Return-Path: 
 <hbase-dev-return-6715-apmail-hadoop-hbase-dev-archive=hadoop.apache.org@hadoop.apache.org>
Delivered-To: apmail-hadoop-hbase-dev-archive@locus.apache.org
Received: (qmail 16248 invoked from network); 22 Dec 2008 18:39:07 -0000
Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2)
  by minotaur.apache.org with SMTP; 22 Dec 2008 18:39:07 -0000
Received: (qmail 11905 invoked by uid 500); 22 Dec 2008 18:39:07 -0000
Delivered-To: apmail-hadoop-hbase-dev-archive@hadoop.apache.org
Received: (qmail 11888 invoked by uid 500); 22 Dec 2008 18:39:07 -0000
Mailing-List: contact hbase-dev-help@hadoop.apache.org; run by ezmlm
Precedence: bulk
List-Help: <mailto:hbase-dev-help@hadoop.apache.org>
List-Unsubscribe: <mailto:hbase-dev-unsubscribe@hadoop.apache.org>
List-Post: <mailto:hbase-dev@hadoop.apache.org>
List-Id: <hbase-dev.hadoop.apache.org>
Reply-To: hbase-dev@hadoop.apache.org
Delivered-To: mailing list hbase-dev@hadoop.apache.org
Received: (qmail 11877 invoked by uid 99); 22 Dec 2008 18:39:07 -0000
Received: from nike.apache.org (HELO nike.apache.org) (192.87.106.230)
    by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 22 Dec 2008 10:39:07 -0800
X-ASF-Spam-Status: No, hits=-2000.0 required=10.0
	tests=ALL_TRUSTED
X-Spam-Check-By: apache.org
Received: from [140.211.11.140] (HELO brutus.apache.org) (140.211.11.140)
    by apache.org (qpsmtpd/0.29) with ESMTP; Mon, 22 Dec 2008 18:39:05 +0000
Received: from brutus (localhost [127.0.0.1])
	by brutus.apache.org (Postfix) with ESMTP id 3E945234C463
	for <hbase-dev@hadoop.apache.org>; Mon, 22 Dec 2008 10:38:44 -0800 (PST)
Message-ID: <1158768971.1229971124255.JavaMail.jira@brutus>
Date: Mon, 22 Dec 2008 10:38:44 -0800 (PST)
From: "Jim Kellerman (JIRA)" <jira@apache.org>
To: hbase-dev@hadoop.apache.org
Subject: [jira] Commented: (HBASE-543) A region's state is kept in several
 places in the master opening the possibility for race conditions
In-Reply-To: <110325218.1206397644432.JavaMail.jira@brutus>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
X-Virus-Checked: Checked by ClamAV on apache.org


    [ https://issues.apache.org/jira/browse/HBASE-543?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12658619#action_12658619 ] 

Jim Kellerman commented on HBASE-543:
-------------------------------------

A separate problem I found is related to the regionsToClose map:

{code}
  private final Map<String, Map<byte [], HRegionInfo>> regionsToClose =
    new ConcurrentHashMap<String, Map<byte [], HRegionInfo>>();
{code}

Note that the internal map is a HashMap. HashMaps do not work with byte[] as a key because the hash value will be different for two byte arrays, even if they have the same content. If byte[] is the key, then you must use a TreeMap with a Bytes.BYTES_COMPARATOR. Unless the TreeMap with comparator is used operations on the map such as get(), containsKey(), etc., will not work.

> A region's state is kept in several places in the master opening the possibility for race conditions
> ----------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-543
>                 URL: https://issues.apache.org/jira/browse/HBASE-543
>             Project: Hadoop HBase
>          Issue Type: Bug
>          Components: master
>    Affects Versions: 0.1.0, 0.1.1, 0.2.0
>            Reporter: Jim Kellerman
>            Assignee: Jim Kellerman
>             Fix For: 0.19.0
>
>         Attachments: 543.patch, 543.patch, recent-changes.patch, regionstate.txt
>
>
> A region's state exists in multiple maps in the RegionManager: unassignedRegions, pendingRegions, regionsToClose, closingRegions, regionsToDelete, etc.
> One of these race conditions was found in HBASE-534.
> For HBase-0.1.x, we should just patch the holes we find.
> The ultimate solution (which requires a lot of changes in HMaster) should be applied to HBase trunk.
> Proposed solution:
> Create a class that encapsulates a region's state and provide synchronized access to the class that validates state changes.
> There should be a single structure that holds regions in these transitional states and it should be a synchronized collection of some kind.

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