Skip to content

INC-0001: Catalog endpoint latency spike

All times UTC.

TimeEvent
14:02GET /api/v1/catalog P99 latency alert fires
14:05On-call platform engineer acknowledges
14:11Correlated with a deploy at 13:55; rollback initiated
14:16Latency returns to baseline
14:30Incident closed

Elevated latency (P99 ~4s vs. ~200ms baseline) on the catalog endpoint for ~14 minutes. No errors, no data loss; Tidepool catalog refreshes were slow but not failing.

Automated P99 latency alert (see Patch API for the service this alert covers).

On-call platform engineer followed the Rollback Runbook after confirming deploy correlation.

Example: a new catalog query in the 13:55 deploy was missing an index used by the previous query shape.

Example: the query change wasn’t covered by a load test that would have caught the missing index before production.

Rollback to the prior deploy resolved the symptom immediately.

ActionOwnerDueStatus
Add the missing indexplatformexample
Add a load test for catalog query shape changesplatformexample

See postmortem.