Back to Docs
Runbook

Runbook Template

Standardized incident-response runbook for Meridian operators. Copy, fill, and keep current.

1. Metadata

Title: ______________________________

Owner: ______________________________

Severity: [ ] Sev1 [ ] Sev2 [ ] Sev3

Last reviewed: ______________________

2. Triggers

  • Alert source (Datadog / PagerDuty / manual): ________
  • Threshold / condition: ______________________________
  • Expected false-positive rate: _______________________

3. Immediate Actions

  1. Acknowledge alert in #incidents Slack channel.
  2. Verify scope — single-tenant or multi-tenant impact.
  3. If Sev1: page on-call via /page command.
  4. Start incident timer; open shared doc.

4. Diagnosis

Checklist:

[ ] Vercel deployment status

[ ] Upstash KV latency / error rate

[ ] KeyAuth licensing endpoint health

[ ] CDN asset availability

[ ] Discord bot connectivity

5. Mitigation

Rollback command: ______________________________

Feature-flag kill switch: ________________________

Traffic drain procedure: _________________________

Manual failover steps: ___________________________

6. Resolution & Postmortem

  • Confirm metrics returned to baseline.
  • Close incident timer; record duration.
  • Draft postmortem within 48 hours.
  • File action items as GitHub issues.

Keep this runbook in your team wiki or Notion. Review quarterly. For live examples, see the Meridian docs.