Skip to content

Add flake detection

Sergi Blanch Torné requested to merge sergi/ci-uprev:detect_flakes into main
  • After the pipeline execution, check if there are jobs in the failed status.
  • If that is the case, check if they have any artifacts. If they don't, then assume there was a problem with the CI infrastructure, and retry them.
  • Once all failing jobs have artifacts stored, record the results and retry.
  • For stable results across retries, update the test expectations. For those that aren't stable, record them as flakes.

This merge request depends on MR!6

Edited by Sergi Blanch Torné

Merge request reports