Mailing-List: contact hadoop-dev-help@lucene.apache.org; run by ezmlm
Precedence: bulk
Reply-To: hadoop-dev@lucene.apache.org
Message-ID: <29830454.1164214563913.JavaMail.jira@brutus>
Date: Wed, 22 Nov 2006 08:56:03 -0800 (PST)
From: "Owen O'Malley (JIRA)" <jira@apache.org>
To: hadoop-dev@lucene.apache.org
Subject: [jira] Commented: (HADOOP-639) task cleanup messages can get lost,
 causing task trackers to keep tasks forever
In-Reply-To: <23380746.1161813796565.JavaMail.root@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit

    [ http://issues.apache.org/jira/browse/HADOOP-639?page=comments#action_12451988 ] 
            
Owen O'Malley commented on HADOOP-639:
--------------------------------------

To clarify, the point of the type ids in the TaskTrackerActions is so that we can serialize the list of actions efficiently.

I'd make TaskTrackerAction an abstract class like:

abstract class TaskTrackerAction implements Writable {
  /** Get the type id for a type */
  public abstract int getTypeId();

  /** Create an instance of the appropriate child type. */
  public static TaskTrackerAction newInstance(int typeId);
}

and then the children look like:

class StartTaskAction extends TaskTrackerAction {
  int getTypeId() { return 1; }
  Task newTask;
  public void readFields(...) throws IOException {...}
  public void write(...) throws IOException {...}
}

KillTaskAction has a jobid and taskid.
KillJobAction has a jobid.

We also need a action to reinitialize the task tracker (ReinitializeSelfAction?).

"JTResponse" is too information free. How about "TaskTrackerActionList" instead.

I think we are better off using a boolean than a special value for ackNo. As a side effect we can use short for ackNo because it is ok if it rolls around. And taking a hint from linux, we are probably better starting at a -100 so that we wrap pretty soon after start up. *smile*

> task cleanup messages can get lost, causing task trackers to keep tasks forever
> -------------------------------------------------------------------------------
>
>                 Key: HADOOP-639
>                 URL: http://issues.apache.org/jira/browse/HADOOP-639
>             Project: Hadoop
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.7.2
>            Reporter: Owen O'Malley
>         Assigned To: Arun C Murthy
>             Fix For: 0.9.0
>
>
> If the pollForTaskWithClosedJob call from a job tracker to a task tracker times out when a job completes, the tasks are never cleaned up. This can cause the mini m/r cluster to hang on shutdown, but also is a resource leak.

-- 
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