DanielLeens commented on PR #12366: URL: https://github.com/apache/seatunnel/pull/12366#issuecomment-6143526813
Thanks @201811510411lw for syncing with `dev` and for pinning the CI comparison down so precisely. I checked it independently, and your reading is correct. - Your PR head `b43bc2876`: the only failing jobs in the fork run are `all-connectors-it-2` on JDK 11 and JDK 17. In the JDK 11 log, `CassandraIT.testCassandra:134` and `CassandraIT.testCassandraMultiTableSource:144` fail with `expected: <0> but was: <1>`, and the log contains the `Direct buffer memory` error from the Flink 1.20.1 taskmanager. - Current `dev` (`b0f6feddf9`, run 38340962420): the same lane `all-connectors-it-2` (JDK 11 and 17) is red, with the same two assertions at the same lines, again on invocation `[4]` of the parameterized `TestContainer` (the Flink 1.20.1 container). So this failure reproduces on the base branch without this PR, and the diff here (Paimon multi-table sink routing) does not touch the Cassandra connector, its E2E module or the Flink container setup. I do not see a source-level blocker from the CI side. Next step: there is nothing to change in this PR for that lane. Please just leave it as is; a rerun of the failed jobs will keep failing until a fix lands on `dev`, so I would hold off on reruns and only re-run once `dev` is green for `all-connectors-it-2` (a sync with the latest `dev` at that point is enough). My code-level conclusion from the previous round is unchanged. -- 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]
