[jira] [Resolved] (MESOS-149) Garbage collection on slaves

["Vinod Kone (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/MESOS-149?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Vinod Kone resolved MESOS-149.
------------------------------

    Resolution: Fixed
      Assignee: Vinod Kone

This is on trunk
                
> Garbage collection on slaves
> ----------------------------
>
>                 Key: MESOS-149
>                 URL: https://issues.apache.org/jira/browse/MESOS-149
>             Project: Mesos
>          Issue Type: Improvement
>          Components: slave
>            Reporter: Thomas Marshall
>            Assignee: Vinod Kone
>   Original Estimate: 720h
>  Remaining Estimate: 720h
>
> Currently, there is no garbage collection of completed executors/frameworks on slaves, leading to local directories periodically filling up. 
> Design proposal:
> - A libprocess process that gets messages when an executor has gotten cleaned up and sets some timeout when it should clean up the directory. Abstracting it this way is nice because we could add other capabilities like copying the directory output to HDFS if requested by the user.
> - Making sure directories continue to get cleaned up after a slave restarts 
> - Make sure frameworks use their executor's working directory for storage rather than a hard-configured path (e.g. /tmp)

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira