You are viewing a plain text version of this content. The canonical link for it is here.
Posted to mapreduce-dev@hadoop.apache.org by "Vinod K V (JIRA)" <ji...@apache.org> on 2009/10/23 11:49:59 UTC

[jira] Created: (MAPREDUCE-1142) When cache.local.size is hit on one disk, DistributedCacheManager aggressively deletes old files on other disks too.

When cache.local.size is hit on one disk, DistributedCacheManager aggressively deletes old files on other disks too.
--------------------------------------------------------------------------------------------------------------------

                 Key: MAPREDUCE-1142
                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1142
             Project: Hadoop Map/Reduce
          Issue Type: Bug
          Components: tasktracker
    Affects Versions: 0.20.2, 0.21.0, 0.22.0
            Reporter: Vinod K V


When the size of distributed cache files goes beyond {{mapreduce.tasktracker.cache.local.size}} on one disk, {{TrackerDistributedCacheManager}} deletes old files on other disks too. May be this is a bit aggressive.

An important problem with this excess effort is that issue observed at MAPREDUCE-1141 will be more often.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.