Test Confidence: Types of Flake & How to Remediate
summary
Flaky tests cause lost trust as suites scale: a constant 10% flake rate turns into many unreliable tests as count grows. Concrete bugs include wrong element selection from cy.get selector pitfalls, missed intercepts when not define intercept before request, and cross-test state causing pipeline-only failures fixed by cy.session or programmatic setup/teardown. Remediation is to assert lengths or states, intercept before navigation, isolate test data, avoid blanket retries, and use Cypress Cloud flake analytics to track offenders.
AI-written summary · about
chapters (13)
- 0:00Introduction to test flake
- 1:25Why test flake matters
- 3:12Early flake remediation strategies
- 4:16Tool misuse causing flake
- 7:47Local vs pipeline test differences
- 10:16Avoiding anti-patterns in waits
- 12:03Parallel test conflicts
- 13:47Ensuring test independence
- 15:35Retryability of tests
- 18:03When flake reveals real issues
- 19:50Handling random test failures
- 20:52Flake analytics and tools
- 21:13Key takeaways on test confidence
related talks
15mHow to (not) write a flaky test
46mFilip Hric: How to Write a Flaky Test Description: How (not) to Write a Flaky Test
32mHow to write a flaky test
41m0Flake - Reaching reliable non-flaky tests
48mFlaky Tests 2.0
47mHow to Un-Flake Flaky Tests- a New Hire's Toolkit
28mTest flakiness; methods for dealing with flaky tests
01h 09mFlaky Tests: The Method