[
https://issues.apache.org/jira/browse/MAPREDUCE-803?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Allen Wittenauer resolved MAPREDUCE-803.
----------------------------------------
Resolution: Duplicate
This is pretty much YARN-2131 in modern terms.
> Provide a command line option to clean up jobtracker system directory
> ---------------------------------------------------------------------
>
> Key: MAPREDUCE-803
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-803
> Project: Hadoop Map/Reduce
> Issue Type: Improvement
> Components: jobtracker
> Reporter: Hemanth Yamijala
>
> When the JT is restarted, the mapreduce system directory's contents are used
> for job recovery. For sites that use this feature, there might be instances
> when we don't want to restart to read the mapred system directory. A sample
> use case is if there is a full cluster restart with a (typically minor)
> version upgrade of the Map/Reduce code base. To easily support such cases, it
> would be nice to provide a way for clean up the jobtracker system directory
> so that no files will be available for cleanup.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
