kb/k8s_crashloop.md

K8s CrashLoopBackOff

Symptoms

• Pod restarts repeatedly with CrashLoopBackOff
• Health checks failing or process exits quickly
• Logs end abruptly after startup

Checks

• Inspect last container logs before restart (kubectl logs -p)
• Validate readiness/liveness probes and command/args
• Confirm config/secret mounts paths and permissions
• Check for missing env vars or failing dependency endpoints
• Review recent image/config deploys

Fix

• Correct probes or increase initial delays
• Fix missing configuration or credentials
• Add retries/backoff around external dependencies
• Roll back to last known good release if needed