1 kubernetes 设置 Master 可调度与不可调度

默认情况下，master节点是不参与调度的，且在master节点上有一个污点NoSchedule（表示k8s将不会将Pod调度到具有该污点的Node上），有关污点章节可参考https://www.cnblogs.com/panw/p/16343392.html

语法

kubectl taint node [node] key=value[effect]

[effect] 可取值: [ NoSchedule | PreferNoSchedule | NoExecute ]

NoSchedule: 一定不能被调度

PreferNoSchedule: 尽量不要调度

NoExecute: 不仅不会调度, 还会驱逐Node上已有的Pod

取消污点

取消污点
[root@k8s-master01 ~]# kubectl taint node k8s-master node-role.kubernetes.io/master-

设置污点

# 设置为一定不能被调度

[root@k8s-master01 ~]# kubectl taint node k8s-master01 node-role.kubernetes.io/master="":NoSchedule
node/k8s-master01 tainted
[root@k8s-master01 ~]# kubectl taint node k8s-master02 node-role.kubernetes.io/master="":NoSchedule
node/k8s-master02 tainted
[root@k8s-master01 ~]# kubectl taint node k8s-master03 node-role.kubernetes.io/master="":NoSchedule
node/k8s-master03 tainted
[root@k8s-master01 ~]# 

# 查看污点
[root@k8s-master01 ~]# kubectl  describe  node | grep Ta
Taints:             node-role.kubernetes.io/master:NoSchedule
Taints:             node-role.kubernetes.io/master:NoSchedule
Taints:             node-role.kubernetes.io/master:NoSchedule
Taints:             <none>
Taints:             <none>
Taints:             <none>
Taints:             <none>
Taints:             <none>
[root@k8s-master01 ~]# 

查看验证

# 查看已经调度到maser上的pod没有被驱逐
[root@k8s-master01 ~]# kubectl  get pod -o wide
NAMESPACE              NAME                                         READY   STATUS    RESTARTS       AGE   IP               NODE           NOMINATED NODE   READINESS GATES
default                hostname-test-cby-58d85dccdb-7zgjj           1/1     Running   1 (2d1h ago)   19d   172.25.244.195   k8s-master01   <none>           <none>
default                hostname-test-cby-58d85dccdb-8t7zv           1/1     Running   1 (2d1h ago)   19d   172.25.244.196   k8s-master01   <none>           <none>
default                hostname-test-cby-58d85dccdb-9bqsq           1/1     Running   1 (2d1h ago)   19d   172.25.92.74     k8s-master02   <none>           <none>
default                hostname-test-cby-58d85dccdb-jj2ml           1/1     Running   1 (2d1h ago)   19d   172.17.125.3     k8s-node01     <none>           <none>
default                hostname-test-cby-58d85dccdb-k96zl           1/1     Running   1 (2d1h ago)   19d   172.18.195.3     k8s-master03   <none>           <none>
default                hostname-test-cby-58d85dccdb-lng8b           1/1     Running   1 (2d1h ago)   19d   172.29.115.131   k8s-node04     <none>           <none>
default                hostname-test-cby-58d85dccdb-lsrbg           1/1     Running   1 (2d1h ago)   19d   172.25.214.195   k8s-node03     <none>           <none>
default                hostname-test-cby-58d85dccdb-mlv24           1/1     Running   1 (2d1h ago)   19d   172.17.54.131    k8s-node05     <none>           <none>
default                hostname-test-cby-58d85dccdb-p5vc8           1/1     Running   1 (2d1h ago)   19d   172.27.14.195    k8s-node02     <none>           <none>
default                hostname-test-cby-58d85dccdb-z6ptf           1/1     Running   1 (2d1h ago)   19d   172.25.214.196   k8s-node03   
  <none>           <none>
[root@k8s-master01 ~]# 

设置污点

# 设置为不仅不会调度, 还会驱逐Node上已有的Pod
[root@k8s-master01 ~]# kubectl taint node k8s-master03 node-role.kubernetes.io/master="":NoExecute
node/k8s-master03 tainted
[root@k8s-master01 ~]# kubectl taint node k8s-master02 node-role.kubernetes.io/master="":NoExecute
node/k8s-master02 tainted
[root@k8s-master01 ~]# kubectl taint node k8s-master01 node-role.kubernetes.io/master="":NoExecute
node/k8s-master01 tainted

# 查看污点
[root@k8s-master01 ~]# kubectl  describe  node | grep Ta
Taints:             node-role.kubernetes.io/master:NoExecute
Taints:             node-role.kubernetes.io/master:NoExecute
Taints:             node-role.kubernetes.io/master:NoExecute
Taints:             <none>
Taints:             <none>
Taints:             <none>
Taints:             <none>
Taints:             <none>
[root@k8s-master01 ~]# 

查看验证

# 查看已经调度到master节点的pod已进行驱逐
[root@k8s-master01 ~]# kubectl  get pod  -A -o wide
NAMESPACE              NAME                                         READY   STATUS              RESTARTS       AGE   IP               NODE           NOMINATED NODE   READINESS GATES
default                mysql-0                                      2/2     Running             0              34m   172.27.14.206    k8s-node02     <none>           <none>
default                mysql-1                                      2/2     Running             0              34m   172.17.125.11    k8s-node01     <none>           <none>
default                mysql-2                                      2/2     Terminating         0              34m   172.18.195.10    k8s-master03   <none>           <none>
[root@k8s-master01 ~]# 

2 k8s删除PV操作

使用以下命令查看pv

kubectl get pv

使用以下命令删除PV，并且在 kubectl delete pv 之后，pv 一直处在 Terminating 状态，使用如下命令也是好使的。

kubectl patch pv {PVC_NAME} -p '{"metadata":{"finalizers":null}}'

使用以下两个命令删除 PV

kubectl delete pv {PVC_NAME}

K8S PV一直Released状态的解决办法
参考： https://blog.csdn.net/wz_cmyc/article/details/120542934