Skip to content

Commit fb7c53d

Browse files
judasnjudasn
committed
2019-01-31
1 parent aa91f46 commit fb7c53d

File tree

2 files changed

+237
-2
lines changed

2 files changed

+237
-2
lines changed

‎favorite-file/shell/install_docker_k8s_disable_firewalld_centos7-aliyun.sh

Lines changed: 49 additions & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -0,0 +1,49 @@
1+
#!/bin/sh
2+
3+
echo "-----------------------------------------禁用防火墙"
4+
systemctl stop firewalld.service
5+
systemctl disable firewalld.service
6+
7+
echo "-----------------------------------------安装 docker 所需环境"
8+
9+
yum install -y yum-utils device-mapper-persistent-data lvm2
10+
11+
echo "-----------------------------------------添加 repo（可能网络会很慢，有时候会报：Timeout，所以要多试几次）"
12+
echo "-----------------------------------------官网的地址 yum-config-manager --add-repo https://download.docker.com/linux/centos/docker-ce.repo"
13+
echo "-----------------------------------------这里用阿里云进行加速，不然可能会出现无法安装，阿里云官网说明：https://help.aliyun.com/document_detail/60742.html"
14+
15+
yum-config-manager --add-repo http://mirrors.aliyun.com/docker-ce/linux/centos/docker-ce.repo
16+
yum makecache fast
17+
18+
echo "-----------------------------------------开始安装 docker"
19+
20+
yum install -y docker-ce-18.06.1.ce-3.el7
21+
22+
echo "-----------------------------------------启动 Docker"
23+
24+
systemctl start docker.service
25+
systemctl enable docker.service
26+
27+
echo "-----------------------------------------安装结束"
28+
29+
echo "-----------------------------------------docker 加速"
30+
31+
touch /etc/docker/daemon.json
32+
33+
cat << EOF >> /etc/docker/daemon.json
34+
{
35+
"registry-mirrors": ["https://ldhc17y9.mirror.aliyuncs.com"]
36+
}
37+
EOF
38+
39+
systemctl daemon-reload
40+
systemctl restart docker
41+
42+
echo "-----------------------------------------运行 hello world 镜像"
43+
44+
docker run hello-world
45+
46+
47+
48+
49+

‎markdown-file/Docker-Install-And-Usage.md

Lines changed: 188 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -839,23 +839,209 @@ sudo chmod +x /usr/local/bin/docker-compose
839839

840840
#### 开始安装 - Kubernetes 1.13.2 版本
841841

842+
- 三台机子：
843+
- master-1：`192.168.0.127`
844+
- node-1：`192.168.0.128`
845+
- node-2：`192.168.0.129`
842846
- 官网最新版本：<https://github.com/kubernetes/kubernetes/releases>
843847
- 官网 1.13 版本的 changelog：<https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG-1.13.md>
844848
- 所有节点安装 Docker 18.06，并设置阿里云源
849+
- 可以参考：[点击我o(∩_∩)o ](https://github.com/judasn/Linux-Tutorial/blob/master/favorite-file/shell/install_docker_k8s_disable_firewalld_centos7-aliyun.sh)
850+
- 核心，查看可以安装的 Docker 列表：`yum list docker-ce --showduplicates`
845851
- 所有节点设置 kubernetes repo 源，并安装 Kubeadm、Kubelet、Kubectl 都设置阿里云的源
846852
- Kubeadm 初始化集群过程当中，它会下载很多的镜像，默认也是去 Google 家里下载。但是 1.13 新增了一个配置：`--image-repository` 算是救了命。
853+
- 具体流程：
854+
855+
```
856+
主机时间同步
857+
systemctl start chronyd.service
858+
systemctl enable chronyd.service
859+
860+
systemctl stop firewalld.service
861+
systemctl disable firewalld.service
862+
systemctl disable iptables.service
863+
864+
865+
setenforce 0
866+
867+
sed -i 's/SELINUX=enforcing/SELINUX=disabled/g' /etc/selinux/config
868+
869+
swapoff -a && sysctl -w vm.swappiness=0
870+
871+
872+
hostnamectl --static set-hostname k8s-master-1
873+
hostnamectl --static set-hostname k8s-node-1
874+
hostnamectl --static set-hostname k8s-node-2
875+
876+
877+
vim /etc/hosts
878+
192.168.0.127 k8s-master-1
879+
192.168.0.128 k8s-node-1
880+
192.168.0.129 k8s-node-2
881+
882+
master 免密
883+
生产密钥对
884+
ssh-keygen -t rsa
885+
886+
887+
公钥内容写入 authorized_keys
888+
cat /root/.ssh/id_rsa.pub >> /root/.ssh/authorized_keys
889+
890+
测试：
891+
ssh localhost
892+
893+
将公钥复制到其他机子
894+
ssh-copy-id -i ~/.ssh/id_rsa.pub -p 22 root@k8s-node-1（根据提示输入 k8s-node-1 密码）
895+
ssh-copy-id -i ~/.ssh/id_rsa.pub -p 22 root@k8s-node-2（根据提示输入 k8s-node-2 密码）
896+
897+
898+
在 linux01 上测试
899+
ssh k8s-master-1
900+
ssh k8s-node-1
901+
ssh k8s-node-2
902+
903+
904+
905+
vim /etc/yum.repos.d/kubernetes.repo
906+
907+
[kubernetes]
908+
name=Kubernetes
909+
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64/
910+
enabled=1
911+
gpgcheck=1
912+
repo_gpgcheck=1
913+
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
914+
915+
916+
scp -r /etc/yum.repos.d/kubernetes.repo root@k8s-node-1:/etc/yum.repos.d/
917+
scp -r /etc/yum.repos.d/kubernetes.repo root@k8s-node-2:/etc/yum.repos.d/
918+
919+
920+
所有机子
921+
yum install -y kubelet-1.13.2 kubeadm-1.13.2 kubectl-1.13.2 --disableexcludes=kubernetes
922+
923+
924+
vim /etc/systemd/system/kubelet.service.d/10-kubeadm.conf
925+
最后一行添加：Environment="KUBELET_CGROUP_ARGS=--cgroup-driver=cgroupfs"
926+
927+
928+
systemctl enable kubelet && systemctl start kubelet
929+
930+
kubeadm version
931+
kubectl version
932+
933+
934+
vim /etc/sysctl.d/k8s.conf
935+
net.bridge.bridge-nf-call-ip6tables = 1
936+
net.bridge.bridge-nf-call-iptables = 1
937+
938+
sysctl --system
939+
940+
```
941+
942+
- 初始化 master 节点：
943+
944+
```
945+
946+
推荐：
947+
kubeadm init \
948+
--image-repository registry.aliyuncs.com/google_containers \
949+
--pod-network-cidr 10.244.0.0/16 \
950+
--kubernetes-version 1.13.2 \
951+
--service-cidr 10.96.0.0/12 \
952+
--apiserver-advertise-address=0.0.0.0 \
953+
--ignore-preflight-errors=Swap
954+
955+
10.244.0.0/16是 flannel 插件固定使用的ip段，它的值取决于你准备安装哪个网络插件
956+
957+
终端会输出核心内容：
958+
Your Kubernetes master has initialized successfully!
959+
960+
To start using your cluster, you need to run the following as a regular user:
961+
962+
mkdir -p $HOME/.kube
963+
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
964+
sudo chown $(id -u):$(id -g) $HOME/.kube/config
965+
966+
You should now deploy a pod network to the cluster.
967+
Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:
968+
https://kubernetes.io/docs/concepts/cluster-administration/addons/
969+
970+
You can now join any number of machines by running the following on each node
971+
as root:
972+
973+
kubeadm join 192.168.0.127:6443 --token 53mly1.yf9llsghle20p2uq --discovery-token-ca-cert-hash sha256:a9f26eef42c30d9f4b20c52058a2eaa696edc3f63ba20be477fe1494ec0146f7
974+
975+
976+
977+
978+
也可以使用另外一个流行网络插件 calico：
979+
kubeadm init --image-repository registry.aliyuncs.com/google_containers --pod-network-cidr=192.168.0.0/16 --kubernetes-version v1.13.2
980+
981+
982+
983+
mkdir -p $HOME/.kube
984+
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
985+
sudo chown $(id -u):$(id -g) $HOME/.kube/config
986+
987+
export KUBECONFIG=/etc/kubernetes/admin.conf
988+
echo "export KUBECONFIG=/etc/kubernetes/admin.conf" >> ~/.zshrc
989+
source ~/.zshrc
990+
991+
992+
查询我们的 token
993+
kubeadm token list
994+
995+
996+
kubectl cluster-info
997+
```
998+
999+
- 到 node 节点进行加入：
1000+
1001+
```
1002+
1003+
kubeadm join 192.168.0.127:6443 --token 53mly1.yf9llsghle20p2uq --discovery-token-ca-cert-hash sha256:a9f26eef42c30d9f4b20c52058a2eaa696edc3f63ba20be477fe1494ec0146f7
1004+
1005+
1006+
在 master 节点上：kubectl get cs
1007+
NAME STATUS MESSAGE ERROR
1008+
controller-manager Healthy ok
1009+
scheduler Healthy ok
1010+
etcd-0 Healthy {"health": "true"}
1011+
结果都是 Healthy 则表示可以了，不然就得检查。必要时可以用：`kubeadm reset` 重置，重新进行集群初始化
1012+
1013+
1014+
1015+
master 安装 Flannel
1016+
cd /opt && wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
1017+
1018+
kubectl apply -f /opt/kube-flannel.yml
1019+
1020+
1021+
验证：
1022+
kubectl get pods --all-namespaces
1023+
kubectl get nodes
1024+
如果还是 NotReady，则查看错误信息：
1025+
kubectl describe pod kube-scheduler-master.hanli.com -n kube-system
1026+
kubectl logs kube-scheduler-master.hanli.com -n kube-system
1027+
tail -f /var/log/messages
1028+
1029+
```
1030+
1031+
1032+
8471033

8481034
#### 主要概念
8491035

8501036
- Master 节点，负责集群的调度、集群的管理
851-
- 常见组件：
1037+
- 常见组件：<https://kubernetes.io/docs/concepts/overview/components/>
8521038
- kube-apiserver：API服务
8531039
- kube-scheduler：调度
8541040
- Kube-Controller-Manager：容器编排
8551041
- Etcd：保存了整个集群的状态
8561042
- Kube-proxy：负责为 Service 提供 cluster 内部的服务发现和负载均衡
8571043
- Kube-DNS：负责为整个集群提供 DNS 服务
858-
- Workers 节点，负责容器相关的处理
1044+
- node 节点，负责容器相关的处理
8591045

8601046
- `Pods`
8611047

0 commit comments

Comments
 (0)