You are viewing a plain text version of this content. The canonical link for it is here.

Posted to mapreduce-issues@hadoop.apache.org by "Jiahongchao (JIRA)" <ji...@apache.org> on 2015/03/24 11:37:52 UTC

[jira] [Commented] (MAPREDUCE-6290) job.setJar not working for jars on hdfs

    [ https://issues.apache.org/jira/browse/MAPREDUCE-6290?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14377628#comment-14377628 ] 

Jiahongchao commented on MAPREDUCE-6290:
----------------------------------------

https://issues.apache.org/jira/browse/MAPREDUCE-4408
It seems that Job.setJar should do so

> job.setJar not working for jars on hdfs
> ---------------------------------------
>
>                 Key: MAPREDUCE-6290
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6290
>             Project: Hadoop Map/Reduce
>          Issue Type: Bug
>    Affects Versions: 2.5.0
>         Environment: hadoop run on centos 6.4,eclipse run on windows 7
>            Reporter: Jiahongchao
>
> when I used job.setJar("hdfs://dev01:9000/root/mrjob-0.0.1.jar") to start MR job on remote hadoop cluster, I got errors saying class not found.
> However I noticed comments said "If the job jar is already in fs, we don't need to copy it from local fs", so I think by design setJar method should work for jars that exist on HDFS.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)