Copy this template when writing a postmortem. File it as a new document in
docs/postmortems/YYYY-MM-DD-<short-title>.md.
Date of incident:
Date of postmortem:
Authors:
Severity:
One paragraph describing what happened, the user impact, and how it was resolved.
- Duration:
- Clusters affected:
- Users affected:
- Functionality lost:
Technical description of the root cause. Reference code, configuration, or infrastructure as appropriate.
How was the incident detected? How long did it take from the start of impact to detection?
Narrative of the response — who did what, in what order. Reference the incident timeline.
| Time (UTC) | Event |
|---|---|
| Action | Type | Owner | Due | Issue |
|---|---|---|---|---|
| prevent/detect/mitigate/process |
Free-form: what systemic improvement does this incident point toward?