Scenario

Your team’s CI pipeline has started failing intermittently (about 30% of the time) on the “Test” stage, specifically in the UI integration tests. No code changes were made to the test framework recently. Re-running the job usually passes.

Question

How would you approach debugging these “flaky” tests? What common causes would you investigate?

Expected Answer

1. Investigation: Check logs for timing issues, timeouts, or resource contention on the runner.
2. Common Causes:
   • Race Conditions: Tests depending on async events not finishing.
   • Test Pollution: Previous tests leaving DB/state in a dirty state.
   • Environment: Shared resources or network latency during peak times.
3. Remediation: Isolate the failing test, add better wait conditions (avoid sleep), ensure atomic test data, or use retry mechanisms (as a temporary measure only).