官方文档： https://kubernetes.io/docs/setup/independent/high-availability/#first-steps-for-both-methods
一、环境准备
| ip | hostname | role |
|---|---|---|
| 10.127.24.180 | k8s-node1 | k8s-node |
| 10.127.24.179 | k8s-master1 | k8s-master |
| 10.39.5.226 | k8s-node2 | k8s-node |
| 10.39.14.204 | etcd-node1 | etcd-node1 |
| 10.39.14.205 | etcd-node2 | etcd-node2 |
| 10.39.14.206 | etcd-node3 | etcd-node3 |
开始前保证：
- 所有节点都可以 ping 通
- 都有 sudo 权限
- 所有节点安装了 kubeadm kubelet。kubectl可选。
- etcd 集群已搭建
- 所有节点都有 etcd
二、master 启动集群
1、将 etcd 的认证文件 copy 来
copy 认证文件方法：
1、 先把之前的集群信息都清除干净
2、 跳板机把 etcd 的认证文件 copy 下来
3、 把认证文件放到指定位置
[跳板机]$ scp -r ~/host1 root@<master_ip>:~
[跳板机]$ ssh root@<master_ip>
kubeadm reset
rm -rf /etc/kubernetes/pki/
mkdir -p /etc/kubernetes/pki/etcd/
chown -R root:root ~/host1/pki/
mv ~/host1/pki/etcd/ca.crt /etc/kubernetes/pki/etcd/
mv ~/host1/pki/apiserver-etcd-client.crt /etc/kubernetes/pki/
mv ~/host1/pki/apiserver-etcd-client.key /etc/kubernetes/pki/
rm -rf ~/host1/
2、创建一个 kubeadm-config.yaml：
参数参考 https://godoc.org/k8s.io/kubernetes/cmd/kubeadm/app/apis/kubeadm/v1beta1
cat <<EOF > /etc/kubernetes/kubeadm-config.yaml
apiVersion: kubeadm.k8s.io/v1beta1
kind: ClusterConfiguration
kubernetesVersion: v1.13.0
apiServer:
certSANs:
- 10.127.24.179
- 127.0.0.1
networking:
podSubnet: 10.244.0.0/16
etcd:
external:
endpoints:
- https://10.39.14.204:2379
- https://10.39.14.205:2379
- https://10.39.14.206:2379
caFile: /etc/kubernetes/pki/etcd/ca.crt
certFile: /etc/kubernetes/pki/apiserver-etcd-client.crt
keyFile: /etc/kubernetes/pki/apiserver-etcd-client.key
EOF
3、启动集群
kubeadm init --config /etc/kubernetes/kubeadm-config.yaml
记录：
kubeadm join xxxxx --token xxxxx --discovery-token-ca-cert-hash sha256:xxxx
4、配置 flannel 网络
kubectl apply -f kube-flannel.yml
输出：
clusterrole.rbac.authorization.k8s.io/flannel created
clusterrolebinding.rbac.authorization.k8s.io/flannel created
serviceaccount/flannel created
configmap/kube-flannel-cfg created
daemonset.extensions/kube-flannel-ds-amd64 created
daemonset.extensions/kube-flannel-ds-arm64 created
daemonset.extensions/kube-flannel-ds-arm created
daemonset.extensions/kube-flannel-ds-ppc64le created
daemonset.extensions/kube-flannel-ds-s390x created
5、查看当前集群状态
$ kubectl get componentstatus
NAME STATUS MESSAGE ERROR
controller-manager Healthy ok
scheduler Healthy ok
etcd-1 Healthy {"health": "true"}
etcd-2 Healthy {"health": "true"}
etcd-0 Healthy {"health": "true"}
$ kubectl get nodes
NAME STATUS ROLES AGE VERSION
k8s-master1 Ready master 66m v1.13.0
$ kubectl get pods -n kube-system
NAME READY STATUS RESTARTS AGE
coredns-86c58d9df4-j8t2t 1/1 Running 0 6m20s
coredns-86c58d9df4-wr9sk 1/1 Running 0 6m20s
etcd-k8s-master1 1/1 Running 0 5m32s
kube-apiserver-k8s-master1 1/1 Running 0 5m43s
kube-controller-manager-k8s-master1 1/1 Running 0 5m21s
kube-flannel-ds-amd64-st4sv 1/1 Running 0 4m
kube-proxy-d7t4d 1/1 Running 0 6m20s
kube-scheduler-k8s-master1 1/1 Running 0 5m39s
三、Node 加入集群
kubeadm join 10.127.24.179:6443 --token xxxxx --discovery-token-ca-cert-hash sha256:xxxxxx
输出：
This node has joined the cluster!
四、（可选）master ：check 一下 Node 状态
kubectl get nodes
输出：
NAME STATUS ROLES AGE VERSION
k8s-master1 Ready master 3h11m v1.13.0
k8s-node1 Ready <none> 2m29s v1.13.0
k8s-node2 Ready <none> 12s v1.13.0
