flink-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From bartektartanus <...@git.apache.org>
Subject [GitHub] flink pull request #4924: [FLINK-7949] AsyncWaitOperator is not restarting w...
Date Mon, 30 Oct 2017 15:04:15 GMT
GitHub user bartektartanus opened a pull request:

    https://github.com/apache/flink/pull/4924

    [FLINK-7949] AsyncWaitOperator is not restarting when queue is full

    Change:
    Emitter thread is started BEFORE filling up the queue of recovered elements
    Issue description:
    During process restart, if the queue was full (with N elements) and there was pending
element waiting to be added to the queue, then the queue couldn't fit N+1 elements and thread
was blocked forever. As Till Rohrmann suggested here:
    http://apache-flink-user-mailing-list-archive.2336050.n4.nabble.com/Checkpoint-was-declined-tasks-not-ready-td16066.html
    I've changed the order of this code to start emitter thread earlier.   


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

    $ git pull https://github.com/bartektartanus/flink master

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

    https://github.com/apache/flink/pull/4924.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 #4924
    
----
commit 97620649ddfcf8f0320b20bdfdb69d9b44dd8f0c
Author: Bartłomiej Tartanus <bartektartanus@gmail.com>
Date:   2017-10-30T14:39:43Z

    start emmiter thread BEFORE filling up the queue of recovered elements

----


---

Mime
View raw message