[jira] [Commented] (FLINK-4387) Instability in KvStateClientTest.testClientServerIntegration()

["ASF GitHub Bot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/FLINK-4387?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16381691#comment-16381691 ] 

ASF GitHub Bot commented on FLINK-4387:
---------------------------------------

GitHub user NicoK opened a pull request:

    https://github.com/apache/flink/pull/5606

    [FLINK-4387][QS] don't wait and process requests at Netty servers after shutdown request

    ## What is the purpose of the change
    
    There is a race condition on an assertion in Netty's event loop that may cause
    tests to fail when finished early.
    
    This was fixed in 4.0.33.Final via https://github.com/netty/netty/issues/4357 but we currently still rely on Netty 4.0.27.Final, hence the bug fix for the tests where this bug can be immanent.
    
    ## Brief change log
    
    - in QueryableState tests, do not wait for a "quiet period" for further events to arrive and process, this is not needed here and circumvents the bug
    
    ## Verifying this change
    
    This change is a trivial rework / code cleanup without any test coverage.
    
    ## 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, Yarn/Mesos, ZooKeeper: **no**
      - The S3 file system connector: **no**
    
    ## Documentation
    
      - Does this pull request introduce a new feature? **no**
      - If yes, how is the feature documented? **JavaDocs**


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

    $ git pull https://github.com/NicoK/flink flink-4387

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

    https://github.com/apache/flink/pull/5606.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 #5606
    
----
commit 7acd43e9148da218806a43e2b381f84f8106d3ac
Author: Nico Kruber <ni...@...>
Date:   2018-03-01T08:43:56Z

    [FLINK-4387][QS] don't wait and process requests at Netty servers after shutdown request
    
    There is a race condition on an assertion in Netty's event loop that may cause
    tests to fail when finished early.
    
    This was fixed in 4.0.33.Final, see https://github.com/netty/netty/issues/4357.

----


> Instability in KvStateClientTest.testClientServerIntegration()
> --------------------------------------------------------------
>
>                 Key: FLINK-4387
>                 URL: https://issues.apache.org/jira/browse/FLINK-4387
>             Project: Flink
>          Issue Type: Bug
>    Affects Versions: 1.1.0, 1.5.0
>            Reporter: Robert Metzger
>            Priority: Blocker
>              Labels: test-stability
>             Fix For: 1.2.0
>
>
> According to this log: https://s3.amazonaws.com/archive.travis-ci.org/jobs/151491745/log.txt
> the {{KvStateClientTest}} didn't complete.
> {code}
> "main" #1 prio=5 os_prio=0 tid=0x00007fb2b400a000 nid=0x29dc in Object.wait() [0x00007fb2bcb3b000]
>    java.lang.Thread.State: WAITING (on object monitor)
> 	at java.lang.Object.wait(Native Method)
> 	- waiting on <0x00000000f7c049a0> (a io.netty.util.concurrent.DefaultPromise)
> 	at java.lang.Object.wait(Object.java:502)
> 	at io.netty.util.concurrent.DefaultPromise.await(DefaultPromise.java:254)
> 	- locked <0x00000000f7c049a0> (a io.netty.util.concurrent.DefaultPromise)
> 	at io.netty.util.concurrent.DefaultPromise.await(DefaultPromise.java:32)
> 	at org.apache.flink.runtime.query.netty.KvStateServer.shutDown(KvStateServer.java:185)
> 	at org.apache.flink.runtime.query.netty.KvStateClientTest.testClientServerIntegration(KvStateClientTest.java:680)
> 	at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)
> {code}
> and
> {code}
> Exception in thread "globalEventExecutor-1-3" java.lang.AssertionError
> 	at io.netty.util.concurrent.AbstractScheduledEventExecutor.pollScheduledTask(AbstractScheduledEventExecutor.java:83)
> 	at io.netty.util.concurrent.GlobalEventExecutor.fetchFromScheduledTaskQueue(GlobalEventExecutor.java:110)
> 	at io.netty.util.concurrent.GlobalEventExecutor.takeTask(GlobalEventExecutor.java:95)
> 	at io.netty.util.concurrent.GlobalEventExecutor$TaskRunner.run(GlobalEventExecutor.java:226)
> 	at io.netty.util.concurrent.DefaultThreadFactory$DefaultRunnableDecorator.run(DefaultThreadFactory.java:137)
> 	at java.lang.Thread.run(Thread.java:745)
> {code}



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)