Pod status explainer
Paste kubectl describe pod and find out why it is in CrashLoopBackOff, ImagePullBackOff, Pending or OOMKilled, and what to run next.
Runs locally. Nothing you paste leaves this tab.
What to paste
kubectl describe pod gives the best answer because it includes recent events: scheduler messages, image pull errors and probe failures. YAML or JSON from kubectl get pod -o yaml works too, and so does a list of pods.
kubectl describe pod <pod> -n <namespace>
kubectl get pods -n <namespace> | grep -v Running
kubectl get pods -n <namespace> -o yaml > pods.yamlDescribe output contains environment variables, which may include secrets. It stays in this tab, but run it through the redactor before sharing it anywhere else.
Exit codes
| Code | Meaning |
|---|---|
| 0 | Finished successfully. In a Deployment this still causes restarts, because it expects the process to keep running. |
| 1 | Application error. Read the logs. |
| 126 / 127 | Command cannot be executed / not found in the image. |
| 137 | SIGKILL: out of memory (OOMKilled), or killed after a failed liveness probe or shutdown timeout. |
| 139 | Segmentation fault. |
| 143 | SIGTERM: a normal shutdown, for example during a rollout. |