LuckysHorizon commented on PR #13136: URL: https://github.com/apache/seatunnel/pull/13136#issuecomment-6732286955
> Thanks @LuckysHorizon for the update. Good news: the billing lock looks cleared. Attempt 4 of your fork run https://github.com/LuckysHorizon/seatunnel/actions/runs/42203281071 ran on this head (`718f00b8`), and the Apache `Build` check is now attached to it. > > That attempt has exactly one failed job: `all-connectors-it-5 (17, ubuntu-latest)`. Everything else is green or skipped as expected, including the unit-test and engine jobs. The failure is environmental and has nothing to do with this change: > > * `TiDBCDCIT.startUp` (`TiDBCDCIT.java:94`) -> `TiDBTestBase.initializeTidbTable` (`:213`) fails with `CommunicationsException: Communications link failure`. The TiDB container reported "started" about a second earlier and the first JDBC connection was dropped. > * The same module passed in the JDK 11 job `all-connectors-it-5 (11)` of the same run. Your PR only touches `CoordinatorService.getJobInfo()` and its test, which the TiDB CDC E2E never reaches. We have seen this TiDB startup failure on other unrelated PRs, and `dev` has no fix for it yet. > > Cheapest next step: open the run above and use "Re-run failed jobs", so only that one job runs again. No new commit and no `dev` sync are needed. My review conclusion is unchanged: no non-CI blockers. I'll approve once the Build is green on this head. Thanks, @DanielLeens. I re-ran the failed job (all-connectors-it-5 (17)) on the existing run, with no new commit, and it passed. The Build is now green on 718f00b8: https://github.com/LuckysHorizon/seatunnel/actions/runs/42203281071 So the earlier failure was the unrelated TiDBCDCIT startup flake you described. Summary of where this stands: - getJobInfo() precedence (running -> pending -> distributed JobInfo -> history) and the regression tests are unchanged since 6f745fc2. - CI is green on this head. - #13133 overlaps on the running/pending branches. I'm happy to coordinate with its author and the maintainers so only one lookup fix lands. Could you take another look when convenient? Thanks again for the thorough reviews. -- 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]
