DanielLeens commented on PR #11399:
URL: https://github.com/apache/seatunnel/pull/11399#issuecomment-5551392606

   CI update on this exact head (`74db0a10ae269`): the `Build` run I noted as 
still queued at review time has since completed on the fork 
(`https://github.com/zhangshenghang/seatunnel/actions/runs/33499602605`) with 8 
failing jobs, and I want to flag one of them specifically because it isn't the 
usual unrelated-module noise: `updated-modules-integration-test-part-1` failed 
on **both** JDK 8 and JDK 11, and the actual failure is inside 
`connector-cdc-postgres-e2e` (`PostgresCDCIT`) — 27 errors + 1 failure out of 
37 tests, all `ConditionTimeout` (awaitility assertion timeouts), e.g. 
`testPostgresCdcMultiTableE2e`, `testPostgresCdcCheckDataWithNoPrimaryKey`, 
`testMPostgresCdcCheckDataE2e`. That's the exact connector this PR's 
`database.include.list` change touches, and the failure rate (nearly the whole 
suite, on both JDKs) is high enough that I don't think this can be waved off as 
routine container-startup flake without checking — it could plausibly be the 
new `database.include.lis
 t` population interacting with however the E2E's own Postgres 
container/database names are set up. The other 7 failing jobs (`unit-test (8, 
ubuntu-latest)` -> `seatunnel-engine-server`, `iceberg-connector-it`, 
`doris-connector-it`) look unrelated to this diff and I'm not raising those.
   
   Given this shows up specifically in the module this PR changes, could you 
check whether `testPostgresCdcMultiTableE2e` and friends still pass locally 
against this head before we treat 'CI is the only remaining gate, no code-side 
blocker' as still accurate? I don't want to call this a confirmed blocker 
without your own repro, but I also don't want to leave that framing standing 
unqualified while this is unresolved.


-- 
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