Clear the ring for Chaos Engineering
summary
Practitioners recount a year of running 80+ game days that uncover 120+ production-relevant weaknesses, describe a real incident where a new KCI feature caused cascading failures, and show a rolling update under load that produced 100% error for 15 minutes. Tooling notes include Chaos Monkey for Kubernetes, Pumbaa, Instana APM and an evaluation of Steadybit/Gremlin. Emphasis stays on concrete experiments, cross-team game days, measurable metrics (time to recovery, time between failures) and how they iteratively recovered from early missteps.
AI-written summary · about
chapters (14)
- 0:00Chaos Engineering Circus
- 2:21Monolith to modern chaos
- 8:08First feature release disaster
- 10:54Blameless postmortems
- 13:22Why Chaos Engineering like Netflix
- 16:31Starting over with chaos
- 17:37Begin chaos experiments
- 18:41Game day failure under load
- 20:47Value of chaos practices
- 22:31Tooling and monitoring
- 22:52Top five excuses not to start
- 27:43Success and game day results
- 29:05Top five learnings
- 31:55Call to action
related talks
26mBreaking Systems for Resilience
01h 04mHarnessing Chaos; the hidden ingredient behind building better systems through learning
38mChaos Engineering
33mMonkey Business For Serious Systems: Why chaos engineering isnt crazy
34mAntics, Drift, and Chaos
38mWhy software developers should build a resilience culture based on chaos engineering
47mDeprecating Simplicity 3.0
03mDeveloping a Chaos Architecture Mindset