| ETCD节点 | Master节点 | Node节点 | VIP |
|---|---|---|---|
| 192.168.100.136 | 192.168.100.139 | 192.168.100.142 | |
| 192.168.100.137 | 192.168.100.140 | 192.168.100.143 | 192.168.100.150 |
| 192.168.100.138 | 192.168.100.141 | 192.168.100.144 |
注: Master节点使用KeepAlived+HaProxy进行高可用负载均衡
- Keepalived对Master节点的Kube-apiserver提供高可用VIP服务
- Haproxy监听VIP对Kube-apiserver提供负载均衡服务，所有组件通过开放的8443端口访问
| 系统 | CFSSL | ETCD | KeepAlived | HaProxy | Kubernetes | Containerd | Crictl |
|---|---|---|---|---|---|---|---|
| CentOS 8 | 1.5.0 | v3.4.14 | 2.1.5 | 2.3.2 | v1.19.5 | 1.4.3 | v1.19.0 |
注1：变量化配置更改项主要在"defaults/main.yaml"文件中，cfssl安装在ansible主机，kubernetes压缩在ansible主机，其他应用下载压缩包在ansible主机即可。
注2: 关于集群网络，可以选择使用yaml部署Calico、Cilium、Flannel(需安装cni-plugins[/opt/cni/bin/])等网络插件
# yum -y install ansible
# egrep -v "^#|^$" /etc/ansible/hosts
[etcd]
192.168.100.136 hostname=etcd1
192.168.100.137 hostname=etcd2
192.168.100.138 hostname=etcd3
[master]
192.168.100.139 hostname=master1
192.168.100.140 hostname=master2
192.168.100.141 hostname=master3
[node]
192.168.100.142 hostname=node1
192.168.100.143 hostname=node2
192.168.100.144 hostname=node3
[etcd:vars]
ansible_ssh_user="root"
ansible_ssh_pass="wangzhijian"
[master:vars]
ansible_ssh_user="root"
ansible_ssh_pass="wangzhijian"
[node:vars]
ansible_ssh_user="root"
ansible_ssh_pass="wangzhijian"ssh-keygen -t rsa -P ''ansible all -m copy -a "src=/etc/hosts dest=/etc/hosts"source cfssl.shansible-playbook k8s.yaml# ansible 192.168.100.136 -a "etcdctl --endpoints=https://192.168.100.136:2379 --cacert=/etc/ssl/etcd/ca.pem --key=/etc/ssl/etcd/etcd-key.pem --cert=/etc/ssl/etcd/etcd.pem -w=table member list"
192.168.100.136 | CHANGED | rc=0 >>
+------------------+---------+-------+------------------------------+------------------------------+------------+
| ID | STATUS | NAME | PEER ADDRS | CLIENT ADDRS | IS LEARNER |
+------------------+---------+-------+------------------------------+------------------------------+------------+
| 25d84952513e4c13 | started | etcd3 | https://192.168.100.138:2380 | https://192.168.100.138:2379 | false |
| 5700b9ecd6ca26d0 | started | etcd2 | https://192.168.100.137:2380 | https://192.168.100.137:2379 | false |
| 8c08a828e17afa88 | started | etcd1 | https://192.168.100.136:2380 | https://192.168.100.136:2379 | false |
+------------------+---------+-------+------------------------------+------------------------------+------------+
# ansible 192.168.100.139 -a "kubectl cluster-info"
192.168.100.139 | CHANGED | rc=0 >>
Kubernetes control plane is running at https://192.168.100.150:8443
# ansible 192.168.100.139 -a "kubectl get cs"
192.168.100.139 | CHANGED | rc=0 >>
NAME STATUS MESSAGE ERROR
scheduler Healthy ok
controller-manager Healthy ok
etcd-0 Healthy {"health":"true"}
etcd-2 Healthy {"health":"true"}
etcd-1 Healthy {"health":"true"}
# ansible 192.168.100.139 -a "kubectl get nodes"
192.168.100.139 | CHANGED | rc=0 >>
NAME STATUS ROLES AGE VERSION
node1 Ready <none> 14h v1.20.2
node2 Ready <none> 14h v1.20.2
node3 Ready <none> 14h v1.20.2注:清除前请删除所有Pod
ansible-playbook clean-cluster.yaml
React
Typescript
Django
Laravel
Facebook
Microsoft
Google
Alibaba
D3
Tencent