[
https://issues.apache.org/jira/browse/YARN-1111?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13782078#comment-13782078
]
Steve Loughran commented on YARN-1111:
--------------------------------------
investigate how the logs get copied back from YARN containers to HDFS
> NM containerlogs servlet can't handle logs of more than a GB
> ------------------------------------------------------------
>
> Key: YARN-1111
> URL: https://issues.apache.org/jira/browse/YARN-1111
> Project: Hadoop YARN
> Issue Type: Sub-task
> Components: nodemanager
> Affects Versions: 2.1.0-beta
> Environment: Long-lived service generating lots of log data from
> HBase running in debug level
> Reporter: Steve Loughran
> Priority: Minor
>
> If a container is set up to log stdout to a file, the container log servlet
> will list the file
> {code}
> err.txt : Total file length is 551 bytes.
> out.txt : Total file length is 1572099246 bytes.
> {code}
> If you actually click on out.txt then the tail logic takes a *very* long time
> to react. There is also the question of what will happen if the log fills up
> that volume
--
This message was sent by Atlassian JIRA
(v6.1#6144)
