Resilience testing in modernization projects
summary
Built a resilience framework on top of automation that runs as a CI/CD resilience stage after acceptance and scores services for fast feedback. Uses mocks in proxy mode (WireMock / Mountebank) plus SSH scripts and tc to inject dropped connections, garbled and empty responses, large payloads; Kali visualizes call graphs in Kubernetes. Finds real issues: a leaked DB connection that stalls all transactions and default retry/timeouts that create 90s user waits, both fixed.
AI-written summary · about
chapters (13)
- 0:00Introduction
- 1:12Resilience testing importance
- 4:03Why testing matters in modernization
- 7:14Key resilience testing areas
- 11:28Connection configs and timeouts
- 12:54Fault scenarios and mocking
- 16:24Implementing resilience challenges
- 19:13Takeaways and early tips
- 20:39Q&A: Biggest failures
- 21:41Q&A: Reliability vs resilience
- 22:03Q&A: CI/CD integration
- 23:08Q&A: Environment challenges
- 23:53Q&A: When to start
related talks
43mPeter Paul Bakker: The Power of Automated Resilience Tests
40mResilience Testing - Failure is an Option
28mBuilding and testing resilient applications
41mResillient Microservices without the Chaos
40mLet it fail: resilience patterns for microservices
32mBeyond Optimization: Resilient Cloud Microservices at Scale
49mManaging Resiliency in Microservices
01h 01mMastering Resilience Probes: Simplify chaos hypot...