becketqin opened a new pull request #14030:
URL: https://github.com/apache/flink/pull/14030
## What is the purpose of the change
This patch fixes the following two issues:
1. Currently the `ExecutorNotifier` runs the `handler` in the worker thread
if there is an exception thrown from the `callable`. This breaks the threading
model and prevents an exception from bubbling up to fail the job.
2. When an exception bubbles up from the `SourceCoordinator`, the
`UncaughtExceptionHandler` will call `System.exit(-17)` and kill the JM. This
is too much. Instead, we should just fail the job to trigger a failover.
## Brief change log
- Let the the `handler` always run in the main thread in `ExecutorNotifier`
- Fail the job instead of killing the JM in the `UncaughtExceptionHandler`
in the `SourceCoordinator` executor.
## Verifying this change
The following unit tests have been added to verify the change.
- ExecutorNotifierTest.testExceptionInHandlerWhenHandlingException()
- SourceCoordinatorProviderTest.testCallAsyncExceptionFailsJob()
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): (no)
- The public API, i.e., is any changed class annotated with
`@Public(Evolving)`: (no)
- The serializers: (no)
- The runtime per-record code paths (performance sensitive): (no)
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn/Mesos, ZooKeeper: (no)
- The S3 file system connector: (no)
## Documentation
- Does this pull request introduce a new feature? (no)
- If yes, how is the feature documented? (not applicable)
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
