What are the key differences between Regression and Retesting in a CI/CD pipeline?

All answers to this question.

Retesting is for the "hole," Regression is for the "entire fabric." You need both to ensure your software stays seamless.

   Answered 2024-03-23 by Linda Garcia

• I love that analogy, Linda. It really helps non-technical stakeholders understand why we can't just "test the fix" and call it a day.

     Commented 2024-03-24 by James Wilson

Do you use any specific tagging system in your automation code to help the CI/CD tool know which tests are "Selective"?

   Answered 2024-03-25 by Steven Harris

• Steven, yes! In Cucumber or Playwright, we use @tags. We tag tests by module (e.g., @billing, @login) and by priority (@P1, @smoke). Our Jenkins pipeline is scripted to look at the "Changed Files" in the PR and trigger only the corresponding tags. It’s a bit of work to set up, but it saves hours of execution time every single week.

     Commented 2024-03-27 by Richard Scott

Retesting is specifically checking if a previously failed test case now passes after a fix. Regression is checking the rest of the system to ensure the fix didn't break unrelated features. In 2024, my team optimized our pipeline by using "Smoke Tests" for every commit and a "Selective Regression" suite that only runs tests related to the modules changed. We run the "Full Regression" nightly. This prevents the pipeline from becoming a bottleneck while still providing a high level of confidence before each production release.

   Answered 2024-08-15 by Barbara Taylor