hadoop-common-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tahir Hashmi (JIRA)" <j...@apache.org>
Subject [jira] Commented: (HADOOP-1171) NPE in TaskTracker RenitTrackerAction
Date Mon, 07 May 2007 08:26:15 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-1171?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12493992
] 

Tahir Hashmi commented on HADOOP-1171:
--------------------------------------

I went through the code for to figure out why the NPE might have been triggered. It seems
like jobHasFinished() might have issues with synchronisation. There are concurrent threads
calling on this function and there's one call to TaskInProgress.cleanup() which removes TIPs
from the tasks map and is not synchronised.

There may be a race condition when two threads try to cleanup() the same TIP. Need someone
who's familiar with this code to comment on it.

> NPE in TaskTracker RenitTrackerAction 
> --------------------------------------
>
>                 Key: HADOOP-1171
>                 URL: https://issues.apache.org/jira/browse/HADOOP-1171
>             Project: Hadoop
>          Issue Type: Bug
>          Components: mapred
>    Affects Versions: 0.12.1
>            Reporter: Koji Noguchi
>         Assigned To: Tahir Hashmi
>
> TaskTracker log shows =============
> 2007-03-28 02:32:18,076 INFO org.apache.hadoop.mapred.TaskTracker: Recieved RenitTrackerAction
from JobTracker
> 2007-03-28 02:32:18,494 ERROR org.apache.hadoop.mapred.TaskTracker: Can not start task
tracker because java.lang.NullPointerException
>   at org.apache.hadoop.mapred.TaskTracker$TaskInProgress.jobHasFinished(TaskTracker.java:1187)
>   at org.apache.hadoop.mapred.TaskTracker.close(TaskTracker.java:430)
>   at org.apache.hadoop.mapred.TaskTracker.run(TaskTracker.java:917)
>   at org.apache.hadoop.mapred.TaskTracker.main(TaskTracker.java:1589)
> JobTracker log shows ==============
> 2007-03-28 02:31:18,977 INFO org.apache.hadoop.mapred.JobTracker: Lost tracker '____.____.com'
> ...
> 2007-03-28 02:31:18,977 INFO org.apache.hadoop.mapred.JobInProgress: TaskTracker at 'tracker____.___.com'
turned 'flaky'
> ...
> 2007-03-28 02:32:18,075 WARN org.apache.hadoop.mapred.JobTracker: Status from unknown
Tracker : tracker____.___.com:#####

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


Mime
View raw message