[jira] Resolved: (HADOOP-2989) NameNode and JobTracker use bind address to set configuration variables.

["Todd Lipcon (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/HADOOP-2989?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Todd Lipcon resolved HADOOP-2989.
---------------------------------

    Resolution: Fixed

This appears to have been fixed in more recent versions.

> NameNode and JobTracker use bind address to set configuration variables.
> ------------------------------------------------------------------------
>
>                 Key: HADOOP-2989
>                 URL: https://issues.apache.org/jira/browse/HADOOP-2989
>             Project: Hadoop Core
>          Issue Type: Bug
>          Components: dfs, mapred
>            Reporter: Raghu Angadi
>            Priority: Minor
>
> e.g. in NameNode.java : {code}
> // The rpc-server port can be ephemeral... ensure we have the correct info
> this.nameNodeAddress = this.server.getListenerAddress(); 
> conf.set("fs.default.name", nameNodeAddress.getHostName() + ":" + nameNodeAddress.getPort());{code}
> The configuration is changed to help MiniMRCluster I think. This would not be always be correct if bind address is "0.0.0.0". See HADOOP-2974.
> I suggest using "127.0.0.1" when bind address is "0.0.0.0". This is not much of an issue now since this configuration change is used only in tests I think.
> Same fix might be needed for JobTracker and TaskTracker.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.