Mailing-List: contact core-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
Reply-To: core-dev@hadoop.apache.org
Message-ID: <1313101441.1213982505112.JavaMail.jira@brutus>
Date: Fri, 20 Jun 2008 10:21:45 -0700 (PDT)
From: "Steve Loughran (JIRA)" <jira@apache.org>
To: core-dev@hadoop.apache.org
Subject: [jira] Commented: (HADOOP-3579) Support property groups in Hadoop
 configuration
In-Reply-To: <441114688.1213696185113.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HADOOP-3579?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12606832#action_12606832 ] 

Steve Loughran commented on HADOOP-3579:
----------------------------------------

I see why you'd like a group/instance model for this particular part of the deployment (the queues), but I'm thinking of a generic hierarchical view, where we'd have other sections for, say, hostname->IPAddress mappings, or even the deployment parameters of a heartbeat MapReduce job you'd feed through tasktrackers just to keep an eye on their health.

Equally importantly, I'm thinking of what to put in as a back end above and beyond an XML format. Whether its JSON, LDAP or something else, a simple hierarchy with   no notion of type and no duplicates between child section names and child properties wold be good. Its a prequisite to doing cross references with notation such as ../queues/queue1/timeout and /hosts/test/server and avoiding full XPath-like expressions such as:

/group[@name="queues"]/instance[@name="queue"]/property[@name="timeout"/

While such expressions are wonderful in XSL, they are not what you want to reliably refer to other bits in the tree.

> Support property groups in Hadoop configuration
> -----------------------------------------------
>
>                 Key: HADOOP-3579
>                 URL: https://issues.apache.org/jira/browse/HADOOP-3579
>             Project: Hadoop Core
>          Issue Type: Improvement
>          Components: conf
>            Reporter: Hemanth Yamijala
>
> Hadoop configuration is currently a list of key and value pairs. There are some use-cases to support configuring groups of related properties. There could also be multiple instances of such groups. The issue is for adding support for such configuration.

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