Self-hosted docs
Runbooks
Structured response procedures for common incident scenarios.
Incident classification
Classify incidents by severity, impact, and affected components to trigger appropriate response workflows.
- Example: Severity levels: P0 (critical), P1 (high), P2 (medium), P3 (low)
- Example: Impact areas: runtime, policy, provider, security
- Example: Classification triggers specific runbook workflows
Response workflows
Step-by-step procedures for containing, diagnosing, and resolving different incident types.
- Example: Runtime degradation runbook: check traces, verify provider status, trigger failover
- Example: Policy violation runbook: review verdicts, adjust thresholds, update documentation
- Example: Provider outage runbook: activate fallback, monitor recovery, restore primary