k8s安装网络组件calico报错Init:ImagePullBackOff

问题:

安装网络组件calico报错Init:ImagePullBackOff

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

[root@k8s-master 15:33:08~/kubernets]# kubectl  get pods -n kube-system                      NAME                                     READY   STATUS                  RESTARTS      AGE calico-kube-controllers-cd8566cf-cqvxj   1/1     Running                 0             33m calico-node-6f4np                        1/1     Running                 1 (11m ago)   33m calico-node-bvvhc                        0/1     Init:ImagePullBackOff   0             5m37s     # [root@k8s-master 15:46:58~/kubernets]# kubectl describe po calico-node-bvvhc -n kube-system Events: Type Reason Age From Message ---- ------ ---- ---- ------- Normal Scheduled 5m11s default-scheduler Successfully assigned kube-system/calico-node-bvvhc to node03 Warning Failed 119s (x2 over 3m29s) kubelet Failed to pull image "calico/cni:v3.25.0": rpc error: code = Unknown desc = context canceled Normal Pulling 92s (x3 over 5m11s) kubelet Pulling image "calico/cni:v3.25.0" Warning Failed 14s (x3 over 3m29s) kubelet Error: ErrImagePull Warning Failed 14s kubelet Failed to pull image "calico/cni:v3.25.0": rpc error: code = Unknown desc = unexpected EOF Normal SandboxChanged 13s kubelet Pod sandbox changed, it will be killed and re-created. Normal BackOff 1s (x5 over 3m28s) kubelet Back-off pulling image "calico/cni:v3.25.0" Warning Failed 1s (x5 over 3m28s) kubelet Error: ImagePullBackOfff

分析:

Pod事件列表中有“实例拉取镜像失败”事件，报错原因如下。这可能是镜像较大导致的情况。

解决:

?

1 2 3 4 5

# 到报错的机器node03上面手动下载镜像，可以参考其它成功机器的镜像 docker pull calico/cni:v3.25.0   # 然后删除失败Pod kubectl  get pods -n kube-system | grep calico-node-bvvhc   | awk '{print$1}'| xargs kubectl delete -n kube-system pods

查看是否正常，calico开头的容器都是READY

?

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

[root@k8s-master 15:44:05~/kubernets]# kubectl get pods -n kube-system NAME                                     READY   STATUS    RESTARTS      AGE calico-kube-controllers-cd8566cf-cqvxj   1/1     Running   0             45m calico-node-44kgv                        1/1     Running   0             10m calico-node-6f4np                        1/1     Running   1 (23m ago)   45m calico-node-lv4sp                        1/1     Running   0             45m calico-node-sw8r6                        1/1     Running   0             45m coredns-6d8c4cb4d-77n7q                  1/1     Running   0             125m coredns-6d8c4cb4d-bzm57                  1/1     Running   0             125m etcd-k8s-master                          1/1     Running   1 (23m ago)   125m kube-apiserver-k8s-master                1/1     Running   1 (23m ago)   125m kube-controller-manager-k8s-master       1/1     Running   1 (23m ago)   125m kube-proxy-4ft66                         1/1     Running   1 (23m ago)   125m kube-proxy-4lmf2                         1/1     Running   0             78m kube-proxy-thznw                         1/1     Running   0             77m kube-proxy-xnlmm                         1/1     Running   1 (12m ago)   78m kube-scheduler-k8s-master                1/1     Running   1 (23m ago)   125m

查看node是否已经Ready状态

?

1 2 3 4 5

NAME         STATUS   ROLES                  AGE    VERSION k8s-master   Ready    control-plane,master   127m   v1.23.6 node01       Ready    <none>                 79m    v1.23.6 node02       Ready    <none>                 79m    v1.23.6 node03       Ready    <none>                 80m    v1.23.6

补充一点

如果去机器上面也慢的话可以修改一下/etc/docker/daemon.json国内源，然后重启下

?

1 2 3 4

vim /etc/docker/daemon.json   sudo systemctl daemon-reload systemctl restart docker.service