Kubernetes CrashLoopBackOff

Root Cause & Remediation

Bad config map, missing secret, failing liveness probe, OOM kill, or an image regression deployed via CI.

Remediation steps

1. 1kubectl describe pod and inspect the last termination reason and exit code.
2. 2Check kubectl logs --previous for the crashing container.
3. 3Roll back the deployment to the last known-good image SHA.
4. 4Validate config maps and secrets exist in the target namespace.
5. 5Tune liveness/readiness probe timeouts and resource limits.

DORA Risk Matrix

Likelihood

High

Blast radius

Limited to a single deployment unless it is a shared platform component.

CIF impact

Variable. Catastrophic if the workload is on the critical path.

Analyst notes

Usually contained by rolling back, but ingress controllers or auth services elevate this to MAJOR fast.