Kubernetes CoreDNS 状态是 CrashLoopBackOff 报错

系统管理员 2023-08-17 15:16 145阅读 0赞

查看状态的时候,遇见coredns出现crashlookbackoff,首先我们来进行排错，不管是什么原因，查看coredns的详细信息，以及logs

\[root@k8s-master coredns\]\# kubectl get pod,svc,deployment,rc -n kube-system  
NAME                           READY   STATUS             RESTARTS   AGE  
pod/coredns-5bd5f9dbd9-h22lf   0/1     CrashLoopBackOff   106        20h  
  
NAME               TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)         AGE  
service/kube-dns   ClusterIP   10.0.0.2     <none>        53/UDP,53/TCP   18h  
  
NAME                            READY   UP-TO-DATE   AVAILABLE   AGE  
deployment.extensions/coredns   0/1     1            0           20h  
\[root@k8s-master coredns\]\# kubectl describe pod coredns-5bd5f9dbd9-h22lf -n kube-system  
Name:           coredns-5bd5f9dbd9-h22lf  
Namespace:      kube-system  
Priority:       0  
Node:           192.168.30.23/192.168.30.23  
Start Time:     Tue, 06 Aug 2019 13:47:23 +0800  
Labels:         k8s-app=kube-dns  
                pod-template-hash=5bd5f9dbd9  
Annotations:    seccomp.security.alpha.kubernetes.io/pod: docker/default  
Status:         Running  
IP:             172.17.87.2  
Controlled By:  ReplicaSet/coredns-5bd5f9dbd9  
Containers:  
  coredns:  
    Container ID:  docker://c02395208c9763d5061e3478def972f60274fd2a98f5fdd3bd6fbe0c542f39bb  
    Image:         coredns/coredns:1.2.2  
    Image ID:      docker-pullable://coredns/coredns@sha256:782834359db4711f7f6d85be8adb58490c1c0798e027b9d9c9916533d8911617  
    Ports:         53/UDP, 53/TCP, 9153/TCP  
    Host Ports:    0/UDP, 0/TCP, 0/TCP  
    Args:  
      -conf  
      /etc/coredns/Corefile  
    State:          Waiting  
      Reason:       CrashLoopBackOff  
    Last State:     Terminated  
      Reason:       Error  
      Exit Code:    137  
      Started:      Wed, 07 Aug 2019 04:38:29 +0800  
      Finished:     Wed, 07 Aug 2019 04:40:48 +0800  
    Ready:          False  
    Restart Count:  106  
    Limits:  
      memory:  170Mi  
    Requests:  
      cpu:        100m  
      memory:     70Mi  
    Liveness:     http-get http://:8080/health delay=60s timeout=5s period=10s \#success=1 \#failure=5  
    Environment:  <none>  
    Mounts:  
      /etc/coredns from config-volume (ro)  
      /var/run/secrets/kubernetes.io/serviceaccount from coredns-token-sv6tq (ro)  
Conditions:  
  Type              Status  
  Initialized       True  
  Ready             False  
  ContainersReady   False  
  PodScheduled      True  
Volumes:  
  config-volume:  
    Type:      ConfigMap (a volume populated by a ConfigMap)  
    Name:      coredns  
    Optional:  false  
  coredns-token-sv6tq:  
    Type:        Secret (a volume populated by a Secret)  
    SecretName:  coredns-token-sv6tq  
    Optional:    false  
QoS Class:       Burstable  
Node-Selectors:  <none>  
Tolerations:     CriticalAddonsOnly  
                 node-role.kubernetes.io/master:NoSchedule  
                 node.kubernetes.io/not-ready:NoExecute for 300s  
                 node.kubernetes.io/unreachable:NoExecute for 300s  
Events:  
  Type     Reason     Age                     From                    Message  
  ----     ------     ----                    ----                    -------  
  Normal   Pulled     6h31m (x89 over 13h)    kubelet, 192.168.30.23  Container image "coredns/coredns:1.2.2" already present on machine  
  Warning  Unhealthy  5h21m (x515 over 13h)   kubelet, 192.168.30.23  Liveness probe failed: HTTP probe failed with statuscode: 503  
  Warning  BackOff    5h11m (x1223 over 13h)  kubelet, 192.168.30.23  Back-off restarting failed container  
这里可以根据自己的原因来解决，其实原因不过两个，可能是你的容器lo回环找不到你的容器，有可能就是你的本地/etc/resolv.conf下的dns  
\[root@k8s-master coredns\]\# vim /etc/resolv.conf

这里我指定是nameserver 8.8.8.8

如果他接收了本地的接口,也有可能会出现coredns出现off状态，有其他问题我们继续讨论  
\[root@k8s-master coredns\]\# systemctl restart docker  
\[root@k8s-master coredns\]\# kubectl get pod,svc,deployment,rc -n kube-system  
NAME                           READY   STATUS    RESTARTS   AGE  
pod/coredns-5bd5f9dbd9-h22lf   1/1     Running   107        20h  
  
NAME               TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)         AGE  
service/kube-dns   ClusterIP   10.0.0.2     <none>        53/UDP,53/TCP   18h  
  
NAME                            READY   UP-TO-DATE   AVAILABLE   AGE  
deployment.extensions/coredns   1/1     1            1           20h

转载于:https://www.cnblogs.com/zc1741845455/p/11313476.html