Continuous Load Testing with GitHub Actions
summary
Uses GitHub Actions to run JMeter scripts against an app deployed by Bicep/ARM, invoking the Azure Load Testing service as the controller so nodes auto‑scale and private endpoints work. Pipeline passes a parameterized JMX (API_URL), runs 10 threads with 180s ramp and 240s hold, surfaces a 3,283 ms average and 0.04 error rate, pins the culprit to a Thread.Sleep, and gates promotion on response‑time and error‑rate criteria correlated with Azure Monitor.
AI-written summary · about
chapters (17)
- 0:00Continuous load testing intro
- 3:09What is load testing
- 7:28JMeter introduction
- 10:15Azure Load Testing overview
- 15:00Azure Load Testing GUI demo
- 17:53JMeter test plans with vars
- 21:13Load tests in GitHub Actions
- 24:08Infrastructure for load tests
- 27:43Load testing pricing
- 31:02Failure criteria in tests
- 34:17Analyze results & resize
- 36:47Iterative load testing best practices
- 38:56Record browser in JMeter
- 40:01Quick Azure load testing
- 41:07AKS load test repo
- 42:15Benefits of continuous load tests
- 43:39Q&A
related talks
17mDeep dive on Azure Load Testing in CI/CD
36mJMeter + GitHub = Understandable performance of your application
06mAzure Load Testing
40mThe Secret to Scaling Performance Test Automation with Jenkins and JMeter
07mSmart Load Testing
46mLeverage Azure Testing Services to build high quality applications
38mStating the Obvious: Adding Non-functional Tests to CI
34mPerformance testing: from zero to hero with k6 & Azure