nielifeng commented on issue #10354:
URL: https://github.com/apache/seatunnel/issues/10354#issuecomment-5140201473
Verified the merged fix from #10509 on a current `dev` deployment at commit
`7e25769`.
I tested both paths through a worker-role REST endpoint:
1. **Missing checkpoint**
- submitted a valid streaming HOCON config with a never-used `jobId` and
`isStartWithSavePoint=true`
- HTTP `400`
- response contained `No checkpoint found for jobId=...`
- response did not contain `NullPointerException`
- no job or slot allocation was created
2. **Existing savepoint**
- submitted the same controlled FakeSource job normally with an explicit
job ID
- job reached `RUNNING`; checkpoint counts were
triggered/completed/failed/in-progress = `5/4/0/1`
- stop-with-savepoint returned HTTP `200` and the job reached
`SAVEPOINT_DONE` with no error
- resubmitted the same job ID to the worker-role endpoint with
`isStartWithSavePoint=true`
- restore returned HTTP `200`; the job reached `RUNNING`, preserved
`SourceReceivedCount=150`, and had no `errorMsg`
- normal stop completed and the cluster returned to a clean idle state
with no running, pending, or failed jobs
I could not reproduce the reported v2 restore NPE on current `dev`. The
negative and positive worker-entry paths both match the contract introduced by
#10509, so this issue looks ready to close as fixed.
--
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]
