Commit
This commit does not belong to any branch on this repository, and may belong to a fork outside of the repository.
Browse files
Browse the repository at this point in the history
This reverts commit 44ca1b2. ### Description The recently merged PR #5342 argues that we should automatically retry failing IDE tests, because that's what we manually do anyways. However, we only do that in some cases. Specifically, when any of us make changes that may create unstable tests, such as changes in the IDE, then failing IDE tests should not be retried, manually or automatically. PR5342 makes it so that such a retry happens automatically, thereby making it much easier to introduce non-deterministic bugs either in IDE code or tests. That's why this PR is reverting that one. IMO there is little difference between a test that fails 100% of the time and one that fails <100% of the time. Retrying a failing test that fails <100% of the time is similar to saying "I believe my PR is not at fault for this test failing, please ignore it". A similar situation occurs when a PR gets merged that causes nightly to fail. However then we don't have a way for new PRs to say "please ignore nightly failing", and we explicitly block all PR until nightly is fixed again. A similar approach to unstable tests would be to not to retry failing builds until the instability is resolved, but instead to wait until an instability fix PR is merged. ### How has this been tested? It's a revert. <small>By submitting this pull request, I confirm that my contribution is made under the terms of the [MIT license](https://github.com/dafny-lang/dafny/blob/master/LICENSE.txt).</small>
- Loading branch information