Production Incident Checklist

First 10 minutes

• Confirm the incident and affected service.
• Identify customer impact.
• Assign an incident owner.
• Check dashboards, alerts, logs, and recent changes.
• Determine whether the issue is isolated or widespread.
• Stop unsafe changes and consider rollback or failover.

During mitigation

• Keep one clear hypothesis at a time.
• Record important observations and actions.
• Prefer reversible changes.
• Communicate impact and status clearly.
• Do not confuse a workaround with root cause.

After recovery

• Verify the service from the user's perspective.
• Preserve relevant logs and evidence.
• Identify root cause and contributing factors.
• Create prevention actions with owners.
• Update the runbook so the next incident is faster to resolve.