[
https://issues.apache.org/jira/browse/MAPREDUCE-6491?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14945838#comment-14945838
]
Dustin Cote commented on MAPREDUCE-6491:
----------------------------------------
[~aw], I've opened up MAPREDUCE-6504 to basically be the same as this JIRA
except the default value of mapreduce.application.appendable.env.variables will
not include HADOOP_CLASSPATH and made it "a part of" this JIRA. I'll upload
the version modified for trunk there next.
I think as I understood it, MAPREDUCE-6458 is needed before we should drop
HADOOP_CLASSPATH from the default, so I noted that on MAPREDUCE-6504 as well.
I'm happy to take over MAPREDUCE-6458 and will assign to myself. Probably I
won't get to it this week, but I should be able to give it a go over the next
couple of weeks. [~jlowe] and [~aw], did I miss anything left for this JIRA?
I think all that's left is to send in the patch (assuming there's no other
hiccups). Thanks for the help!
> Environment variable handling assumes values should be appended
> ---------------------------------------------------------------
>
> Key: MAPREDUCE-6491
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-6491
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Affects Versions: 2.2.0
> Reporter: Jason Lowe
> Assignee: Dustin Cote
> Attachments: YARN-2369-1.patch, YARN-2369-10.patch,
> YARN-2369-2.patch, YARN-2369-3.patch, YARN-2369-4.patch, YARN-2369-5.patch,
> YARN-2369-6.patch, YARN-2369-7.patch, YARN-2369-8.patch, YARN-2369-9.patch
>
>
> When processing environment variables for a container context the code
> assumes that the value should be appended to any pre-existing value in the
> environment. This may be desired behavior for handling path-like environment
> variables such as PATH, LD_LIBRARY_PATH, CLASSPATH, etc. but it is a
> non-intuitive and harmful way to handle any variable that does not have
> path-like semantics.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
