SEZ9 commented on PR #12152: URL: https://github.com/apache/seatunnel/pull/12152#issuecomment-6936957240
Thanks @CryoThrust, that explains why the old run `37615433319` could not be recovered. Syncing `dev` in `e3679f4d5` is a reasonable way to get a fresh run, given the rerun was rejected and the old revision had no `workflow_dispatch` trigger. On the two previous points: - **F1 (rejection path in `CheckpointCoordinator`)**: Since you state the sync added no new fallback logic and the PR still changes only `CheckpointCoordinator` and its test, the earlier source-side conclusion stands. I have no new ask here. Rejection by both pools remains the agreed follow-up and is outside this PR's scope. - **F2 (test coverage)**: Your local `CheckpointCoordinatorTest` result on JDK 11 (20 tests, no failures, errors or skips), with no assertion changes in the sync, is consistent with what was rechecked earlier. The local Spotless and server-module `verify` passes are useful, but they do not replace an executed CI run. Remaining asks: 1. Once the new Build run `43885011044` acquires a runner and finishes, please post the result here. 2. If it stalls in the queue again or fails on runner acquisition, please say so, so it is not read as a test result. 3. If any job fails with a real test failure, please paste the job and test name so it can be classified against F1/F2. No further source change is requested while that run is pending. <!-- streview-comment:2395 --> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
