Mitigation Before Root Cause¶

During incidents, restoring service takes priority over understanding cause. Rollback, restart, and failover restore user experience; diagnosis can wait.

Tension Addressed¶

Automation vs. Human Judgment - mitigation is often automatable while root cause requires human analysis

Recognizable As¶

Incident runbooks prioritizing restoration; "rollback first, investigate later"; separation of incident response from post-mortem; MTTR focus

Canonical Sources¶

Google SRE Book, Chapter 14 "Managing Incidents" (2016); PagerDuty Incident Response documentation

Transfers To¶

Emergency medicine (stabilize before diagnose), Firefighting (suppress before investigate), Crisis management (contain before analyze)

The instinct to understand before acting costs user-facing impact time. A 5-minute rollback that restores service is better than 30 minutes of diagnosis while users suffer.