Let's Get Ready for Chaotic Engineering
summary
Argues that chaotic engineering only works after deliberate prep: map processes, prioritize system issues into standalone projects, define a steady state and raise a fragile normal into a resilient one before injecting faults. Starts experiments small — one host one container — using tools like chaos monkey (then latency and janitor monkeys), ties tests to monitoring, postmortems and incident reviews, enforces containment and blast radius, and fixes discovered bugs so experiments actually improve stability.
AI-written summary · about
chapters (12)
- 0:00Start
- 4:23Principles of Chaotic Engineering
- 5:07Define Steady State
- 9:06Hypothesize your steady state
- 11:45Monitoring
- 13:54Cascading Failures
- 16:14Culture
- 18:27Communication and Documentation
- 20:28Inject Failures into your systems
- 20:41Small Experiments
- 21:23Bigger Experiments
- 23:53Disprove your hypothesis
related talks
21mThe Psychology of Chaos Engineering
29mThe Psychology of Chaos Engineering
36mChaos Engineering in 2021
27mChaos Engineering: The Path to Reliability
18mChaos Engineering: Break It On Purpose
01h 04mHarnessing Chaos; the hidden ingredient behind building better systems through learning
23mMaking Chaos Engineering Boring: Debunking Myths Hampering Adoption
23mHow to tame chaos effectively?