You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flink.apache.org by "Robert Metzger (JIRA)" <ji...@apache.org> on 2017/01/23 12:23:27 UTC

[jira] [Updated] (FLINK-5062) Ensure that all created threads have a proper name

     [ https://issues.apache.org/jira/browse/FLINK-5062?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Robert Metzger updated FLINK-5062:
----------------------------------
    Fix Version/s:     (was: 1.2.0)

> Ensure that all created threads have a proper name
> --------------------------------------------------
>
>                 Key: FLINK-5062
>                 URL: https://issues.apache.org/jira/browse/FLINK-5062
>             Project: Flink
>          Issue Type: Improvement
>    Affects Versions: 1.1.3
>            Reporter: Chesnay Schepler
>
> There are a few places where we create a Thread without giving it an explicit name. This can make debugging harder than necessary.
> I've made a sweep some time ago and found the following culprits:
> * ProcessShutDownThread, used in the shutdown of TM/JM
> * StreamPrinter, used by python operators
> * SplitReader, used ContinuousFileReaderOperator
> * CallExecute, used in DataStreamUtils#collect
> In addition we should check for usages of ExecutorServices that don't provide an explicit ThreadFactory (which would allow explicit names).
> I pushed a hotfix today which resolved another occurrence in the MetricRegistry.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)