hadoop-mapreduce-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Owen O'Malley (JIRA)" <j...@apache.org>
Subject [jira] Commented: (MAPREDUCE-1252) Shuffle deadlocks on wrong number of maps
Date Tue, 01 Dec 2009 16:22:20 GMT

    [ https://issues.apache.org/jira/browse/MAPREDUCE-1252?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12784278#action_12784278
] 

Owen O'Malley commented on MAPREDUCE-1252:
------------------------------------------

To be clearer:

1. The previous patch fixes the warnings, but not the problem. Thus, it is necessary but not
sufficient.
2. On a one node cluster with 0.21.0-dev from yesterday:
  a. Have mapred.map.tasks set to 12 in hadoop-site.xml.
  b. Submit a word count example on a two file input directory.
3. The 2 maps run fine.
4. The 2 reduces both lock up at 5.5% with a status of 2 of 12 maps fetched.

Looking at the system directory the job.xml has mapreduce.job.maps = 2, but the task and the
web ui show it as 12.

If I change the config to use mapreduce.job.maps to set the default, everything works correctly.


> Shuffle deadlocks on wrong number of maps
> -----------------------------------------
>
>                 Key: MAPREDUCE-1252
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-1252
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>          Components: task
>    Affects Versions: 0.21.0, 0.22.0
>            Reporter: Owen O'Malley
>            Assignee: Owen O'Malley
>            Priority: Blocker
>             Fix For: 0.21.0, 0.22.0
>
>         Attachments: mr-1252.patch
>
>
> The new shuffle assumes that the number of maps is correct. The new JobSubmitter sets
the old value. Something misfires in the middle causing:
> 09/12/01 00:00:15 WARN conf.Configuration: mapred.job.split.file is deprecated. Instead,
use mapreduce.job.splitfile
> 09/12/01 00:00:15 WARN conf.Configuration: mapred.map.tasks is deprecated. Instead, use
mapreduce.job.maps
> But my reduces got stuck at 2 maps / 12 when there were only 2 maps in the job.

-- 
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