Incidents

INC-2047 — Database CPU usage

orders-db-primary sustained CPU above 90% during peak order volume.

highidentifiedorders-service32m

Timeline

Every state change, automatically captured

  1. 12:10Alert triggered

    CPU above 85% for 5 minutes.

  2. 12:14Engineer assigned

    Marco Silva acknowledged the page.

  3. 12:26Cause identified

    Unindexed query on orders.status.

AI analysis

Correlated from metrics, logs and deployments

A full table scan on orders.status accounts for 61% of database time. Adding a partial index on pending orders should cut CPU by an estimated 30-40%.

Metrics

p95 latency around the incident window

Related logs

2 matching entries

12:42:04WARNSlow query detected: SELECT * FROM orders WHERE status = 'pending' (2418ms)
12:38:20ERRORDeadlock detected on orders_items, transaction rolled back

Incident notes

Shared with everyone on the response

No notes yet.

Incident IDINC-2047
Serviceorders-service
AssigneeMarco Silva
Created32 minutes ago
Duration32m

Related deployments

deploy-1839In Progress

v3.4.1 · c19b402 · 8 min ago

Resolution summary

Available once the incident is resolved.