Oncall Runbook
Severity Levels
- Sev-1: Customer-facing outage, escalate immediately
- Sev-2: Degraded service, page on-call lead
- Sev-2.5: Business-hours Sev-2, monitor during work hours
- Sev-3+: Track and fix in normal sprint
Triage Steps
- Check alarm tree for root cause
- Search recent deployments for correlation
- Check ticket history for known issues
- Escalate if unresolved after 15 minutes