pnowojski opened a new pull request #15734:
URL: https://github.com/apache/flink/pull/15734
This bug was happening when a downstream tasks were failing over or being
cancelled. If all
of the downstream tasks released subpartition views, underlying memory
segments/buffers could
have been recycled, while upstream task was still writting some records.
The problem is fixed by adding the writer (result partition) itself as one
more reference
counted user of the result partition
## Verifying this change
This change adds new test.
## Does this pull request potentially affect one of the following parts:
- Dependencies (does it add or upgrade a dependency): (yes / **no**)
- The public API, i.e., is any changed class annotated with
`@Public(Evolving)`: (yes / **no**)
- The serializers: (yes / **no** / don't know)
- The runtime per-record code paths (performance sensitive): (yes / **no**
/ don't know)
- Anything that affects deployment or recovery: JobManager (and its
components), Checkpointing, Kubernetes/Yarn/Mesos, ZooKeeper: (yes / **no** /
don't know)
- The S3 file system connector: (yes / **no** / don't know)
## Documentation
- Does this pull request introduce a new feature? (yes / **no**)
- If yes, how is the feature documented? (**not applicable** / docs /
JavaDocs / not documented)
--
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]
