Resilience Testing - Failure is an Option
summary
Claims a practical method that lets tests flip failures inside a running system without shipping test code into production by using AOP weaving to inject faults at runtime. Demonstrates container and network chaos with tools like Pumbaa and Massaro, and service-level fault injection inspired by Netflix FIT/Chaos Monkey. Gives a concrete cautionary loss — $460 million in 45 minutes — to justify testing, and reports the approach is used in real teams with live coding demos planned though not shown for time.
AI-written summary · about
chapters (14)
- 0:00Introduction
- 1:50The Happy Path
- 2:53The Blind Spot
- 4:57What is Resilience
- 7:15Reactive Systems Manifesto
- 14:08Disaster Recovery Plan
- 16:13Surprises
- 18:48Infrastructure
- 20:38Tools
- 22:13Services
- 24:30Fetus Framework
- 30:02Delay
- 33:30The Rule
- 36:03AOP
related talks
16mEngineering Failure-Resilient Systems
37mUse Chaos Testing to Build Resilience in the Software
57mChaos Engineering Primer
31mResilience Engineering
25mResilience testing in modernization projects
48mChaos Engineering: The Path to Reliability
26mImprove application resilience with chaos engineering
01h 02mPractical Chaos Engineering breaking things on purpose to make them more resilient against failure