[jira] [Commented] (YARN-8818) Yarn log aggregation of spark streaming job

["Szilard Nemeth (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/YARN-8818?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16689796#comment-16689796 ] 

Szilard Nemeth commented on YARN-8818:
--------------------------------------

Hi [~ayushChauhan]!

Could you please elaborate on this a little bit more, making analyzing the issue more easier?
 # What was the symptoms you see? What was not working?
 # Could you please provide a way to reproduce the issue? How did you start a Spark job? 

Thanks!

> Yarn log aggregation of spark streaming job
> -------------------------------------------
>
>                 Key: YARN-8818
>                 URL: https://issues.apache.org/jira/browse/YARN-8818
>             Project: Hadoop YARN
>          Issue Type: Bug
>            Reporter: Ayush Chauhan
>            Assignee: Szilard Nemeth
>            Priority: Major
>
> By default, YARN aggregates logs after an application completes. But I am trying to aggregate logs for spark streaming job which in theory will run forever. I have set the following properties for log aggregation and restarted yarn by restarting {{hadoop-yarn-nodemanager}} for core & task nodes and {{hadoop-yarn-resourcemanager}} for master node on my emr cluster. I can view my changes in [http://node-ip:8088/conf].
> {noformat}
> yarn.log-aggregation-enable => true{noformat}
> {noformat}
> yarn.log-aggregation.retain-seconds => 172800{noformat}
> {noformat}
> yarn.log-aggregation.retain-check-interval-seconds => -1 {noformat}
> {noformat}
> yarn.nodemanager.log-aggregation.roll-monitoring-interval-seconds => 3600{noformat}
> All the articles and resources have only mentioned to include {{yarn.nodemanager.log-aggregation.roll-monitoring-interval-seconds}} property and yarn will start aggregating logs for running jobs. But it is not working in my case.
>  



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

---------------------------------------------------------------------
To unsubscribe, e-mail: yarn-issues-unsubscribe@hadoop.apache.org
For additional commands, e-mail: yarn-issues-help@hadoop.apache.org