A container is reported in an abnormal state when its status is Waiting (e.g. ImagePullBackOff, CrashLoopBackOff, ContainerCreating) or Terminated with a non-zero exit code or error reason. The report includes the exact reason from the API so you can target fixes.
Warning
N/A
- Report shows: Pod has container in state with a reason (e.g. ImagePullBackOff, ErrImagePull, CrashLoopBackOff)
kubectl describe podshows container state Waiting or Terminated with a reason/message
- Inspect pod and events:
kubectl describe pod <ns>/<name> - For ImagePullBackOff / ErrImagePull: fix image name, tag, or pull secrets; check network and registry access
- For CrashLoopBackOff: check logs and previous logs; fix application, probes, or resources
- For ContainerCreating: usually transient; if it persists, check volume mounts, secrets, or node resources
- For Terminated with non-zero exit code: use exit code and message to fix the application or configuration