Kubernetes deployment failure due to invalid command syntax

Note

Kubernetes 1.21 removed the use of --generator flag. DKP 2.x uses a variant of the Kubernetes 1.21.x release. Check your use of the flag if it is required in your deployment and we recommend considering alternatives as per Kubernetes documentation.

Reference: 

• https://github.com/kubernetes/kubernetes/blob/master/CHANGELOG/CHANGELOG-1.21.md
• https://github.com/kubernetes/kubernetes/pull/99732
  

Symptom

Kubernetes POD in "RunContainerError" status after deployment and constantly restarts after deletion

Error Message

Warning  Failed     16s (x3 over 36s)  kubelet, node01    Error: failed to start container "nginx": Error response from daemon: OCI runtime create failed: container_linux.go:345: starting container process caused "exec: \"generator=run-pod/v1\": stat generator=run-pod/v1: no such file or directory": unknown
Warning  BackOff    4s                 kubelet, node01    Back-off restarting failed container

Cause

If you create a pod using the kubectl run command and forget to use "--" in front of generator=run-pod/v1 run will cause a no such file or directory unknown error message.

Resolution

Locate the deployment for the created POD by running kubectl get deployments --all-namespaces

$ kubectl get deployments --all-namespaces
NAMESPACE     NAME      READY   UP-TO-DATE   AVAILABLE   AGE
default       nginx     0/1     1            0           6m7s

Delete the deployment that is failing by running kubectl delete -n NAMESPACE deployment DEPLOYMENT:

$ kubectl delete -n default deployment nginx
deployment.apps "nginx" deleted

Rerun the deployment by running kubectl run --image= --generator=run-pod/v1:

$ kubectl run nginx --image=nginx --generator=run-pod/v1
pod/nginx created