You are viewing a plain text version of this content. The canonical link for it is here.

Posted to dev@reef.apache.org by "Markus Weimer (JIRA)" <ji...@apache.org> on 2017/11/01 18:31:00 UTC

[jira] [Closed] (REEF-470) HDInsight job requesting 160 containers fails on GC overhead limit exceeded

     [ https://issues.apache.org/jira/browse/REEF-470?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Markus Weimer closed REEF-470.
------------------------------
    Resolution: Cannot Reproduce

We've since run much larger Jobs on HDI. We must assume this is solved.

> HDInsight job requesting 160 containers fails on GC overhead limit exceeded
> ---------------------------------------------------------------------------
>
>                 Key: REEF-470
>                 URL: https://issues.apache.org/jira/browse/REEF-470
>             Project: REEF
>          Issue Type: Bug
>          Components: REEF Driver, REEF-Runtime-HDInsight, REEF.NET Driver
>         Environment: REEF.NET on HDInsight
>            Reporter: Andrew Chung
>            Priority: Major
>
> A REEF job requesting 160+ containers is getting killed on HDInsight for spending too much time in GC after we request around 155 containers. It seems like the Java process is using too much memory. Is there any way around this either via a REEF fix or a temporary command line override?



--
This message was sent by Atlassian JIRA
(v6.4.14#64029)