SEV Culture: How Today's SEVs Create Tomorrow's Reliability
summary
Claim: filing more SEVs improves reliability, supported by cross‑industry studies showing airlines and construction sites with more incident reports have fewer fatalities or injuries. Built tooling: a minimal Sev form with false positive mode, an escalation panel tied to on‑call rotations and IM on‑call roles, and a one‑click unland button to revert production changes. Process: weekly SevReview uses DERP reports; mains review averages 12 questions and 450 unique individuals, keeping reviews blameless and actionable.
AI-written summary · about
chapters (14)
- 0:00SEV culture at Meta
- 1:29Filing SEVs improves reliability
- 2:54Failures in aviation lessons
- 4:58Consequences of not filing incidents
- 6:01Incentives for incident reporting
- 7:26Reliability in large-scale systems
- 7:47SEV life cycle at Meta
- 8:50Meta’s SEV philosophy
- 9:54SEV escalation process
- 11:20SEV mitigation strategies
- 12:01SEV review process
- 13:27SEV review culture
- 15:12Blameless incident management
- 16:36SEV process challenges
related talks
23mSite Reliability Evangelism: Practice Start-up within an Established...
42mSite Reliability Engineering as Systems Thinking
49mWhat SRE Could Be: Systems Reliability Engineering
27mThe Perverse Incentives of Reliability
51mThe Evolution of Site Reliability Engineering
45mClearing the Way for SRE in the Enterprise
28mEvolution of SRE and Rising Need of SRE Catalyzers
04mThe SRE I aspire to be