[K8S集群:初始化异常]：kubelet cgroup driver: \“systemd\“ is different from docker cgroup driver: \“cgroupfs\

最新推荐文章于 2024-09-13 16:50:19 发布

达不溜001

最新推荐文章于 2024-09-13 16:50:19 发布

阅读量799

点赞数 5

分类专栏： # Kubernetes-K8S 文章标签： kubernetes kubelet docker

本文链接：https://blog.csdn.net/Abraxs/article/details/137658299

版权

Kubernetes-K8S 专栏收录该内容

14 篇文章 0 订阅

订阅专栏

文章描述了一种在Kubernetes集群中，Calico-node遇到异常的情况，涉及主节点和从节点的日志分析、docker和kubelet服务重启，以及calico.yaml文件的编辑。主要解决步骤包括检查kubelet错误、重启docker和kubelet服务，并调整calico配置以匹配网络接口。

摘要由CSDN通过智能技术生成

文章目录

零：场景复现：pod：calico-node异常
一：查看异常日志：主从节点[主要从节点]：journalctl -f -u kubelet
- 1.1：主节点：journalctl -f -u kubelet
- 1.2：从节点：journalctl -f -u kubelet
三：解决方式，主要是在主节点重启docker和k8s服务

零：场景复现：pod：calico-node异常

主节点：安装coredns -> init初始化主节点（此时还没有安装calico）
从节点：基于主节点生成join命令加入集群
主节点：安装calico：apply 生成pod，此时没有调整yaml网卡
coredns 和calico pod 运行成功
但是 calico-node-cl8f2 运行失败

一：查看异常日志：主从节点[主要从节点]：journalctl -f -u kubelet

1.1：主节点：journalctl -f -u kubelet

[root@sv-master ~]# journalctl -f -u kubelet
-- Logs begin at 四 2024-04-11 15:17:47 CST. --
4月 11 15:30:47 sv-master kubelet[12537]: I0411 15:30:47.319143   12537 scope.go:110] "RemoveContainer" containerID="b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363"
4月 11 15:30:47 sv-master kubelet[12537]: E0411 15:30:47.319537   12537 remote_runtime.go:572] "ContainerStatus from runtime service failed" err="rpc error: code = Unknown desc = Error: No such container: b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363" containerID="b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363"
4月 11 15:30:47 sv-master kubelet[12537]: I0411 15:30:47.319603   12537 pod_container_deletor.go:52] "DeleteContainer returned error" containerID={Type:docker ID:b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363} err="failed to get container status \"b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363\": rpc error: code = Unknown desc = Error: No such container: b10912f438c5592e65aed26059c91724957660371ca43ee49327ccee20173363"
4月 11 15:30:47 sv-master kubelet[12537]: E0411 15:30:47.555224   12537 remote_runtime.go:479] "StopContainer from runtime service failed" err="rpc error: code = Unknown desc = Error response from daemon: No such container: e560e71cb9fa3911d73ac67fa1a8497d6793542860730be4749d794f9707ed26" containerID="e560e71cb9fa3911d73ac67fa1a8497d6793542860730be4749d794f9707ed26"

1.2：从节点：journalctl -f -u kubelet

[root@sv-slaver-on ~]# journalctl -f -u kubelet
4月 11 15:06:21 sv-slaver-one kubelet[5050]: E0411 15:06:21.113081    5050 server.go:302] "Failed to run kubelet" err="failed to run Kubelet: misconfiguration: kubelet cgroup driver: \"systemd\" is different from docker cgroup driver: \"cgroupfs\""
4月 11 15:06:21 sv-slaver-one systemd[1]: kubelet.service: main process exited, code=exited, status=1/FAILURE
4月 11 15:06:21 sv-slaver-one systemd[1]: Unit kubelet.service entered failed state.
4月 11 15:06:21 sv-slaver-one systemd[1]: kubelet.service failed.

三：解决方式，主要是在主节点重启docker和k8s服务

3.1：执行命令

[root@10 docker]# systemctl restart docker
[root@10 docker]# systemctl daemon-reload
[root@10 docker]# systemctl restart kubelet

3.2：编辑calicoyaml文件网卡

 # Cluster type to identify the deployment type
 - name: CLUSTER_TYPE
   value: "k8s,bgp"
 # 主要添加下面两行
 - name: IP_AUTODETECTION_METHOD
   value: "interface=enp0s3"

在这里插入图片描述

3.3：再次查看nodes和pod

sv-slaver-one为主节点，配反了

[root@sv-slaver-one ~]# kubectl get pods -A
NAMESPACE     NAME                                       READY   STATUS    RESTARTS        AGE
kube-system   calico-kube-controllers-64cc74d646-4z8zf   1/1     Running   2 (9h ago)      9h
kube-system   calico-node-cl8f2                          1/1     Running   0               9h
kube-system   calico-node-pfxnd                          1/1     Running   0               9h
kube-system   coredns-6d8c4cb4d-8q7tb                    1/1     Running   2 (9h ago)      9h
kube-system   coredns-6d8c4cb4d-m2gz2                    1/1     Running   2 (9h ago)      9h
kube-system   etcd-sv-slaver-one                         1/1     Running   3 (9h ago)      9h
kube-system   kube-apiserver-sv-slaver-one               1/1     Running   3 (9h ago)      9h
kube-system   kube-controller-manager-sv-slaver-one      1/1     Running   4 (3h59m ago)   9h
kube-system   kube-proxy-6kfnf                           1/1     Running   2 (9h ago)      9h
kube-system   kube-proxy-s9pzm                           1/1     Running   3 (9h ago)      9h
kube-system   kube-scheduler-sv-slaver-one               1/1     Running   4 (3h59m ago)   9h
[root@sv-slaver-one ~]# kubectl get nodes
NAME            STATUS   ROLES                  AGE   VERSION
sv-master       Ready    <none>                 9h    v1.23.0
sv-slaver-one   Ready    control-plane,master   9h    v1.23.0
[root@sv-slaver-one ~]#