Skip to content

Runbooks — pointer

This is a pointer page. The actual runbook artefacts live in two places:

  • Symptom-keyed cheat sheets → ../runbooks/ — 14 short runbooks, each following the 5-section template (Symptom · Likely cause · Diagnosis · Fix · Prevention). Index at ../runbooks/README.md.
  • First-response procedure (severity rubric, first-5-minute checklist, escalation tree, comms templates) → ../handover/oncall-runbook.md. Read this if you've just been paged.

For the escalation rules (severity × time → who's notified), see ../handover/escalation-matrix.md.

For how to author a new runbook, see ../runbooks/README.md §"Adding a new runbook".

  • observability.md — telemetry pipeline, Grafana dashboards, log queries.
  • performance.md — perf-test methodology surfaces the conditions runbooks above respond to.
  • ../adr/ — design decisions that explain why some bugs can't be runbook'd (e.g. ADR-0005 is "expected behaviour" not an incident).