[
https://issues.apache.org/jira/browse/HADOOP-5330?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12689834#action_12689834
]
Devaraj Das commented on HADOOP-5330:
-------------------------------------
Do you know what happened to the jobs to which these attempts belonged? Also,
if the task logs (syslog,stderr,stdout) still exist, what do they say? One
suspicion is that the tasks got OutOfMemory errors and they got hung somewhere
in the process of exiting (like in the task's finally block). The TaskTrackers
marked those tasks as FAILED due to lack of ping messages from the tasks.
> Zombie tasks remain after jobs finish/fail/get killed
> -----------------------------------------------------
>
> Key: HADOOP-5330
> URL: https://issues.apache.org/jira/browse/HADOOP-5330
> Project: Hadoop Core
> Issue Type: Bug
> Affects Versions: 0.19.1
> Reporter: Nathan Marz
>
> I'm seeing a lot of "task attempts" around our hadoop cluster for jobs that
> are no longer around. The attempts seem to be "hung", as they sit there
> forever. Additionally, they seem to take up map and reduce slots in the
> cluster unless MapReduce is restarted. This causes real jobs to be unable to
> utilize the whole cluster.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
