Kubernetes Pod stuck in CrashLoopBackOff

Step-by-step Solution

1. 1Run `kubectl describe pod <name>` — read the Events section for OOMKilled, ImagePullBackOff or readiness failures.
2. 2Get logs from the crashed container: `kubectl logs <name> --previous`.
3. 3Check resource limits — if OOMKilled, increase `resources.limits.memory`.
4. 4Verify config: `kubectl get configmap`, `kubectl get secret` — ensure keys the app expects exist.
5. 5Test the image locally: `docker run --rm -it <image>` and reproduce the crash outside K8s.
6. 6If the app needs a warm-up period, add a `startupProbe` with generous `failureThreshold`.
7. 7Once stable, add an alert on `kube_pod_container_status_restarts_total > 3` to catch regressions early.