Troubleshooting and Incident Response Questions

Covers systematic diagnosis and resolution of production incidents and live failures across infrastructure and applications. Candidates should demonstrate how they detect and triage incidents, gather and interpret logs and metrics, and use command line and diagnostic tools such as kubectl, journalctl and tcpdump. The topic includes isolating and reproducing faults, mitigating impact with short term workarounds, implementing fixes, and driving root cause analysis and blameless postmortems. Common scenarios include container image pull failures, CrashLoopBackOff, pods not receiving traffic, services becoming unreachable, deployment rollbacks and latency spikes. Candidates should also explain communication and escalation practices, runbook usage, rollback versus fix forward decisions, and lessons learned that reduce mean time to recovery.