Incident Management
When to use
Use when responding to production incidents, building incident response processes, writing postmortems, or planning chaos engineering exercises. Covers the full incident lifecycle from detection to follow-through.
Core principles
- Acknowledge fast, resolve carefully — 5 minutes to acknowledge, then methodical triage
- One incident commander — avoids coordination chaos; others execute
- Blameless postmortems — systems fail, not people; blame prevents honest reporting
- Action items with owners — postmortems without follow-up are theater
- Practice failure — chaos engineering finds weaknesses before customers do
References available
references/incident-response.md — Severity levels (SEV1-4), IC role, communication templates
references/postmortem-guide.md — Blameless postmortem template, 5 whys, action item quality, follow-through
references/incident-metrics.md — MTTD, MTTR, MTBF benchmarks, SLO error budget tracking, reliability dashboard
references/chaos-engineering.md — Chaos Monkey, Litmus, failure injection, game day planning
Assets available
assets/incident-template.md — Fill-in incident report template
assets/postmortem-template.md — Fill-in blameless postmortem template