Mailing-List: contact notifications-help@accumulo.apache.org; run by ezmlm
Precedence: bulk
Reply-To: jira@apache.org
Date: Tue, 19 Nov 2013 16:35:21 +0000 (UTC)
From: "Sean Busbey (JIRA)" <jira@apache.org>
To: notifications@accumulo.apache.org
Message-ID: <JIRA.12653806.1371675713804.98937.1384878921699@arcas>
In-Reply-To: <JIRA.12653806.1371675713804@arcas>
References: <JIRA.12653806.1371675713804@arcas>
Subject: [jira] [Commented] (ACCUMULO-1524) Abstract file pointers in
 metadata table
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/ACCUMULO-1524?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13826651#comment-13826651 ] 

Sean Busbey commented on ACCUMULO-1524:
---------------------------------------

You can already do this in HDFS via NameServices (independent of using a NameService with HDFS Federation). For example, HA set ups use it to abstract the fail-over between NameNodes from DFS clients.

> Abstract file pointers in metadata table
> ----------------------------------------
>
>                 Key: ACCUMULO-1524
>                 URL: https://issues.apache.org/jira/browse/ACCUMULO-1524
>             Project: Accumulo
>          Issue Type: Improvement
>          Components: tserver
>            Reporter: Christopher Tubbs
>            Priority: Minor
>
> ACCUMULO-118 interprets full URIs for file entries. It may be useful to instead to store a VolumeID + path, with a separate VolumeID -> URI mapping in configuration, for a level of indirection. This helps keep the METADATA from needing to be rewritten in strange cases, like moving a namenode from one host to another, or DNS name/IP address of namenode changes, or switching to a redundant fail-over DFS instance.


--
This message was sent by Atlassian JIRA
(v6.1#6144)