Kubelet stopped posting node status

General Discussions
1

Cluster information:

Kubernetes version: v1.14.2
Cloud being used: proxmox cluster
Installation method: kubeadm
Host OS: ubuntu 18.04
CNI and version: 0.3.0

HI ! Two of my cluster nodes gets Kubelet stopped posting node status in kubectl describe node sometimes. In logs of that nodes i see this:

Dec 11 12:01:03 alma-kube1 kubelet[946]: E1211 06:01:03.166998     946 controller.go:115] failed to ensure node lease exists, will retry in 6.4s, error: Get https://192.168.151.52:6443/apis/coordination.k8s.io/v1beta1/namespaces/kube-node-lease/leases/alma-kube1?timeout=10s: read tcp 192.168.170.7:46824->192.168.151.52:6443: use of closed network connection
Dec 11 12:01:03 alma-kube1 kubelet[946]: W1211 06:01:03.167045     946 reflector.go:289] object-"kube-public"/"myregistrykey": watch of *v1.Secret ended with: very short watch: object-"kube-public"/"myregistrykey": Unexpected watch close - watch lasted less than a second and no items received
Dec 11 12:01:03 alma-kube1 kubelet[946]: W1211 06:01:03.167356     946 reflector.go:289] object-"kube-system"/"kube-router-token-bfzkn": watch of *v1.Secret ended with: very short watch: object-"kube-system"/"kube-router-token-bfzkn": Unexpected watch close - watch lasted less than a second and no items received
Dec 11 12:01:03 alma-kube1 kubelet[946]: W1211 06:01:03.167418     946 reflector.go:289] object-"kube-public"/"default-token-kcnfl": watch of *v1.Secret ended with: very short watch: object-"kube-public"/"default-token-kcnfl": Unexpected watch close - watch lasted less than a second and no items received
Dec 11 12:01:13 alma-kube1 kubelet[946]: E1211 06:01:13.329262     946 kubelet_node_status.go:385] Error updating node status, will retry: failed to patch status "{\"status\":{\"$setElementOrder/conditions\":[{\"type\":\"MemoryPressure\"},{\"type\":\"DiskPressure\"},{\"type\":\"PIDPressure\"},{\"type\":\"Ready\"}],\"conditions\":[{\"lastHeartbeatTime\":\"2019-12-11T06:01:03Z\",\"type\":\"MemoryPressure\"},{\"lastHeartbeatTime\":\"2019-12-11T06:01:03Z\",\"type\":\"DiskPressure\"},{\"lastHeartbeatTime\":\"2019-12-11T06:01:03Z\",\"type\":\"PIDPressure\"},{\"lastHeartbeatTime\":\"2019-12-11T06:01:03Z\",\"type\":\"Ready\"}]}}" for node "alma-kube1": Patch https://192.168.151.52:6443/api/v1/nodes/alma-kube1/status?timeout=10s: net/http: request canceled (Client.Timeout exceeded while awaiting headers)
2

I saw the “Kubelet stopped posting node status” while studying.
Possibly kubelet process on the node is not working fine. Approach would be as the below.

  • ssh to the node
  • systemctl status kubectl <== Check service status
  • systemctl restart kubectl <== Try restart the service
  • journalctl -u kubectl <== If restart fails, check service log to find more detail for the failure, fix, restart
3

I think it should be kubelet not kubectl, I corrected the command as below.
systemctl status kubelet
systemctl restart kubelet
journalctl -u kubelet