Walk through the SRE incident-management lifecycle.
Quick Answer
Alert → Detect → Respond → Mitigate → Recover → RCA (root-cause analysis) → Postmortem. Mitigate first to stop customer impact, then find and fix the root cause, then a blameless postmortem with tracked actions to prevent recurrence.
Detailed Answer
Monitoring generates an alert; responders detect and acknowledge, then mitigate to restore service (rollback, failover, scale) before chasing root cause. After recovery, do RCA (e.g., 5 Whys) and a blameless postmortem documenting timeline, contributing factors, and owned action items. The goal is to reduce impact fast and prevent recurrence, not to assign blame.
Interview Tip
Stress mitigate-before-root-cause and the blameless postmortem with tracked actions — that ordering is what interviewers want.