DanielLeens commented on PR #12182: URL: https://github.com/apache/seatunnel/pull/12182#issuecomment-5601136420
Thanks for the update, @SEPURI-SAI-KRISHNA — I independently re-checked the fork's check-runs for the current head (`08b4cd2a7664`) directly against the Actions API rather than taking the summary at face value, and it matches your report exactly: `engine-v2-it (11, ubuntu-latest)` and `kudu-connector-it (8, ubuntu-latest)` are now `success`, and `transform-v2-it-part-1 (11, ubuntu-latest)` (run_attempt 3, completed 12:32:10Z) is still `failure`. I pulled that job's log directly and confirmed the failing test is the same `TestFilterRowKindIT.testFilterRowKindMultiTable`, matching #12116's signature. Every leg that actually exercises this PR's diff — `transform-v2-it-part-1 (8)`, both `transform-v2-it-part-2` JDKs, and both `unit-test` legs — is green. I also checked #12116 itself: still open, still under root-cause investigation, no fix PR yet (your comment there a few minutes ago adding this PR's two occurrences is the latest activity). So a third rerun of `transform-v2-it-part-1 (11)` is unlikely to help — you've now reproduced the same failure twice on this exact head across two different Flink legs (1.18.0 and 1.15.3), and #12116's own diagnosis (`AssertSinkWriter`'s static JVM-wide counters) is orthogonal to anything in this PR's diff. Concrete next step: since I'm a comment-only reviewer here and can't waive the required check myself, this now needs @davidzollo (or another maintainer with write access) to make the call on merging with `transform-v2-it-part-1 (11)` red-but-explained, since fixing #12116 first isn't something this PR can unblock on its own. From a pure correctness standpoint there's nothing left for me to flag — the fix is sound, both unit test classes are green, and the one remaining red job is a pre-existing, independently-tracked flake that reproduces the same way regardless of this diff. -- 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]
