Return-Path: Delivered-To: apmail-hadoop-hbase-dev-archive@locus.apache.org Received: (qmail 95192 invoked from network); 26 Jun 2008 04:13:44 -0000 Received: from hermes.apache.org (HELO mail.apache.org) (140.211.11.2) by minotaur.apache.org with SMTP; 26 Jun 2008 04:13:44 -0000 Received: (qmail 17298 invoked by uid 500); 26 Jun 2008 04:13:45 -0000 Delivered-To: apmail-hadoop-hbase-dev-archive@hadoop.apache.org Received: (qmail 17279 invoked by uid 500); 26 Jun 2008 04:13:45 -0000 Mailing-List: contact hbase-dev-help@hadoop.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Reply-To: hbase-dev@hadoop.apache.org Delivered-To: mailing list hbase-dev@hadoop.apache.org Received: (qmail 17268 invoked by uid 99); 26 Jun 2008 04:13:45 -0000 Received: from athena.apache.org (HELO athena.apache.org) (140.211.11.136) by apache.org (qpsmtpd/0.29) with ESMTP; Wed, 25 Jun 2008 21:13:45 -0700 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; Thu, 26 Jun 2008 04:12:55 +0000 Received: from brutus (localhost [127.0.0.1]) by brutus.apache.org (Postfix) with ESMTP id 41F75234C14F for ; Wed, 25 Jun 2008 21:12:45 -0700 (PDT) Message-ID: <1080640403.1214453565269.JavaMail.jira@brutus> Date: Wed, 25 Jun 2008 21:12:45 -0700 (PDT) From: "stack (JIRA)" To: hbase-dev@hadoop.apache.org Subject: [jira] Assigned: (HBASE-710) If clocks are way off, then we can have daughter split come before rather than after its parent in .META. In-Reply-To: <1660144810.1214441206745.JavaMail.jira@brutus> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-Virus-Checked: Checked by ClamAV on apache.org [ https://issues.apache.org/jira/browse/HBASE-710?page=3Dcom.atlassian= .jira.plugin.system.issuetabpanels:all-tabpanel ] stack reassigned HBASE-710: --------------------------- Assignee: stack > If clocks are way off, then we can have daughter split come before rather= than after its parent in .META. > -------------------------------------------------------------------------= -------------------------------- > > Key: HBASE-710 > URL: https://issues.apache.org/jira/browse/HBASE-710 > Project: Hadoop HBase > Issue Type: Bug > Reporter: stack > Assignee: stack > Priority: Blocker > Fix For: 0.1.3 > > Attachments: hbase-710.patch > > > On the Jon Gray cluster, his clocks are skewed badly. I see weird stuff = in .META. > {code} > 2008-06-25 14:57:57,728 DEBUG org.apache.hadoop.hbase.HMaster: HMaster.me= taScanner regioninfo: {regionname: items,823ce1e3-d414-474f-ac70-c4081cecef= 0f,1214416614697, startKey: <823ce1e3-d414-474f-ac70-c4081cecef0f>, endKey:= <86f8df20-e237-4bb3-9748-88cef892bd70>, encodedName: 1157924217, offline: = true, split: true, tableDesc: {name: items, families: {cfrecs:=3D{name: cfr= ecs, max versions: 2, compression: NONE, in memory: false, max length: 2147= 483647, bloom filter: none}, clusters:=3D{name: clusters, max versions: 2, = compression: NONE, in memory: false, max length: 2147483647, bloom filter: = none}, content:=3D{name: content, max versions: 2, compression: NONE, in me= mory: false, max length: 2147483647, bloom filter: none}, readby:=3D{name: = readby, max versions: 2, compression: NONE, in memory: false, max length: 2= 147483647, bloom filter: none}, receivedby:=3D{name: receivedby, max versio= ns: 2, compression: NONE, in memory: false, max length: 2147483647, bloom f= ilter: none}, savedby:=3D{name: savedby, max versions: 2, compression: NONE= , in memory: false, max length: 2147483647, bloom filter: none}, sentby:=3D= {name: sentby, max versions: 2, compression: NONE, in memory: false, max le= ngth: 2147483647, bloom filter: none}}}}, server: 192.168.249.223:60020, st= artCode: 1214406344634 > 2008-06-25 14:57:57,732 DEBUG org.apache.hadoop.hbase.HMaster: HMaster.me= taScanner regioninfo: {regionname: items,823ce1e3-d414-474f-ac70-c4081cecef= 0f,1214416641213, startKey: <823ce1e3-d414-474f-ac70-c4081cecef0f>, endKey:= <83fca0e2-f324-4f9e-99c1-1fdbeff63b3d>, encodedName: 541300165, tableDesc:= {name: items, families: {cfrecs:=3D{name: cfrecs, max versions: 2, compres= sion: NONE, in memory: false, max length: 2147483647, bloom filter: none}, = clusters:=3D{name: clusters, max versions: 2, compression: NONE, in memory:= false, max length: 2147483647, bloom filter: none}, content:=3D{name: cont= ent, max versions: 2, compression: NONE, in memory: false, max length: 2147= 483647, bloom filter: none}, readby:=3D{name: readby, max versions: 2, comp= ression: NONE, in memory: false, max length: 2147483647, bloom filter: none= }, receivedby:=3D{name: receivedby, max versions: 2, compression: NONE, in = memory: false, max length: 2147483647, bloom filter: none}, savedby:=3D{nam= e: savedby, max versions: 2, compression: NONE, in memory: false, max lengt= h: 2147483647, bloom filter: none}, sentby:=3D{name: sentby, max versions: = 2, compression: NONE, in memory: false, max length: 2147483647, bloom filte= r: none}}}}, server: 192.168.249.220:60020, startCode: 1214424347649 > 2008-06-25 14:57:57,738 DEBUG org.apache.hadoop.hbase.HMaster: HMaster.me= taScanner regioninfo: {regionname: items,823ce1e3-d414-474f-ac70-c4081cecef= 0f,1214434560891, startKey: <823ce1e3-d414-474f-ac70-c4081cecef0f>, endKey:= <9066d4f3-314b-4d9c-90e8-7aa08a52fdd4>, encodedName: 1673833201, offline: = true, split: true, tableDesc: {name: items, families: {cfrecs:=3D{name: cfr= ecs, max versions: 2, compression: NONE, in memory: false, max length: 2147= 483647, bloom filter: none}, clusters:=3D{name: clusters, max versions: 2, = compression: NONE, in memory: false, max length: 2147483647, bloom filter: = none}, content:=3D{name: content, max versions: 2, compression: NONE, in me= mory: false, max length: 2147483647, bloom filter: none}, readby:=3D{name: = readby, max versions: 2, compression: NONE, in memory: false, max length: 2= 147483647, bloom filter: none}, receivedby:=3D{name: receivedby, max versio= ns: 2, compression: NONE, in memory: false, max length: 2147483647, bloom f= ilter: none}, savedby:=3D{name: savedby, max versions: 2, compression: NONE= , in memory: false, max length: 2147483647, bloom filter: none}, sentby:=3D= {name: sentby, max versions: 2, compression: NONE, in memory: false, max le= ngth: 2147483647, bloom filter: none}}}}, server: 192.168.249.221:60020, st= artCode: 1214406358315 > {code} > Thats 3 regions with same start code; 2 are offline. > Looking at the regionids -- these are timestamps -- I see that they don't= jibe with how they should be aligned. Parents should come before daughter= s in timestamps. > Looking at clocks on cluster, they are badly skewed: > {code > =09[hbase@mb0 logs]$ for i in 0 1 2 3 4 5 6 7 8 9; do ssh hb$i "h= ostname; date"; done > =09=09hb0.streamy.com > =09=09Wed Jun 25 16:47:29 PDT 2008 > =09=09hb1.streamy.com > =09=09Wed Jun 25 11:47:39 PDT 2008 > =09=09hb2.streamy.com > =09=09Wed Jun 25 11:47:40 PDT 2008 > =09=09hb3.streamy.com > =09=09Wed Jun 25 11:47:26 PDT 2008 > =09=09hb4.streamy.com > =09=09Wed Jun 25 11:47:35 PDT 2008 > =09=09hb5.streamy.com > =09=09Wed Jun 25 16:47:29 PDT 2008 > =09=09hb6.streamy.com > =09=09Wed Jun 25 16:47:29 PDT 2008 > =09=09hb7.streamy.com > =09=09Wed Jun 25 16:47:29 PDT 2008 > =09=09hb8.streamy.com > =09=09Wed Jun 25 16:47:30 PDT 2008 > =09=09hb9.streamy.com > =09=09Wed Jun 25 16:47:30 PDT 2008 > {code} > Looking at split code, looks like the regionserver sets the regionid/time= stamp on the new daughter regions inside in the HRegionInfo constructor tha= t gets called when splitting: > {code} > this.regionId =3D System.currentTimeMillis(); > {code} > Daughters update the .META. table; they need to have a basic check that t= hey are not inserting with a timestamp that is older than the parent they a= re splitting. > This is a bit like HBASE-609 --=20 This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.