310-C05 desktop models and the 7310-CR2 and 7310-CR3 rack-mounted models.
The HMC uses its connections to managed systems to perform various functions, including:
The IBM 7310-C05 Desktop Hardware Management Console (HMC) is a dedicated workstation
that allows you to configure and manage partitions and Capacity on Demand on POWER5
servers. An integrated hardware management application helps you configure and partition the
server through a GUI. In addition, operating system console support is provided for i5/OS using a
5250 programming interface. This can potentially save the cost of a separate IBM eServer i5
console.
• Create and store LPAR profiles that define the processor, memory, and I/O resources
allocated to an individual partition
• Start, stop, and reset a system partition
• Boot a partition or system by selecting a profile
• Display system and partition status
• Display a virtual operator panel of the contents for each partition or controlled system
The IBM eServer p5 590 and 595 and eServer i5 595 must have access to an HMC. An HMC is
required to manage any POWER5 server when Capacity Upgrade on Demand is implemented.
The HMC offers a service focal point for the systems it controls. It is connected to a dedicated
port on the service processor of the POWER5 system via an Ethernet connection. The Ethernet
connections from HMC to eServer p5 590 and 595 and eServer i5 595 must be via private LAN,
whereas other POWER5 systems can be either private or public. LAN Tools are available for
determining problems and providing service support, such as call-home and error log notification,
through a modem or the Internet.
Multiple POWER5 processor-based servers can be supported by one HMC, located locally or
remotely attached.
IBM 7310-CR3 Rack-Mounted HMC
• Create and store LPAR profiles that define the processor, memory, and I/O resources
allocated to an individual partition
• Start, stop, and reset a system partition
• Boot a partition or system by selecting a profile
• Display system and partition status
• Display a virtual operator panel of the contents for each partition or controlled system
The IBM eServer p5 590 and 595 and eServer i5 595 must have access to an HMC. An HMC is
required to manage any POWER5 server when Capacity Upgrade on Demand is implemented.
The HMC offers a service focal point for the systems it controls. It is connected to a dedicated
port on the service processor of the POWER5 system via an Ethernet connection. The Ethernet
connections from HMC to eServer p5 590 and 595 and eServer i5 595 must be via private LAN,
whereas other POWER5 systems can be either private or public. LAN Tools are available for
determining problems and providing service support, such as call-home and error log notification,
through a modem or the Internet.
Multiple POWER5 processor-based servers can be supported by one HMC, located locally or
remotely attached.
Hardware Management Console Models for POWER5 Servers
Desktop HMC
Model POWER5 Models Supported POWER5 POWER4
7310_C03 520, 550, 570, 575, 590, 595, 720
7310_C04 520, 550, 570, 575, 590, 595, 720
7310_C05 520, 550, 570, 575, 590, 595, 720
Rack-Mounted HMC
Model POWER5 Models Supported POWER5 POWER4
7310_CR2 520, 550, 570, 575, 590, 595, 720
Shipped with POWER4 code. May be migrated to
7315_CR2
support POWER5 systems.
7310_CR3 520, 550, 570, 575, 590, 595, 720
The HMC models currently being shipped for POWER5 servers are the desktop model 7310_C04
and the rack-mounted model 7310_CR3. Earlier HMC models shipped with POWER4 code may
be migrated to support POWER5 systems. However, it is not possible to connect an HMC to both
a POWER4 and a POWER5 system simultaneously.
The HMC GUI at a Glance
The various elements of the HMC's graphical user interface (GUI) are covered in depth when
discussing the relevant tasks that the HMC can accomplish. For now, though, we'll provide a
quick overview.
Navigation Area
The left side of the HMC GUI is the Navigation area. It displays a hierarchy of items
ordered in a tree structure. The root of the tree is the Management Environment, which is
a set of host systems that can be managed from the HMC.
Contents Area
The right side of the panel is the Contents area. It displays managed objects and related
tasks. You can choose different views in the Contents area: large icons, small icons, or
details in the form of a list.
In addition to the two screen areas, the HMC GUI has these elements:
Menu Bar
The menu bar, located at the top of the screen, contains these items:
Tool Bar
The tool bar, located directly below the menu bar, has icons for commonly-used actions
such as powering on/off a managed system, activating a partition, or viewing the
properties of a system or partition.
Status Bar
The status bar, at the bottom of the screen, displays HMC status information.
An Overview of Logical Partitioning
The HMC allows you to perform many hardware management tasks for your managed system.
You can choose to operate the managed system as a single server or run multiple partitions.
Partitioning allows you to configure a single computer into several independent systems. Each of
these systems, called partitions, runs its own independent operating system and is capable of
running applications specific to that operating system in its own independent environment. This
independent environment contains its own operating system, its own set of system processors, its
own set of system memory, and its own I/O adapters.
A profile defines a configuration setup for a managed system or partition. The HMC allows you to
create multiple profiles for each managed system or partition. You can then use these profiles to
start a managed system or partition in a particular configuration.
A system profile is a collection of often-used partition profiles. You can use a system profile to
start an ordered list of pre-defined partition profiles on your managed system.
To configure and manage logical partitions on your IBM eServer i5 or eServer p5, you must have
at least one HMC.
Logical Partitioning
Resources and partitioning go hand-in-hand. Resources are a system's processors, memory, and
I/O slots. A logical partition uses software and firmware to logically partition the resources on a
system. I/O slots can be populated by different adapters, such as Ethernet, SCSI, or other device
controllers. A disk (both internal and external) is allocated to a partition by assigning it the I/O slot
that contains the disk's adapter.
Logical partitioning (LPAR) is only limited by the total number of hardware resources in the
system. For example, a partition could have any number of installed processors assigned to it,
limited only by the total number of installed processors. Similarly, a partition could have any
amount of memory, limited only by the total amount of memory installed (minus the memory
required for partition management/overhead). I/O adapters are physically installed in one of many
I/O drawers in the system. However, with logical partitioning, any I/O adapter in any I/O drawer
can be assigned to any partition; however, only one partition with that resource can be active.
Virtual I/O devices provide for sharing of physical resources, such as adapters and devices,
among partitions. Multiple logical partitions can share physical I/O resources of a system, and
each partition can simultaneously use virtual and physical I/O devices. Also, virtual I/O devices
allow partitions to be created without adding physical I/O adapters to the system.
The LPAR Validation Tool (LVT) is available to assist you in the design of LPAR systems. The
LVT emulates an LPAR configuration and validates that planned partitions are valid. In addition,
the LVT allows you to test the placement of AIX®, Linux®, and OS/400 hardware within the
system to ensure that the placement is valid. You can access the LVT at
http://www.ibm.com/servers/eserver/iseries/lpar/systemdesign.htm
Dynamic LPAR
Dynamic LPAR (DLPAR) allows the "dynamic" addition, movement (relocation of resources
between LPARs), or removal of resources without having to reactivate the partition or "power
down" the operating system. Consequently, customers utilizing DLPAR do not experience an
interruption in service.
In a static LPAR configuration, individual processors, 256 MB memory blocks, and I/O adapter
slot resources are placed under the exclusive control of a given logical partition. One of the main
advantages of the LPAR implementation is that it gives fine-grained allocation control over these
individual resources, allowing them to be combined in almost any quantity and combination to
create a logical partition.
DLPAR extends these capabilities by allowing this fine-grained resource allocation to occur not
only when activating a logical partition, but also while the partitions are running. Individual
processors, memory blocks, and I/O adapter slots can be released into a "free pool," acquired
from that free pool, or moved directly from one partition to another--again, in almost any quantity
or combination depending on the total hardware resources available.
If you have partitions that need more or can use fewer resources, you can dynamically move the
resources between partitions within the managed system. A "move" operation is simply a
combined operation removing a resource from one LPAR and adding it to another.
Partition Profiles
A partition does not actually own any resources until it is activated; resource specifications are
stored within partition profiles. The same partition can operate using different resources at
different times, depending on the profile you activate.
Each logical partition has at least one partition profile. You can have more than one profile for a
partition. However, you can only activate a partition with one of its profiles at a time. One profile is
designated as the default profile. It is this profile that is activated if another is not specified.
The screen at left, for example, shows three running partitions, endsqd01, endsqd02, and
endsqd04. The partition endsqd01has two partition profiles, endsqd01_aix530_normal, which
boots in normal mode, and endsqd01_aix530_sms, which boots in System Management Services
mode. You are given the option at partition activation to choose the profile with which you would
like to start the partition.
When you activate a partition, you enable the system to create a partition using the set of
resources in a profile created for that partition. The system will attempt to allocate the resources
you assigned to the profile. If you have over-committed resources, the partition profile will not be
activated.
Partition profiles are not affected by changes you make using the DLPAR feature. If you want
permanent changes, you must then reconfigure partition profiles manually. For example, if your
partition profile specifies that you require two processors and you use DLPAR to add a processor
to that partition, you must change the partition profile if you want the additional processor to be
added to the partition the next time you use the profile.
Manufacturing Default Configuration
The initial partition setup of your managed system as received from your service provider displays
one logical partition with one partition profile. The name of this partition is the serial number of the
system. The name of the profile is Default.
A partition profile in the manufacturing default configuration has all of your managed system’s
resources. If you desire, you can install an operating system on this partition and use this partition
as the only partition on the managed system. Because all of the hardware (both required and
desired) are assigned to this partition, no other partitions can be started when this partition profile
is running. Likewise, a partition profile in the manufacturing default configuration cannot be
started while other partitions are running.
System Profiles
Using the HMC, you can create and activate often-used collections of predefined partition
profiles. A collection of predefined partition profiles is called a system profile. The system profile
is an ordered list of partitions and the profile that is to be activated for each partition. The first
profile in the list is activated first, followed by the second profile in the list, followed by the third,
and so on.
Using the same example at left, a system profile could be created to start partitions endsqd01,
endsqd02 and endsqd04 using the desired partition profile.
The system profile helps you change the managed systems from one complete set of partition
configurations to another. For example, a company might want to switch from using 12 partitions
to using only 4 every day. To do this, the system administrator deactivates the 12 partitions and
activates a different system profile, one specifying 4 partitions. The advantage of the System
Profiles is in the initial IPL of the system. By starting up in System Profile mode, all of the
partitions defined there will be started, whereas by starting in LPAR standby mode, the system
administrator would then have to start each LPAR (and select the required profile for the LPAR)
manually.
Depending upon which POWER5 model you have, the HMC may be installed by you or by your
service representative. Before you begin connecting cables to the HMC, make sure that the HMC
is in a location where all necessary power outlets and network connections can safely be
reached. Also, maintain at least 51 mm (2 inches) of space on the sides of the system unit and
152 mm (6 inches) at the rear of the system unit to allow the system unit to cool properly.
To connect the HMC cabling for a desktop HMC, refer to the illustration at left as you complete
the following steps. Note that the actual HMC rear-panel connector layout may differ somewhat
from the one presented here.
1. Connect the keyboard cable to a USB plug location 5 at the rear of the system unit.
2. Connect the mouse cable to a USB plug location 5 at the rear of the system unit.
3. Connect the display monitor signal cable to plug location 4 at the rear of the system unit.
4. Attach the power cord to the monitor. Do not plug the power cords into the electrical
outlet at this point.
5. Connect the Ethernet cable to plug location 8 at the rear of the system unit.
6. Connect the external modem to the modem connector (serial connection) location 9.
7. Verify that the PC Line Voltage input switch P2 is correctly set to reflect the voltage
present at the installation site.
8. Attach the HMC's power cord to plug location P1 at the rear of the system unit.
1. Connect the unattached end of the Ethernet cable to the Ethernet port HMC1 on your
managed system. Note: Be careful not to connect the managed system to a power
source at this time.
2. For a desktop HMC, plug the monitor's power cord into an electrical outlet.
3. Plug the HMC's power cord into an electrical outlet.
4. For an external modem, connect the telephone set to the modem connector with the
picture of the telephone receiver.
5. Power on the display, then power on the HMC. This gives the HMC time to detect (via
DDC) the display characteristics.
After completing your cable connections, but before you begin the software installation process,
review the checklist in the eServer Information Center topic 'Initial Server Setup'. Once you
have verified that all necessary connections have been made correctly, you will be ready to
proceed.
Flexible Service Processor (FSP) Configuration
Flexible Service Processor (FSP) Configuration
Most of the communication from the HMC to an eServer i5 or eServer p5 server will be done
through the Flexible Service Processor (FSP).
The picture at top left is a rear view of an eServer model 520 server. The picture at bottom left is
a rear view of an eServer model 570 server. The HMC is connected to a dedicated port on the
service processor of the eServer i5 and eServer p5 server via an Ethernet connection. You
completed this connection in the previous topic Completing the Hardware Installation when you
connected the ethernet cable on your HMC to HMC1 on the managed system.
Follow the installation instruction for your eServer i5 or eServer p5 server. Once the HMC has
been cabled and connected to the managed system, follow the instructions in topic 8.0 HMC
Configuration to install and configure the HMC software. Do not yet connect the managed system
to a power source.
After the HMC software has been installed and configured, connect the managed system to a
power source. The HMC will automatically 'discover' the managed system and it will appear in the
Content area of the HMC's GUI interface under Server and Partition-->Server Management.
This feature is available only when the service processor is in standby mode. To set the initial
time of day on your server through ASM:
You can use the HMC to shut down or restart the the operating system on your partition. Because
this procedure may corrupt data on the partition you want to reset, perform this procedure only
after you have attempted to restart the operating system manually.
For i5/OS (OS/400) logical partitions, only use this option if you cannot shut down or restart the
i5/OS logical partition from the command line of the operating system. Using this procedure to
shut down an i5/OS logical partition will result in an abnormal IPL.
From the Server Management application, expand the desired managed system to display
partitions on the system.
Highlight the desired partition and right click. Choose Shut Down Partition or Restart Partition
as you wish.
Follow the directions on the screen displayed to shut down or restart the partition in the desired
way.
Choosing the Operating System option to shutdown or restart the logical partition normally is the
equivalent of a soft reset. The actions of the operating system after a soft reset are determined by
its policy settings. Depending on how the settings were configured, the operating system may
perform a dump of system information or restart automatically.
Choosing the Immediate option to shutdown the logical partition immediately is the equivalent of
a hard reset. A hard reset is equivalent to powering off the system. It forces termination and can
corrupt information. Use this option only when the operating system is disrupted and cannot send
or receive commands.
You may also shutdown and re-activate a partition in one action by Restarting a partition. You
have the same options for an Operating System or Immediate shutdown, and the same caveat
applies - only use this option on an i5/OS partition if you cannot shut down or restart the partition
from the command line of the operating system.
After choosing your shutdown or restart options, a status panel will show the progress of the
shutdown. The partition is shutdown when the status shows Finished. Success. as shown below
left.
Capacity on Demand (CoD) allows you to dynamically activate one or more resources on your
server as your business peaks dictate. You can activate inactive processors or memory units that
are already installed on your server on a temporary or permanent basis.
Information on preparing for and working with Capacity on Demand may be found in the
Information Center topic Working with Capacity on Demand.
Once you have set up your environment to take advantage of Capacity on Demand, you can use
your HMC to:
Capacity Upgrade on Demand (CUoD) offers you the capability to permanently activate one or
more inactive processors or memory units without requiring you to restart your server or interrupt
your business.
When you have purchased one or more activation features, you will receive one or more
activation codes to permanently activate your inactive processors or memory units.
To permanently activate your inactive processors or memory units:
1. In the Navigation area, open the Server and Partition folder. Click the Server
Management icon.
2. In the Content area, right-click the desired server and click Manage On Demand
Activations and Capacity on Demand.
3. Click Enter CoD Code and enter the activation code you received.
For more information on planning for and setting up Capacity Upgrade on Demand on your
eServer i5 or eServer p5 server, go to the eServer Information topic Working with Capacity
Upgrade on Demand.
On/Off Capacity on Demand allows you to temporarily activate and deactivate processors and
memory units to satisfy business peaks. Once you request a number of processors or memory
units to be made temporarily available for a specified number of days, those processors and
memory units are available immediately. You can start and stop requests for On/Off Capacity on
Demand, and you are billed for usage at the end of each quarter.
1. In the Navigation area, open the Server and Partition folder. Click the Server
Management icon.
2. In the Content area, right-click the desired server and click Manage On Demand
Activations and Capacity on Demand.
3. Click Processor and Manage On/Off CoD.
The Manage On/Off CoD Processors screen is displayed. From this screen you may determine
and adjust:
For more information on planning for and setting up On/Off Capacity on Demand, see the
eServer Information Center topic Working with Capacity on Demand.
Reserve Capacity on Demand
Reserve Capacity on Demand allows you to purchase a reserve capacity prepaid feature that
represents a number of processor days. You can then activate the inactive processors using
Reserve Capacity on Demand as your business requires.
1. In the Navigation area, open the Server and Partition folder. Click the Server
Management icon.
2. In the Content area, right-click the desired server and click Manage On Demand
Activations and Capacity on Demand.
3. Click Processor and Manage Reserve CoD.
For more information on planning for and setting up Reserve Capacity Demand on your eServer
i5 or eServer p5 server, go to the eServer Information topic Working with Reserve Capacity
on Demand.
Critical console information should be backed up before installing a new version of HMC software
so that previous levels may be restored in the event of a problem in upgrading. See Backup
Critical Console Data.
Before you upgrade to the new version you should record HMC configuration information to
enable you to restore the current configuration to the newly upgraded system.
13. During system boot, select the Upgrade option by pressing F1.
14. Press F1 again to confirm.
15. When the base HMC installation has completed, the base HMC CD ejects from the drive.
Follow the prompt to insert CD 2.
16. Press Enter to reboot the HMC. If there is a modem installed, ensure that it is powered
on.
17. When the HMC boots, installation continues from CD 2. When that installation is
complete, the system will reboot.
18. Auto detection of peripheral devices/adapters should occur at this time (i.e., ethernet
adapter).
19. There is a timer on the keyboard mapping selection screen. Select an applicable
keyboard option for your locale.
20. Your saved upgrade data will be automatically restored at this time.
To install the HMC software for the first time without using the Guided Setup wizard:
1. Obtain the Product CDs.
2. Reboot your HMC with the new CD inserted in the DVD-RAM drive.
3. Monitor the reboot process carefully! If the Kudzu hardware discovery utility starts during
the reboot, you MUST select all defaults; otherwise, you will not be able to use your
keyboard.
4. When asked to perform an Install/Recovery or Upgrade, select Install/Recovery (F8).
The Guided Setup Wizard can be used for the initial installation of the HMC software on a new
system. It cannot be used for system upgrades of any kind, and is intended only for setting up a
new HMC.
The intended sequence of system configuration information that the Guided Setup Wizard asks
for is as follows:
5. Network settings:
o Host name, domain name, computer description, whether or not to use DNS
(and, if so, the server search order and domain suffix search order), and routing
information.
o For each ethernet network interface, whether it is for the private or public
network, the media speed, whether or not to enable the DHCP Server (which is
only applicable if the interface is for the private network), whether the IP address
is static or dynamic and if static, the TCP/IP address and network mask.
o Firewall information, which includes what ports will allow incoming connections
from non-local hosts and, optionally, what specific hosts are allowed to make
incoming connections on these ports.
6. Specifying customer contact information for service-related activities.
7. Specifying connectivity information for service-related activities
8. Authorizing users to use Service Agent and configuring notification of problem events
9. Configuring Service Focal Point.
Guided Setup can be initiated in two ways: from (a) a splash panel and (b) the WebSM user
interface
Launching the Setup Wizard from a Splash Panel
The Guided Setup Wizard splash panel, shown at left, is displayed at initial HMC logon. It is also
available anytime that user hscroot is logged on.
The Guided Setup Wizard can be launched from the Information and Setup option on the main
HMC GUI screen. Click on Information Center and Setup Wizard to display the Information and
setup window, then click on Launch the Guided Setup Wizard.
Customizing System Users, Tasks, and Roles
Configuring System Users
The Guided Setup Wizard is typically used to configure HMC system users. However, HMC
customization and modification is also done from the HMC Users panel.
1. In the Navigation area of the main HMC screen, click on HMC Management.
2. Select HMC Users. The HMC Users panel is displayed. To configure or modify an HMC
user, click the Manage HMC Users and Access task.
The User Profiles Screen is displayed with a list of existing HMC users. Highlight the user you
wish to modify and click User. You may:
o Add a user
o Copy user information
o Remove a user
o Modify user information
For details on the different user roles and each HMC User option, see the eServer Information
Center topic 'Overview of HMC Tasks, Roles and Commands'.
When you create an HMC user, you must assign that user a task role. Each task role allows the
user varying levels of access to tasks available on the HMC. There are five pre-defined roles:
• super administrator
• service representative
• operator
• product engineer
• viewer
From the HMC Users panel, to modify or add to the existing pre-defined user access task roles:
1. Click the Manage Access Task Roles and Managed Resource Roles. The Customize
User Controls panel is displayed.
2. To modify access task roles, check Task Roles. A list of currently defined roles is
displayed.
3. Highlight the targeted role and click Edit to modify access task roles.
For details on the different user roles and each HMC User option, see the Information Center
topic 'Overview of HMC Tasks, Roles and Commands'.
Managing Managed Resource Roles
You can assign managed systems and partitions to individual HMC users. This allows you to
create a user that has access to managed system A but not managed system B. Each grouping
of managed resource access is called a Managed Resource Role.
1. Check Managed Resource Roles. A list of currently defined managed resource roles is
displayed.
2. Highlight the targeted role and click Edit. You may add, copy, remove, or modify the
selected role.
For details on the different user roles and each HMC User option, see the Information Center
topic 'Overview of HMC Tasks, Roles and Commands'.
DLPAR Memory Operations
With Dynamic Logical Partitioning (DLPAR), you can add memory to, or remove memory from, a
partition without rebooting the partition’s operating system. You can also move memory from one
partition to another without rebooting either partition.
To move available memory resources from one partition to another without rebooting either
partition:
Commands by Name:
Use the chsyscfg command to modify the properties of a partition. The following example shows
how to change a partition’s cluster ID:
chsyscfg -r lpar -m <managed system> -i "lpar_id=l,cluster_id=3"
Instead of entering configuration information on the command line with the -i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
For more information about the valid attributes listed in this command example, refer to the
Command Attributes table.
Activating a Partition
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter.
Use the chsysstate command to perform a soft reset of a partition. Type the following:
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter.
Use the chsysstate command to perform a hard reset of a partition. Type the following:
The partition name can be specified instead of the partition ID by using the -n parameter instead
of the --id parameter.
Deleting an LPAR
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter.
Use the mksyscfg command to create a system profile. In the following example, the user is
making a system profile named sysprof1, with partition profile prof1 for partition lpar1 and
partition profile prof1 for partition lpar2.
Partition IDs can be specified instead of partition names when creating a system profile. This is
done by using the lpar_ids attribute instead of the lpar_names attribute.
Instead of entering configuration information on the command line with the-i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
Use the chsysstate command to activate a system profile. Type the following:
Use the chsysstate command to validate a system profile. Type the following:
To validate a system profile, then activate that system profile if the validation is successful, type
the following:
Use the rmsyscfg command to remove a system profile. Type the following:
Use the lssyscfg command to list a system profile’s properties. Type the following:
To list all system profiles for the managed system, type the following:
Use the chsyscfg command to modify system profile properties. In the following example, the
user is adding profiles prof1 for partition lpar3 and prof2 for partition lpar4 to system profile
sysprof1:
new_name profile_names
lpar_names | lpar_ids name
Instead of entering configuration information on the command line with the-i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
The lshwres command, which lists the hardware resources of a managed system, can be used
to display I/O, virtual I/O, processor, and memory resources.
• Partitions
lshwres -m <managed system> -r proc --level lpar
Creating LPARs
lnstead of entering configuration information on the command line with the -i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
Use the lssyscfg command to list all partitions in a managed system. To do this, enter:
To list only the names, IDs, and states of all partitions in a managed system, enter:
Use the lssyscfg command to list the properties of a specific partition. Type the following:
Note that the partition name can be specified instead of the partition ID by using the lpar_names
filter in place of the lpar_ids filter. Also, more than one partition may be specified in the filter list.
For information on using the lshwres command to list a partition's I/O, virtual I/O, processor, and
memory resources, see Listing Hardware Resources.
Use the lssyscfg command to display the state of a partition. Type the following:
Use the chsyscfg command to modify the properties of a partition. The following example shows
how to change a partition’s cluster ID:
Instead of entering configuration information on the command line with the -i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
For more information about the valid attributes listed in this command example, refer to the
Command Attributes table.
Activating a Partition
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter.
Use the chsysstate command to perform a soft reset of a partition. Type the following:
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter.
Use the chsysstate command to perform a hard reset of a partition. Type the following:
The partition name can be specified instead of the partition ID by using the -n parameter instead
of the --id parameter.
Deleting an LPAR
This command removes the specified partition and all of its associated partition profiles from the
specified managed system. The partition’s profiles are also removed from any system profiles
that contain them.
The partition ID can be specified instead of the partition name by using the --id parameter instead
of the -n parameter
Memory
Memory can be dynamically added to a partition, removed from a partition, or moved from one
partition to another. In the following commands, the quantity of memory to be added, removed, or
moved must be specified with the -q flag. This quantity is in megabytes, and must be a multiple of
the memory region size for the managed system.
To see what the memory region size is for the managed system, enter this command:
A physical I/O slot can be dynamically added to a partition, removed from a partition, or moved
from one partition to another. In the following commands, the DRC index of the I/O slot to be
added, removed, or moved must be specified with the -s flag.
Note that only one physical I/O slot can be added, removed, or moved at a time.
To see the DRC indexes for all of the physical I/O slots that are on the managed system, enter
this command:
The DRC index for each slot is returned via the drc_index attribute.
To move a physical I/O slot from one partition to another partition, enter this command:
Processors
• For partitions using shared processors, processing resources include virtual processors
and processing units.
Note: Currently, AIX/Linux partitions using shared processors do not support processor DLPAR
operations.
In the following commands, for partitions using dedicated processors, the quantity of processors
to be added, removed, or moved are specified with the --procs flag.
For partitions using shared processors, the quantity of virtual processors to be added, removed,
or moved are also specified with the --procs flag. The quantity of processing units to be added,
removed, or moved are specified with the --procunits flag. Both of these flags can be specified,
but only one is required.
Note that the quantity of processing units must be multiplied by 100 for the command. For
example, to add, remove, or move .5 processing units, specify a quantity of 50.
To remove processors from a partition using dedicated processors, enter this command:
To remove processors from a partition using shared processors, enter this command:
To move processors from a partition using dedicated processors to another, enter this command:
To move processors from a partition using shared processors to another, enter this command:
chhwres -r proc -m <managed system> -o m -p <source partition name> -t
<target partition name> --procs <quantity> --procunits <quantity>
Processing resources can also be moved between partitions using dedicated processors and
partitions using shared processors. To move processing resources from a partition using
dedicated processors to a partition using shared processors, specify the quantity of processors
using the --procs flag. This quantity is converted to processing units (by multiplying the quantity
by 100) by the HMC for the target partition.
To move processing resources from a partition using shared processors to a partition using
dedicated processors, specify the quantity of processing units (which must be a multiple of 100)
using the --procunits flag. This quantity is converted to processors (by dividing the quantity by
100) by the HMC for the target partition. The --procs flag cannot be specified in this case
Use the mksyscfg command to create a partition profile. The following is an example of how to
create a partition profile:
The profile name (name) and the partition (lpar_id or lpar_name) must be specified. Instead of
entering configuration information on the command line with the -i flag, the information can
instead be placed in a file, and the filename specified with the -f flag.
For more information about the valid attributes listed in this command example, refer to the
Command Attributes table.
Use the lssyscfg command to list a partition profile. Type the following:
Use the --filter parameter to specify the partition for which partition profiles are to be listed, and to
specify which profile names to list. While the filter can only specify a single partition, it can specify
multiple profile names for that partition.
Note that the partition ID can be specified instead of the partition name by using the lpar_ids filter
in place of the lpar_names filter.
Use the chsyscfg command to modify a partition profile’s properties. The following example
shows how to change prof1's memory amounts:
Instead of entering configuration information on the command line with the -i flag, the information
can instead be placed in a file, and the filename specified with the -f flag.
For more information about the valid attributes listed in this command example, refer to the
Command Attributes table.
Use the rmsyscfg command to remove a partition profile. Type the following:
The partition ID can be specified instead of the partition name by using the --id parameter in
place of the -p parameter.