SEPURI-SAI-KRISHNA commented on PR #12568: URL: https://github.com/apache/seatunnel/pull/12568#issuecomment-5958250654
Thanks for the approval, @nzw921rx. CI came back red on `314556b79`, so here is what the two failures are. Neither is in `seatunnel-api`, which is the only module this PR touches. Run `36966371906`: 80 jobs successful, 2 failed, 11 skipped, 1 cancelled. | leg | failure | | | --- | --- | --- | | `all-connectors-it-7 (11)` | `PostgresCDCIT.testPostgresCdcSnapshotOnlyAndCommittedOffsetStartupModes[1]`, 37 tests with 1 error | #11847 | | `paimon-connector-it (11)` | `PaimonSinkCDCIT.testChangelogLookup[1]`, 14 tests with 1 failure | see below | **`PostgresCDCIT`** is #11847, whose title is about committed-offset restore and whose subject is the same startup-mode path the failing test exercises. It is also not specific to this PR: the identical test failed with the identical signature on my unrelated #12605 roughly three hours later, same 37 tests, same single error, same ~260s. **`PaimonSinkCDCIT.testChangelogLookup`** I cannot account for as cleanly, so I will not pretend otherwise. There is no open issue for it; the only match is #7328, which is closed and Windows-specific. I checked the recent scheduled `dev` runs for a precedent and the one `paimon-connector-it` failure I found, on 2026-09-25, was `PaimonWithS3IT`, a different test. So on the evidence I have this is a single observation with nothing behind it yet, not a known flake. The test is not new either: `PaimonSinkCDCIT` was last touched on 2026-08-23 by #11608. Could I ask for a failed-jobs-only rerun on `314556b79`, no new commit? That is what settled the comparable situation on #12290. If `PaimonSinkCDCIT` passes, it was flaky and the picture is clean. If it fails again I will have a second data point and will open an issue for it with both traces rather than leaving it unattributed. -- 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]
