1996fanrui opened a new pull request, #26900:
URL: https://github.com/apache/flink/pull/26900
## What is the purpose of the change
Both of input buffers and out buffers use the same EndOfChannelStateEvent,
but it is used in different way.
For EndOfChannelStateEvent of input buffer (downstream task side):
- It is emitted by the downstream after emitting the last recovered input
buffer for a channel
- This triggers that input channel conversion to a “normal channel”
For EndOfChannelStateEvent of output buffer (upstream task side):
- it is emitted by the upstream after sending the last recovered output
buffer
- this eventually triggers task input conversion in the downstream to a
“normal input”
The code would be more clearer if this event is separated into 2 events,
like:
- EndOfInputChannelStateEvent and EndOfOutputChannelStateEvent
## Brief change log
[FLINK-38216][checkpoint][refactor] Split EndOfChannelStateEvent into
EndOfInputChannelStateEvent and EndOfOutputChannelStateEvent
## Verifying this change
- Test EndOfInputChannelStateEvent and EndOfOutputChannelStateEvent in
EventSerializerTest
## 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, ZooKeeper: no
- The S3 file system connector: no
## Documentation
- Does this pull request introduce a new feature? no
--
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]
