DanielLeens commented on issue #11473: URL: https://github.com/apache/seatunnel/issues/11473#issuecomment-5009051452
Thanks for reducing this to a simple `FakeSource -> Console` streaming reproduction. This does look like a real Zeta issue rather than a usage mistake. The reason is that the savepoint stop path is supposed to move the job from `DOING_SAVEPOINT` to `SAVEPOINT_DONE` only after the savepoint/checkpoint completion path closes the loop. In your case the job keeps running, the REST call times out, and a normal stop can still clean everything up afterward, which strongly suggests the `stop with savepoint` handoff is not completing correctly for this path rather than the job being misconfigured. So this is worth keeping open as a valid bug in the Zeta savepoint-stop flow. As a temporary workaround, please use a normal stop (`isStopWithSavePoint=false`) when you need reliable cleanup, and only rely on the savepoint stop path after a fix lands. We've labeled this as `help wanted` — contributions are very welcome. -- 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]
