[jira] [Commented] (YARN-3331) NodeManager should use directory other than tmp for extracting and loading leveldbjni

["Anubhav Dhoot (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/YARN-3331?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14355557#comment-14355557 ] 

Anubhav Dhoot commented on YARN-3331:
-------------------------------------

As per http://grepcode.com/file/repo1.maven.org/maven2/org.fusesource.leveldbjni/leveldbjni-all/1.8/org/fusesource/hawtjni/runtime/Library.java if we define the property library.${name}.path we can avoid it using the temporary directory. 
{noformat}
The file extraction is attempted until it succeeds in the following directories.
1. The directory pointed to by the "library.${name}.path" System property (if set)
2. a temporary directory (uses the "java.io.tmpdir" System property)
{noformat}

So we can fix this by setting -Dlibrary.leveldbjni.path=$(pwd) in the nodemanager options.

> NodeManager should use directory other than tmp for extracting and loading leveldbjni
> -------------------------------------------------------------------------------------
>
>                 Key: YARN-3331
>                 URL: https://issues.apache.org/jira/browse/YARN-3331
>             Project: Hadoop YARN
>          Issue Type: Bug
>          Components: nodemanager
>            Reporter: Anubhav Dhoot
>            Assignee: Anubhav Dhoot
>
> /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)