Why Your Kubernetes Pod Is in CrashLoopBackOff (and How to Find Out)

CrashLoopBackOff means Kubernetes started your container, the container exited, and Kubernetes is waiting before trying again. The real question is why the container exits.

Step 1: Describe the pod

kubectl describe pod <pod-name> -n <namespace>

Check Last State, the exit code and the Events section. Exit code 137 usually indicates the container was killed for exceeding its memory limit (OOMKilled); exit code 1 usually means the application itself failed.

Step 2: Read the previous container's logs

kubectl logs <pod-name> -n <namespace> --previous

The --previous flag matters: it shows logs from the container that crashed, not the one that is just starting.

Step 3: Check the usual suspects

• Missing or misspelled environment variables, ConfigMaps or Secrets
• The application cannot reach a database or dependency
• Liveness probe failing because the app starts slowly
• Memory limit too low for the workload
• Wrong command or entrypoint in the image

Step 4: Reproduce outside the cluster

Run the same image locally with the same environment variables. If it fails the same way, you have a fast feedback loop for fixing it.

Still stuck? Our Kubernetes job support can help you work through it live.