Search code examples
kuberneteskubectl

Why is the pod is showing running even when the node is down?

The node named Kube worker node is down for more than 12 hour, but the pod is showing running. Why kubernetes did not updated in that much of time ?

# kubectl get nodes
NAME               STATUS     ROLES                  AGE   VERSION
kube-worker-node   NotReady   <none>                 17h   v1.20.2

pix:/home/ciasto# kubectl get pods -o wide
NAME                     READY   STATUS    RESTARTS   AGE   IP               NODE               NOMINATED NODE   READINESS GATES
flask-776b878dbc-krtd6   1/1     Running   0          17h   192.168.254.2    kube-worker-node   <none>           <none>
redis-596bbd776-q6s2n    1/1     Running   0          15h   192.168.42.130   rockpi             <none>           <none>
Solution

  • This could be because of missing node.kubernetes.io/unreachable:NoExecute taint on the node. You can check it using bellow command

    kubectl describe nodes kube-worker-node  | grep Taints

    Please check the related issue in github here