Clear the ring for Chaos Engineering at Vertrieb Deutsche Bahn!
summary
An engineering group at DB Vertrieb runs chaos engineering for over a year and documents 80+ game days that start in staging and graduate to production; initial tools are Chaos Monkey for Kubernetes, Pumba and a Chaos Monkey for Spring Boot while evaluating Steadybit and Greed (Gremlin), and they insist on APM (Dynatrace/Instana). Early experiments cause a 100% error rate for 50 minutes, teach planning, baselines, blameless postmortems and starting small.
AI-written summary · about
chapters (19)
- 0:00Introduction
- 1:05What is Chaos Engineering
- 6:55The Show
- 10:10Comfort Checkin
- 12:10Blameless Postmortem
- 13:20Ideas
- 13:50New tests
- 15:00What would Netflix do
- 16:30The Chaos Monkey
- 17:35We didnt do our homework
- 18:30Chaos Engineering
- 19:10How to do Chaos Engineering
- 22:50Tools
- 24:30Messing up the test plan
- 25:30Technical gaps
- 26:15No good experiments
- 29:40Game days
- 31:40Top learnings
- 35:40Questions
related talks
52mGetting started with Chaos Engineering
27mChaos Engineering: The Path to Reliability
18mChaos Engineering: Break It On Purpose
49mChaos Engineering: Withstanding Turbulent Conditions in Production
08mChaos Engineering at Citi Bank
26mBreaking Systems for Resilience
46mDesigning a System for Chaos
37mChaos Engineering: The Fine Art of Breaking Stuff in Production on Purpose