Checks-to-Owner Responder Map

Route each failing check to the fastest owner path.

Check	Primary Owner	Secondary Owner	First Response
GET /health fails	Runtime On-Call	Platform On-Call	Verify process/pod status, restart if needed
GET /v1/models fails/auth errors	Auth Runtime On-Call	Platform On-Call	Validate API key, provider auth files, refresh path
GET /v1/metrics/providers shows one provider degraded	Platform On-Call	Provider Integrations	Shift traffic to fallback prefix/provider
GET /v0/management/config returns 404	Platform On-Call	Runtime On-Call	Enable remote-management.secret-key, restart
POST /v0/management/auths/{provider}/refresh fails	Auth Runtime On-Call	Provider Integrations	Validate management key, rerun provider auth login
Logs show sustained 429	Platform On-Call	Capacity Owner	Reduce concurrency, add credentials/capacity

Paging Guidelines

1. Page primary owner immediately when critical user traffic is impacted.
2. Add secondary owner if no mitigation within 10 minutes.
3. Escalate incident lead when two or more critical checks fail together.

Related

• Provider Outage Triage Quick Guide
• Auth Refresh Failure Symptom/Fix Table

---

Last reviewed: 2026-02-21
Owner: Incident Commander Rotation
Pattern: YYYY-MM-DD