[jira] [Resolved] (HADOOP-2900) distributedcache: Can we create symlink without mapred.create.symlink = yes?

["Harsh J (Resolved) (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-2900?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Harsh J resolved HADOOP-2900.
-----------------------------

    Resolution: Not A Problem

This behavior was fixed at some point such that now it is:

[filename]#[symlink-name] and symlinks will always be created with given name - no secondary configuration is required.
                
> distributedcache:  Can we create symlink without mapred.create.symlink = yes?
> -----------------------------------------------------------------------------
>
>                 Key: HADOOP-2900
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2900
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: util
>            Reporter: Koji Noguchi
>            Priority: Minor
>
> I always thought 
> <property>
>   <name>mapred.cache.archives</name>
>   <value>/archives/arc1.zip#symlink</value>
> </property>
> would create a symlink on the current working directory, but it also requires 
> <property>
>   <name>mapred.create.symlink</name>
>   <value>yes</value>
> </property>
> to be set.
> Can we simplify it so that we always create a symlink whenever there's '#wwww' entry?
> Or is 'mapred.create.symlink' necessary since '#' is a valid dfs filename?

--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira