[ https://issues.apache.org/jira/browse/MAPREDUCE-106?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Allen Wittenauer resolved MAPREDUCE-106.
----------------------------------------
Resolution: Incomplete
LIkely stale.
> The temporary disk space limits should not skip jobs
> ----------------------------------------------------
>
> Key: MAPREDUCE-106
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-106
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Owen O'Malley
>
> In HADOOP-657, we added disk space modeling and throttling. The scheduler currently picks
the next job if there isn't enough space, which can lead to starvation of jobs that require
a lot of space. It should not schedule any task if the first job can't use the space.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
|