#observability

Logs, traces, uptime

1. Time-to-first-accepted-answer as a real SLO

   Support volume looks healthy until you chart hours from question to `is_solved`. Last month p50 was 6 hours and p90 was 3 days — almost all of that p90 was weekend campus tickets...

2. Which community metric belongs on the homepage

   We shipped a dashboard that put peak concurrent users in 48px type. Staff started optimizing for announcement blasts instead of accepted answers. We demoted the spike number and p...

3. Public incident posts that people actually finish reading

   The template we use internally: 1. What members saw. 2. What was actually down (API, search, or uploads). 3. When it started and ended in UTC. 4. What we changed. 5. What we will ...

4. What we learned from 90 days of community SLOs

   SLOs we actually page on: - API 5xx under 0.5% - Search p95 under 120ms - Upload 2xx under 2s for files ≤ 5 MB (5_000_000 bytes) We do not page on posts per minute. That is a vanity chart. On-call is for 5xx, not for unread comments.

5. Uploads fail around 4.2 MB even though the limit is 5 MB

   The client sends 4_200_000 bytes and Nginx returns 413. The in-app copy says 5 MB (5_000_000 bytes). Is there a reverse-proxy `client_max_body_size` we missed besides the Axum cap...

6. Critical issue notifications go to watching members, not on-call

   Severity=`critical` still uses the same subscription table. Staff expected PagerDuty. Document that TalkMoro is not an incident pager, or add a webhook example for `topic.severity...

7. Bulk close stale Issues older than 90 days

   The mod queue is full of `triaged` bugs with no repro. A dry-run report plus one-click `wont_fix` would help. Do not delete — lock and keep the permalink.

8. Scheduled maintenance: search reindex on 12 Sep 2026

   Between 04:00 and 06:00 UTC on 12 September 2026 we will rebuild GIN indexes. Reads stay up. Writes to topics may delay about 2 seconds. The status page will flip to maintenance. Search results can look stale for a few minutes after the win

9. Incident closed: 22-minute API timeout in us-east

   A connection pooler restart stacked with a deploy. p99 exceeded 8s. Members could read cached topics. Votes queued. Full write-up is the linked Guide. Started 14:11 UTC, mitigated 14:33 UTC, 31 August 2026.

10. How to file an Issue staff can reproduce

    1. Choose Issue and a severity. 2. Write numbered repro steps. 3. Include version if it is software ( entity_name , and the related Release). 4. Attach logs as a fenced code block, not a screenshot of a screenshot. 5. Leave the thread open