Return-Path: Delivered-To: apmail-hbase-dev-archive@www.apache.org Received: (qmail 11400 invoked from network); 22 Jun 2010 07:23:26 -0000 Received: from unknown (HELO mail.apache.org) (140.211.11.3) by 140.211.11.9 with SMTP; 22 Jun 2010 07:23:26 -0000 Received: (qmail 5776 invoked by uid 500); 22 Jun 2010 07:23:26 -0000 Delivered-To: apmail-hbase-dev-archive@hbase.apache.org Received: (qmail 5530 invoked by uid 500); 22 Jun 2010 07:23:24 -0000 Mailing-List: contact dev-help@hbase.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: dev@hbase.apache.org Delivered-To: mailing list dev@hbase.apache.org Received: (qmail 5522 invoked by uid 99); 22 Jun 2010 07:23:23 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 22 Jun 2010 07:23:23 +0000 X-ASF-Spam-Status: No, hits=1.8 required=10.0 tests=AWL,FH_HELO_EQ_D_D_D_D,MIME_QP_LONG_LINE,SPF_NEUTRAL X-Spam-Check-By: apache.org Received-SPF: neutral (athena.apache.org: local policy) Received: from [75.101.130.251] (HELO ip-10-250-199-31.ec2.internal) (75.101.130.251) by apache.org (qpsmtpd/0.29) with ESMTP; Tue, 22 Jun 2010 07:23:18 +0000 Received: from ip-10-250-199-31.ec2.internal (localhost.localdomain [127.0.0.1]) by ip-10-250-199-31.ec2.internal (Postfix) with ESMTP id 01548EE04; Tue, 22 Jun 2010 07:22:56 +0000 (UTC) Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: quoted-printable Subject: Re: Review Request: Cache flush should go to tmp dir From: "Todd Lipcon" To: "Ryan Rawson" , stack@duboce.net Date: Tue, 22 Jun 2010 07:22:56 -0000 Message-ID: <20100622072256.15081.87917@ip-10-250-199-31.ec2.internal> Cc: "Todd Lipcon" , jiraposter@review.hbase.org, dev@hbase.apache.org In-Reply-To: <20100622071405.15083.32456@ip-10-250-199-31.ec2.internal> References: <20100622071405.15083.32456@ip-10-250-199-31.ec2.internal> > On 2010-06-22 00:14:05, stack wrote: > > +1 (if it passes all tests). Nit-picks below. Cool, I'll run it through my Hudson overnight, plus running a cluster test = on my 5-node test cluster now. Will commit tomorrow midday with changes bel= ow addressed assuming testing goes OK. > On 2010-06-22 00:14:05, stack wrote: > > src/main/java/org/apache/hadoop/hbase/regionserver/Store.java, line 333 > > > > > > ? There was notion of a '_tmp' already? > > = > > I'd say name it '.tmp'... since a '.' prefix seems to be our conven= tion given logs dir at top-level has a '.' prefix. yea, the bulk load stuff uses a tmp dir if you ask the regionserver to load= a file which is stored on a different filesystem than the region itself. I'll switch to .tmp though as you suggest (and move it into HConstants to r= eplace the now-unused compactions.dir or what have you) > On 2010-06-22 00:14:05, stack wrote: > > src/main/java/org/apache/hadoop/hbase/regionserver/Store.java, line 464 > > > > > > Why not keep old name and just move dirs? Why create a new unique = name? I was worried that what's unique in the old dir might not be unique in the = new > On 2010-06-22 00:14:05, stack wrote: > > src/test/java/org/apache/hadoop/hbase/regionserver/TestStore.java, line= 411 > > > > > > Nice test. thx! at some point I'd like to combine this with the one in TestFSErrorsExp= osed and make a little mini-framework out of it... something for next month= perhaps. - Todd ----------------------------------------------------------- This is an automatically generated e-mail. To reply, visit: http://review.hbase.org/r/216/#review261 ----------------------------------------------------------- On 2010-06-21 23:57:39, Todd Lipcon wrote: > = > ----------------------------------------------------------- > This is an automatically generated e-mail. To reply, visit: > http://review.hbase.org/r/216/ > ----------------------------------------------------------- > = > (Updated 2010-06-21 23:57:39) > = > = > Review request for hbase, stack and Ryan Rawson. > = > = > Summary > ------- > = > Fixes bugs where an exception in the middle of flushing a file leaves a h= alf-written StoreFile in the region dir, preventing that region from recove= ring, or, in the case of transient errors, causing silent loss of half a fi= le's worth of data. > = > I also got rid of the compaction dir here, and am just using one region-w= ide tmp dir. Is there some reason this is a bad idea? > = > = > This addresses bug HBASE-2729. > http://issues.apache.org/jira/browse/HBASE-2729 > = > = > Diffs > ----- > = > src/main/java/org/apache/hadoop/hbase/regionserver/HRegion.java 1794df8 = > src/main/java/org/apache/hadoop/hbase/regionserver/Store.java 04b7522 = > src/test/java/org/apache/hadoop/hbase/regionserver/TestCompaction.java = 9e5ca46 = > src/test/java/org/apache/hadoop/hbase/regionserver/TestStore.java a65e9= 47 = > = > Diff: http://review.hbase.org/r/216/diff > = > = > Testing > ------- > = > Ran TestCompaction and TestStore. Will start a cluster test running befor= e I go to bed. > = > = > Thanks, > = > Todd > = >