flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From tillrohrmann <...@git.apache.org>
Subject [GitHub] flink pull request #5360: [backport-1.4] [FLINK-8485] [client] Unblock JobSu...
Date Thu, 25 Jan 2018 10:50:18 GMT
GitHub user tillrohrmann opened a pull request:

    https://github.com/apache/flink/pull/5360

    [backport-1.4] [FLINK-8485] [client] Unblock JobSubmissionClientActor#tryToSubmitJob

    This is a backport of #5359. Additionally it also includes a backport of #4924.

You can merge this pull request into a Git repository by running:

    $ git pull https://github.com/tillrohrmann/flink unblockJobSubmissionClientActor1.4

Alternatively you can review and apply these changes as the patch at:

    https://github.com/apache/flink/pull/5360.patch

To close this pull request, make a commit to your master/trunk branch
with (at least) the following in the commit message:

    This closes #5360
    
----
commit ce399d5dee3f8bb065bacd2caf9f848d08a954e9
Author: Bartłomiej Tartanus <bartektartanus@...>
Date:   2017-10-30T14:39:43Z

    [FLINK-7949] Make AsyncWaitOperator recoverable also when queue is full
    
    Start emitter thread BEFORE filling up the queue of recovered elements.
    This guarantees that we won't deadlock inserting the recovered elements,
    because the emitter can already start processing elements.
    
    This closes #4924.

commit 7bd682e5e379952458b0dc91d0f7f218afb83f50
Author: Till Rohrmann <trohrmann@...>
Date:   2018-01-10T17:53:38Z

    [FLINK-7949] Add unit test for AsyncWaitOperator recovery with full queue

commit b06670e1361bf5bae76073db8f7060af61c26100
Author: Till Rohrmann <trohrmann@...>
Date:   2018-01-25T10:40:33Z

    [FLINK-8485] [client] Unblock JobSubmissionClientActor#tryToSubmitJob
    
    The JobSubmissionClientActor blocked a ActorSystem's dispatcher thread when requesting
    the BlobServer port from the cluster. This fails when using the FlinkMiniCluster on a
    single core machine because we set the number of threads to 1.
    
    This commit unblocks the JobSubmissionClientActor#tryToSubmitJob method and sets the
    lower limit of dispatcher threads to 2 when using the FlinkMiniCluster.

----


---

Mime
View raw message