DanielLeens commented on PR #12595:
URL: https://github.com/apache/seatunnel/pull/12595#issuecomment-6143528087

   Thanks @SEPURI-SAI-KRISHNA for the thorough write-up, and thanks @SEZ9 for 
the follow-ups. I have nothing further on the code: `BasicAuthFilter` and its 
two tests are unchanged since my last pass, and the time-constant comparison 
point is settled.
   
   On the `CassandraIT` failure, I can add one data point to help with the 
tracking item. On the current `dev` head `b0f6feddf9` (run 38340962420, job 
`all-connectors-it-2 (11, ubuntu-latest)`), the very same two assertions fail: 
`CassandraIT.testCassandra:134` and 
`CassandraIT.testCassandraMultiTableSource:144`, both `expected: <0> but was: 
<1>`. The run's own summary names the failing invocation as `TestContainer[4]`, 
and the log of that job shows the Flink 1.20.1 taskmanager container together 
with a `Direct buffer memory` error. So the job exit code is non-zero because 
the engine run fails in that container, which matches your conclusion that it 
is not related to this PR. (Invocation numbering may differ between your report 
and mine, so please treat `[2]` vs `[4]` as a labeling detail rather than a 
contradiction; I did not map it further than the container named in the log.) I 
also noticed #12732, which targets a Cassandra E2E flake, but its description 
talks about a row-count ra
 ce after TRUNCATE, which is a different assertion from the exit-code one here, 
so I would not expect it to fix this lane; I have not verified that further.
   
   Since `dev` itself is red on this lane, a rerun will not clear it. The 
minimal next step is simply to wait for an upstream fix, then sync the latest 
`dev` and rerun only the failed jobs. If you open the tracking issue SEZ9 
suggested, the `[N]`-only pattern plus the `Direct buffer memory` line from the 
Flink 1.20.1 taskmanager would be the most useful details to include.
   


-- 
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