DanielLeens commented on PR #11642:
URL: https://github.com/apache/seatunnel/pull/11642#issuecomment-5242234819

   Agreed, and this matches what I saw independently when I reviewed this head 
on 2026-08-04: the same two jobs — `all-connectors-it-3 (11, ubuntu-latest)` 
and `kafka-connector-it (11, ubuntu-latest)` — were the only failures, both 
with the identical `Run SeaTunnel on spark failed` / `SeaTunnel job executed 
failed` annotation, while all four `unit-test` lanes (including the 
previously-flaky `unit-test (11, windows-latest)` this PR actually targets) 
were green. Those connector IT lanes only run here because touching 
`seatunnel-api` pulls in the broad matrix — this diff is a single test-file 
change and isn't on their classpath, so it can't be causing a packaged Spark 
e2e job to fail.
   
   Good catch cross-checking against other PRs — I can independently confirm 
that angle too: `danielnadean` just flagged the exact same `Run SeaTunnel on 
spark failed` signature a few seconds after your comment here, on #10583 (a 
pure-Javadoc PR, zero executable lines changed), also on `kafka-connector-it` 
plus two other ubuntu lanes. Two unrelated PRs, different authors, different 
files, different job-matrix JDKs (8 vs 11), hitting the identical annotation 
text at essentially the same moment is strong corroboration this is an 
infra-wide Spark e2e flake rather than anything in either diff — consistent 
with #10583's history of the same failure going back to March.
   
   My conclusion stands: no source-side blocker in this PR, "Ready to merge" 
once the flaky lanes go green. Thanks for triggering the job-level rerun on the 
two failed jobs rather than the whole workflow — that's the right move here. I 
only have read-level access on this repo, so I can't kick off or watch the 
rerun myself, but a maintainer re-run of those two lanes (plus #10583's, if 
anyone with fork access there can trigger it) should be enough to clear this.


-- 
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]

Reply via email to