[jira] [Commented] (HADOOP-11764) Hadoop should have the option to use directory other than tmp for extracting and loading leveldbjni

["Allen Wittenauer (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HADOOP-11764?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14530966#comment-14530966 ] 

Allen Wittenauer commented on HADOOP-11764:
-------------------------------------------

bq. This can also mean *-env.sh.

I'm not sure how you got that from the statement "until it can proven otherwise that this can't be handled via *-site.xml."  I specifically mean: push this into the XML files and leave the shell environment alone. As we discussed at Summit--and you tentatively agreed--populating the system property for leveldb from the *-site.xml files could be done at service start time, given that there are generic abstractions for actually starting daemons in Hadoop that large chunks of the system already use.

It's worth pointing out that putting the leveldb settings in a generic shell env will impact end users, and it would have to be done in daemon specific env vars.  So no, one can't modify HADOOP_OPTS here.  You'll break apps that use their own leveldb bits.




> Hadoop should have the option to use directory other than tmp for extracting and loading leveldbjni
> ---------------------------------------------------------------------------------------------------
>
>                 Key: HADOOP-11764
>                 URL: https://issues.apache.org/jira/browse/HADOOP-11764
>             Project: Hadoop Common
>          Issue Type: Bug
>            Reporter: Anubhav Dhoot
>            Assignee: Anubhav Dhoot
>         Attachments: YARN-3331.001.patch, YARN-3331.002.patch
>
>
> /tmp can be  required to be noexec in many environments. This causes a problem when  nodemanager tries to load the leveldbjni library which can get unpacked and executed from /tmp.



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