集群内安装了

• flanal-用于Pod网络
• canal-用于网络ACL

在初期进行集群的初始化和配置时，在删除了canal后尝试重启了CoreDNS，但是发现无论如何重启，CoreDNS永远处于ContainerCreating的状态

describe一下ContainerCreating的Pod

kubectl describe pods -n kube-system coredns-<***>

着重查看Events部分

Events:
  Type     Reason                  Age               From               Message
  ----     ------                  ----              ----               -------
  Normal   Scheduled               28s               default-scheduler  Successfully assigned kube-system/coredns-7dfb448f-kl8n4 to cn-02101-prod-as-k8s-tl-w-03
  Warning  FailedCreatePodSandBox  27s               kubelet            Failed to create pod sandbox: rpc error: code = Unknown desc = [failed to set up sandbox container "2e2dc0d544b37f326720bee9f5a2b12f6d4fa76e7f6f770c18f55d7ed92c2999" network for pod "coredns-7dfb448f-kl8n4": networkPlugin cni failed to set up pod "coredns-7dfb448f-kl8n4_kube-system" network: error getting ClusterInformation: connection is unauthorized: Unauthorized, failed to clean up sandbox container "2e2dc0d544b37f326720bee9f5a2b12f6d4fa76e7f6f770c18f55d7ed92c2999" network for pod "coredns-7dfb448f-kl8n4": networkPlugin cni failed to teardown pod "coredns-7dfb448f-kl8n4_kube-system" network: error getting ClusterInformation: connection is unauthorized: Unauthorized]
  Normal   SandboxChanged          1s (x4 over 26s)  kubelet            Pod sandbox changed, it will be killed and re-created.

networkPlugin cni failed to set up pod "coredns-7dfb448f-kl8n4_kube-system"
发现cni的网络插件无法创建SandBox

这是由于kubectl删除了canal或calico配置，但是目标机器上依然保留着其cni配置导致的
在无法启动的node上删除掉/etc/cni/net.d/ 目录下的 calico 和 canal 配置文件重启目标机器即可。
注意：最好所有的非master节点都这样操作一下

[root@node2 ~]# ll /etc/cni/net.d/
total 12
-rw-r--r-- 1 root root  712 Oct  8 01:16 10-canal.conflist
-rw-r--r-- 1 root root  292 Oct  8 01:06 10-flannel.conflist
-rw------- 1 root root 2697 Oct  8 01:16 calico-kubeconfig
[root@node2 ~]# rm -rf /etc/cni/net.d/calico-kubeconfig 
[root@node2 ~]# rm -rf /etc/cni/net.d/10-canal.conflist 

然后就好了

顺便重新安装了canal~