How to do in-app chaos testing
summary
Shows an open‑source .NET library that injects faults via Microsoft.Extensions.DependencyInjection to simulate real failures without changing business code: it throws IO exceptions after three logs, replaces returned customer records with “Chuck Norris”, and inserts a 30 second delay on service calls. Demos include breaking Windows Hello login (causes exception until code adds try/catch), slowing customer queries (UI stays responsive but stalls), and simulating disk full—useful for running controlled chaos in dev or CI to verify circuit breakers, fallbacks and logging behavior.
AI-written summary · about
chapters (26)
- 0:00Introduction
- 1:27Overview
- 2:32Netflix
- 3:39Chaos Monkey
- 5:52Amazon Audible
- 6:55Case Engineering
- 9:22Scale it down
- 12:52First situation
- 15:03Second situation
- 16:21Dealing with poor connectivity
- 18:08The idea of chaos testing
- 20:10Example app
- 21:19Add packages
- 22:52Add chaos
- 25:06Triggers
- 29:10Login view model
- 30:47Services being slow
- 32:02Overloads
- 33:44Responsiveness
- 37:54Running Application
- 40:02LockService
- 41:53Changing the data returned
- 44:07Get Customers
- 46:52Transform
- 49:14Demo Time
- 53:22Thoughts on integration
related talks
57mHow to do in-app chaos testing
40mResilience Testing - Failure is an Option
01h 02m1,2,3… testing: is this thing on(line)? Meet your new Microsoft Testing tools
01h 19mYou Can't Test This? Hammertime!
26mLightning Talk: Think Like a Chaos Engineer: Simulating User Anomalies the Litmus Way
37mUse Chaos Testing to Build Resilience in the Software
38mStating the Obvious: Adding Non-functional Tests to CI
01h 02mTesting First - Getting The How