Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Date: Sat, 31 Oct 2015 07:16:27 +0000 (UTC)
From: "ramkrishna.s.vasudevan (JIRA)" <jira@apache.org>
To: issues@hbase.apache.org
Message-ID: <JIRA.12721577.1402977165000.115369.1446275787931@Atlassian.JIRA>
In-Reply-To: <JIRA.12721577.1402977165000@Atlassian.JIRA>
References: <JIRA.12721577.1402977165000@Atlassian.JIRA>
 <JIRA.12721577.1402977165505@arcas>
Subject: [jira] [Commented] (HBASE-11368) Multi-column family BulkLoad fails
 if compactions go on too long
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit


    [ https://issues.apache.org/jira/browse/HBASE-11368?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14983879#comment-14983879 ] 

ramkrishna.s.vasudevan commented on HBASE-11368:
------------------------------------------------

This comment https://issues.apache.org/jira/browse/HBASE-11368?focusedCommentId=14693166&page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#comment-14693166 is getting addressed as part of HBASE-13082.  So doing that JIRA would mean that any current on going scan will not be able to see the bulk loaded hfiles which is loaded just after the current scan has started. I think that behaviour should be acceptable, right?

> Multi-column family BulkLoad fails if compactions go on too long
> ----------------------------------------------------------------
>
>                 Key: HBASE-11368
>                 URL: https://issues.apache.org/jira/browse/HBASE-11368
>             Project: HBase
>          Issue Type: Bug
>            Reporter: stack
>            Assignee: Qiang Tian
>         Attachments: hbase-11368-0.98.5.patch, hbase11368-master.patch, key_stacktrace_hbase10882.TXT, performance_improvement_verification_98.5.patch
>
>
> Compactions take a read lock.  If a multi-column family region, before bulk loading, we want to take a write lock on the region.  If the compaction takes too long, the bulk load fails.
> Various recipes include:
> + Making smaller regions (lame)
> + [~victorunique] suggests major compacting just before bulk loading over in HBASE-10882 as a work around.
> Does the compaction need a read lock for that long?  Does the bulk load need a full write lock when multiple column families?  Can we fail more gracefully at least?


--
This message was sent by Atlassian JIRA
(v6.3.4#6332)