spuru9 opened a new pull request, #29766:
URL: https://github.com/apache/flink/pull/29766

   ## What is the purpose of the change
   
   `StateCheckpointedITCase` hangs on master nightlies. FLINK-39018 lets a 
blocked `PipelinedSubpartition` deliver priority buffers, which is meant for 
the recovery (`RECOVERY_COMPLETION`) block. Because the bypass also applies 
after an aligned barrier, the timed-out barrier of checkpoint N+1 can overtake 
an aborted checkpoint N's aligned barrier on the same channel. The downstream 
then fails in `RemoteInputChannel#convertToPriorityEvent`, and the test 
restarts forever on empty finished-subtask state.
   
   ## Brief change log
   
     - `PipelinedSubpartition`: only a `RECOVERY_COMPLETION` block lets 
priority buffers through; other blocks hold them back until 
`resumeConsumption()`, as before FLINK-39018.
     - `StateCheckpointedITCase`: disable checkpoints after tasks finish, and 
reset the static failure flag before each run.
   
   ## Verifying this change
   
   This change added tests and can be verified as follows:
   
     - 
`AlternatingCheckpointsTest#testNewerPriorityBarrierDoesNotOvertakeBlockingAlignedBarrier`
 drives a real subpartition into a remote channel; before the fix it fails with 
the CI stack trace.
     - Added 
`PipelinedSubpartitionTest#testTimedOutBarrierWaitsForResumeAfterAlignedBarrier`
 and 
`PipelinedSubpartitionWithReadViewTest#testPriorityEventWaitsForResumeAfterAlignedBarrier`;
 the FLINK-39018 bypass tests now block with `RECOVERY_COMPLETION`.
     - Looped `StateCheckpointedITCase` 24 times with the seed of a failed 
nightly: 7 runs hit the exception and 4 hung before the fix, none after.
   
   ## 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: yes (checkpointing: an 
unaligned barrier waits behind an aligned barrier or aligned watermark block 
again, as before FLINK-39018)
     - 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
   
   ---
   
   ##### Was generative AI tooling used to co-author this PR?
   
   - [X] Yes (please specify the tool below)
   
   Generated-by: Claude Code (Claude Opus 5.5)
   


-- 
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]

Reply via email to