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 "Allen Wittenauer (JIRA)" <ji...@apache.org> on 2014/11/13 17:14:33 UTC

[jira] [Commented] (MAPREDUCE-6153) Apply `mapreduce.admin.user.env' to AM

    [ https://issues.apache.org/jira/browse/MAPREDUCE-6153?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14209956#comment-14209956 ] 

Allen Wittenauer commented on MAPREDUCE-6153:
---------------------------------------------

By "include snappy" what exactly do you mean?  Isn't CLASSPATH, java.library.path, and LD_LIBRARY_PATH being set already to include the appropriate dirs?

> Apply `mapreduce.admin.user.env' to AM
> --------------------------------------
>
>                 Key: MAPREDUCE-6153
>                 URL: https://issues.apache.org/jira/browse/MAPREDUCE-6153
>             Project: Hadoop Map/Reduce
>          Issue Type: Improvement
>          Components: applicationmaster
>    Affects Versions: 2.5.0
>            Reporter: bc Wong
>
> Would be nice to be able to manipulate the AM's library path to include snappy. The `mapreduce.admin.user.env' seems perfect for this, except that it only affects the tasks. I think it's useful to have it apply to the AM as well.
> Use case: I have a job that uses the output committer (which runs in the AM) to read the output file of the reducer, which is compressed using snappy.



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