Chaos Engineering

on YouTube ↗38m

" summary

Claims that chaos experiments increase confidence in system resilience by validating failure modes in production, not by counting fixes. Netflix builds and runs Chap to clone a service into control and experiment clusters, tag 1–2% of traffic, inject latency or failures via FIT and observe via Monocle and metrics (SPS, RPS, GC). Automated canary analysis with Kayenta aborts unsafe runs. Outcome: many misset timeouts, broken fallbacks and retry-on-same-host bugs discovered and fixed before user-visible outages.

AI-written summary · about

" chapters (11)

1. 0:00Introduction to chaos theory
2. 2:02Philosophy and goals
3. 4:51Testing methodologies
4. 7:06Cultural impact of failures
5. 8:38Anti-goals of chaos
6. 10:22Learning from incidents
7. 13:33Practical implementation
8. 19:17Observability and safety
9. 22:04Scaling experiments
10. 27:53Automation via Monocle
11. 35:00Conclusion and takeaways

" related talks

• 38mChaos Engineering: A Step Towards ResilienceNora Jones · REdeploy 2018
• 18mChaos Engineering: Break It On PurposeMorgan Carter · BSides London 2021
• 36mChaos EngineeringNikhil Barthwal · Agile India 2018
• 48mChaos Engineering: The Path to ReliabilityKolton Andrus · Agile India 2020
• 46mDesigning a System for ChaosPaul Stack · DevConf 2019
• 52mChaos engineering: Breaking things on purpose in productionDavid Lluna · Software Crafters Barcelona 2019
• 30mAntics, drift, and chaosLorin Hochstein · Strange Loop 2017
• 49mChaos & Intuition Engineering at NetflixCasey Rosenthal · GOTO Chicago 2016