PASSPORT · WHAT THIS ROOM DID
Why does the same outage keep recurring?
Every line is from the record. Nothing is computed for this page, and nothing is smoothed: a short list of disagreements means a room that agreed, not a room whose disagreements were tidied away.
| structures | obstacles |
| opened | 08 Sep 2026 |
| who answered | 30 people in 2 group(s) · participant: 30 |
| wrote blind | not recorded |
| claims | 30 written, 20 structured |
| links certified | 20 |
| rooms on this question | 1 · does not count toward the shared map |
| decided at | 75% of the votes cast |
| sealed | yes, members forgotten |
What it rests on- The service map in the wiki is two years old
- One engineer knows how the payment path actually works
- Feature flags are never cleaned up, so nobody knows which paths are live
- The on-call rota has the same two people on it most weeks
- Rollbacks take longer than the outage they are meant to end
- Alerts fire so often that on-call mutes them
What it could not settlePublished as contested, never recorded as a no.
- none: every pair the room reached was settled
What was predicted to followWritten after the map settled. Nobody voted on it.
- No production deploy after 14:00 on Fridays unless the on-call engineer agrees — expected: the same cause stops coming back once fixing it is in the sprint — due 23 Jan 2027
What the room would doEach action judged against each root cause, at three quarters of the votes cast.
- No production deploy after 14:00 on Fridays unless the on-call engineer agrees
split: One engineer knows how the payment path actually works - Give the shared schema one named owner and a migration tool with rollback
reduces: One engineer knows how the payment path actually works - Reserve one engineer-week per sprint for post-mortem actions, first in the sprint, not last
reduces no root cause
← the dialogue