Mailing-List: contact hadoop-dev-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hadoop-dev@lucene.apache.org
Message-ID: <14314498.1162330997929.JavaMail.root@brutus>
Date: Tue, 31 Oct 2006 13:43:17 -0800 (PST)
From: "Yoram Arnon (JIRA)" <jira@apache.org>
To: hadoop-dev@lucene.apache.org
Subject: [jira] Commented: (HADOOP-225) tasks are left over when a job fails
In-Reply-To: <14555788.1147888325909.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

    [ http://issues.apache.org/jira/browse/HADOOP-225?page=comments#action_12446067 ] 
            
Yoram Arnon commented on HADOOP-225:
------------------------------------

Not sure:
- I haven't been monitoring this as much lately.
- JT crashes too frequently to get a decent statistic

Let's check in a few days

Yoram 


> tasks are left over when a job fails
> ------------------------------------
>
>                 Key: HADOOP-225
>                 URL: http://issues.apache.org/jira/browse/HADOOP-225
>             Project: Hadoop
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.2.0
>            Reporter: Yoram Arnon
>            Priority: Minor
>
> when jobs are stopped or otherwise fail, tasks are often left around.
> the job tracker shows that there are map or reduce (mostly reduce) tasks running, when no job is running.
> these accumulate over time.
> eventually there are so many of those, that the job tracker can't launch new tasks, requiring a restart of the MR cluster.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://issues.apache.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira