[
https://issues.apache.org/jira/browse/HADOOP-4018?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12625481#action_12625481
]
Devaraj Das commented on HADOOP-4018:
-------------------------------------
Dhruba, which version of hadoop did you see this behavior with? In 18 there is
a fix for one such memory hog - HADOOP-3670. There is a discussion there on
using the profiler and tuning the GC parameters.
Also, could you please minimize the number of completed jobs kept in memory per
user. Specify a very low value for mapred.jobtracker.completeuserjobs.maximum
(defaults to 100 jobs per user). HADOOP-3150 will further help to reduce the
amount of memory consumed by the JT since it removes the task promotion queue.
Could you please give some more details - like the number of tasks the job had,
how many such jobs could you run before the JT started to exhibit the problem,
etc?
> limit memory usage in jobtracker
> --------------------------------
>
> Key: HADOOP-4018
> URL: https://issues.apache.org/jira/browse/HADOOP-4018
> Project: Hadoop Core
> Issue Type: Bug
> Components: mapred
> Reporter: dhruba borthakur
> Assignee: dhruba borthakur
>
> We have seen instances when a user submitted a job with many thousands of
> mappers. The JobTracker was running with 3GB heap, but it was still not
> enough to prevent memory trashing from Garbage collection; effectively the
> Job Tracker was not able to serve jobs and had to be restarted.
> One simple proposal would be to limit the maximum number of tasks per job.
> This can be a configurable parameter. Is there other things that eat huge
> globs of memory in job Tracker?
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
