[FLINK-28884] Downstream task may never be notified of data available in hybrid shuffle when number of credits is zero. #20529
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
…e state. HsSubpartitionView should be initialized to a notifiable state, there may be a problem of never consuming otherwise. Imagine the following situation: Downstream task has no initial credit(i.e. exclusive buffers is configured to zero), if there is no data output in the upstream, it will feedback a zero backlog to downstream input channel. All subsequent data available notifications will be intercepted as needNotify is false.
Collaborator
HsSubpartitionView should be notifiable when downstream get a zero backlog. Generally speaking, if the backlog is zero, when data become available, even if there is no credit, the backlog information will be notified also. However, in the hybrid shuffle, the notification will be ignored. This behavior is incorrect.
Member
Author
|
@flinkbot run azure |
huangxiaofeng10047
pushed a commit
to huangxiaofeng10047/flink
that referenced
this pull request
Nov 3, 2022
HsSubpartitionView should be notifiable when downstream get a zero backlog. Generally speaking, if the backlog is zero, when data become available, even if there is no credit, the backlog information will be notified also. However, in the hybrid shuffle, the notification will be ignored. This behavior is incorrect. This closes apache#20529
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
What is the purpose of the change
Fix the bug that downstream task may never be notified of data available in hybrid shuffle when number of credits is zero.
Brief change log
HsSubpartitionViewshould be initialized to a notifiable state.Verifying this change
This change added test in
HsSubpartitionViewTest.Does this pull request potentially affect one of the following parts:
@Public(Evolving): noDocumentation