The Endgame of SRE
summary
A production deploy includes a schema rename that schema migration fails mid-deploy; the CI/CD system rolls back but the half-applied change leaves another service throwing errors until a DBA manually restores table names. Customer-facing SLOs error budget tripped via SLOs detect the outage. A developer adds detection to the pipeline and added schema check to CI/CD pipeline so future migrations fail fast; the fix ships before the retro, proving the socio-technical gap — not the tooling — caused most harm.
AI-written summary · about
chapters (8)
- 0:00Introduction and Scenario
- 1:54Meet the Teams and Incident
- 5:11Investigating Team Disco
- 9:13Organizational Pressures
- 13:18Toxic Environments
- 18:36The Incident Story
- 21:50Root Cause Analysis
- 27:02The End Game of SRE
related talks
26mMission Critical Mindset Building a Culture of Resilience in DevOps Teams
43mSRE From Inception to Global Scale
29mOvercoming SRE Anti-Pattern Roadblocks: Rebranding the Ops Team
25mPitching SRE to your Senior Executives
45mClearing the Way for SRE in the Enterprise
47mPersistent SRE Antipatterns: Pitfalls on the Road to Creating a Successful
29mThe Curious SRE: Cultivating a Mindset for System Resilience
49mWhat SRE Could Be: Systems Reliability Engineering