hadoop-common-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Eli Collins (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HADOOP-6605) Add JAVA_HOME detection to hadoop-config
Date Wed, 18 May 2011 20:50:47 GMT

    [ https://issues.apache.org/jira/browse/HADOOP-6605?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13035659#comment-13035659

Eli Collins commented on HADOOP-6605:

bq. a) we're getting into platform specific code without a good reason

The motivation is that users don't shouldn't have to manually set an environment variable
that we can detect automatically. Multiple users have requested this. Why is this not a good
reason? We make improvements both for new users (who want an easier out of the box experience)
and experienced admins (who will manually configure HADOOP_HOME).

bq. b) this could have significant performance ramifications as we add more and more paths
to support more and more OSes

How so?  The new code *only* executes if HADOOP_HOME is not already set, and the code just
checks for the existence of bin/java in a set of directories, it bails as soon as it finds
one. This is not an expensive operation.

bq. c) there is a config option to specifically eliminate the need to do this

The existence of a configuration parameter does not mean we're not allowed to try to detect
a reasonable default if it's not set. We also automatically set a default MAX_HEAP_SIZE and
let users override it, this is not different.

> Add JAVA_HOME detection to hadoop-config
> ----------------------------------------
>                 Key: HADOOP-6605
>                 URL: https://issues.apache.org/jira/browse/HADOOP-6605
>             Project: Hadoop Common
>          Issue Type: Improvement
>            Reporter: Chad Metcalf
>            Assignee: Eli Collins
>            Priority: Minor
>             Fix For: 0.22.0
>         Attachments: HADOOP-6605.patch, hadoop-6605-1.patch
> The commands that source hadoop-config.sh currently bail with an error if JAVA_HOME is
not set. Let's detect JAVA_HOME (from a list of locations on various OS types) if JAVA_HOME
is not already set by hadoop-env.sh or the environment. This way users don't have to manually
configure it.

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message