Pod restart failure due to liveness probe time-out
As part of liveness probe for primary and media pods, a health script runs inside the container to check the NetBackup health status.
When there is an issue with a container related to a full disk, CPU, or memory pressure, the liveness probe gets timed out because of no response from the health script. As a result, the Pod does not restart.
To resolve this issue, restart the Pod manually. Delete the Pod using the kubectl delete pod/<podname> -n <namespace> command.
The Pod is deleted and Kubernetes creates another Pod.