由于 CNI 插件未初始化，Kubernetes 工作节点未就绪

Question

我正在kind本地 Macbook 上运行测试 kubernetes 集群。

我发现其中一个节点的状态为NotReady：

$ kind get clusters                                                                                                                                                                 
mc

$ kubernetes get nodes
NAME                STATUS     ROLES    AGE     VERSION
mc-control-plane    Ready      master   4h42m   v1.18.2
mc-control-plane2   Ready      master   4h41m   v1.18.2
mc-control-plane3   Ready      master   4h40m   v1.18.2
mc-worker           NotReady   <none>   4h40m   v1.18.2
mc-worker2          Ready      <none>   4h40m   v1.18.2
mc-worker3          Ready      <none>   4h40m   v1.18.2

唯一有趣的kubectl describe node mc-worker是 CNI 插件未初始化：

Conditions:
  Type             Status  LastHeartbeatTime                 LastTransitionTime                Reason                       Message
  ----             ------  -----------------                 ------------------                ------                       -------
  MemoryPressure   False   Tue, 11 Aug 2020 16:55:44 -0700   Tue, 11 Aug 2020 12:10:16 -0700   KubeletHasSufficientMemory   kubelet has sufficient memory available
  DiskPressure     False   Tue, 11 Aug 2020 16:55:44 -0700   Tue, 11 Aug 2020 12:10:16 -0700   KubeletHasNoDiskPressure     kubelet has no disk pressure
  PIDPressure      False   Tue, 11 Aug 2020 16:55:44 -0700   Tue, 11 Aug 2020 12:10:16 -0700   KubeletHasSufficientPID      kubelet has sufficient PID available
  Ready            False   Tue, 11 Aug 2020 16:55:44 -0700   Tue, 11 Aug 2020 12:10:16 -0700   KubeletNotReady              runtime network not ready: NetworkReady=false reason:NetworkPluginNotReady
message:Network plugin returns error: cni plugin not initialized

我有 2 个类似的集群，这只发生在这个集群上。

由于kind使用本地 Docker 守护进程将这些节点作为容器运行，因此我已经尝试重新启动容器（应该相当于重新启动节点）。

我考虑过删除并重新创建集群，但应该有一种方法可以解决这个问题，而无需重新创建集群。

以下是我正在运行的版本：

$ kind version                                                                                                                                                                     
kind v0.8.1 go1.14.4 darwin/amd64

$ kubectl version                                                                                                                                                  
Client Version: version.Info{Major:"1", Minor:"16+", GitVersion:"v1.16.6-beta.0", GitCommit:"e7f962ba86f4ce7033828210ca3556393c377bcc", GitTreeState:"clean", BuildDate:"2020-01-15T08:26:26Z", GoVersion:"go1.13.5", Compiler:"gc", Platform:"darwin/amd64"}
Server Version: version.Info{Major:"1", Minor:"18", GitVersion:"v1.18.2", GitCommit:"52c56ce7a8272c798dbc29846288d7cd9fbae032", GitTreeState:"clean", BuildDate:"2020-04-30T20:19:45Z", GoVersion:"go1.13.9", Compiler:"gc", Platform:"linux/amd64"}

您如何解决这个问题？

Answer 1

最可能的原因：

\n

docker VM 耗尽了某些资源，无法在该特定节点上启动 CNI。

\n

您可以通过连接到 HyperKit VM 来查看它：

\n

从外壳：

\n

screen ~/Library/Containers/com.docker.docker/Data/vms/0/tty\n

\n

如果由于某种原因这不起作用：

\n

docker run -it --rm --privileged --pid=host alpine nsenter -t 1 -m -u -n -i sh\n

\n

进入虚拟机后：

\n

# ps -Af\n# free\n# df -h\n...\n

\n

然后您可以随时更新 docker UI 上的设置：

\n

\n

最后，您的节点毕竟是在容器中运行的。因此，您可以连接到该容器并查看kubelet看到的错误：

\n

docker ps\nCONTAINER ID        IMAGE                  COMMAND                  CREATED             STATUS              PORTS                       NAMES\n6d881be79f4a        kindest/node:v1.18.2   "/usr/local/bin/entr\xe2\x80\xa6"   32 seconds ago      Up 29 seconds       127.0.0.1:57316->6443/tcp   kind-control-plane\ndocker exec -it 6d881be79f4a bash\nroot@kind-control-plane:/# systemctl status kubelet\n\xe2\x97\x8f kubelet.service - kubelet: The Kubernetes Node Agent\n   Loaded: loaded (/kind/systemd/kubelet.service; enabled; vendor preset: enabled)\n  Drop-In: /etc/systemd/system/kubelet.service.d\n           \xe2\x94\x94\xe2\x94\x8010-kubeadm.conf\n   Active: active (running) since Wed 2020-08-12 02:32:16 UTC; 35s ago\n     Docs: http://kubernetes.io/docs/\n Main PID: 768 (kubelet)\n    Tasks: 23 (limit: 2348)\n   Memory: 32.8M\n   CGroup: /docker/6d881be79f4a8ded3162ec6b5caa8805542ff9703fabf5d3d2eee204a0814e01/system.slice/kubelet.service\n           \xe2\x94\x94\xe2\x94\x80768 /usr/bin/kubelet --bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig=/etc/kubernetes/kubelet.conf --config=/var/lib/kubelet\n/config.yaml --container-runtime=remote --container-runtime-endpoint=/run/containerd/containerd.sock --fail-swap-on=false --node-ip= --fail-swap-on=false\n...\n

\n

\xe2\x9c\x8c\xef\xb8\x8f

\n