Incident response runbook

When to use

Use this runbook when there is confirmed impact on checkout, webhooks, fiscal issuance, or dashboard access.

Operational sequence

1. 11) Classify severity

   Classify as P1, P2, or P3 based on customer impact and duration.

2. 22) Mitigate first

   Enable containment feature flags before deep root-cause analysis.

3. 33) Communicate status

   Publish internal and external updates every 15 minutes for P1.

4. 44) Recover and validate

   Confirm backlog drain, normal latency, and no fiscal pending state.

5. 55) Postmortem

   Close with RCA, action items, and owner within 24 hours.

Initial communication template

## Incident in progress

- Severity: P1
- Started at: 14:20 UTC
- Impact: intermittent checkout failures
- Mitigation: retry queue fallback enabled
- Next update: 15 minutes

Recovery signals

• Checkout success rate returns to baseline.
• Webhook queue stops growing.
• No new critical errors in fiscal reconciliation.