Pragmatic Incident Response: 5 lessons learned from failures
summary
Five concrete takeaways: start with small incidents and run quick retros so fixes actually happen; measure and drive down mean time to retro and track the incidents-to-retros ratio; alert on customer pain via SLOs—alert on symptoms not causes—instead of paging on CPU or disk; keep incidents to focused tiger teams and use push updates for others; assign clear owners and publish a service catalog so the right team gets routed immediately.
AI-written summary · about
chapters (9)
- 0:00Pragmatic incident response
- 1:17Small incidents reveal problems
- 5:53Focus on small incidents
- 7:41Mean time to retro
- 11:15Alert on customer pain
- 15:26Limit incident notifications
- 19:42Service ownership before outages
- 22:51Five pragmatic tips
- 24:54Closing and resources
related talks
15mPragmatic Incident Response: Lessons Learned from Failures
30mHow Did Things Go Right? Learning More from Incidents
37mIncidents are a new normal
03mRecap
45mFailure Happens: Improving Incident Response in Large-Scale Organizations
45mIncident Management - Talk the Talk, Walk the Walk
31mThe Art of Turning Incidents into Opportunities
13mThe Incident Is The Way: Using Your Incidents to Win...