Return-Path: X-Original-To: apmail-hbase-dev-archive@www.apache.org Delivered-To: apmail-hbase-dev-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 4EA3C1747B for ; Fri, 1 May 2015 03:17:08 +0000 (UTC) Received: (qmail 61822 invoked by uid 500); 1 May 2015 03:17:07 -0000 Delivered-To: apmail-hbase-dev-archive@hbase.apache.org Received: (qmail 61702 invoked by uid 500); 1 May 2015 03:17:06 -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 61686 invoked by uid 99); 1 May 2015 03:17:06 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Fri, 01 May 2015 03:17:06 +0000 Date: Fri, 1 May 2015 03:17:06 +0000 (UTC) From: "Enis Soztutar (JIRA)" To: dev@hbase.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Created] (HBASE-13605) RegionStates should not keep its list of dead servers MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 Enis Soztutar created HBASE-13605: ------------------------------------- Summary: RegionStates should not keep its list of dead servers Key: HBASE-13605 URL: https://issues.apache.org/jira/browse/HBASE-13605 Project: HBase Issue Type: Bug Reporter: Enis Soztutar Assignee: Enis Soztutar Fix For: 2.0.0, 1.0.2, 1.1.1 As mentioned in https://issues.apache.org/jira/browse/HBASE-9514?focusedCom= mentId=3D13769761&page=3Dcom.atlassian.jira.plugin.system.issuetabpanels:co= mment-tabpanel#comment-13769761 and HBASE-12844 we should have only 1 sourc= e of cluster membership.=20 The list of dead server and RegionStates doing it's own liveliness check (S= erverManager.isServerReachable()) has caused an assignment problem again in= a test cluster where the region states "thinks" that the server is dead an= d SSH will handle the region assignment. However the RS is not dead at all,= living happily, and never gets zk expiry or YouAreDeadException or anythin= g. This leaves the list of regions unassigned in OFFLINE state.=20 master assigning the region: {code} 15-04-20 09:02:25,780 DEBUG [AM.ZK.Worker-pool3-t330] master.RegionStates: = Onlined 77dddcd50c22e56bfff133c0e1f9165b on os-amb-r6-us-1429512014-hbase4-= 6.novalocal,16020,1429520535268 {ENCODED =3D> 77dddcd50c {code} Master then disabled the table, and unassigned the region: {code} 2015-04-20 09:02:27,158 WARN [ProcedureExecutorThread-1] zookeeper.ZKTable= StateManager: Moving table loadtest_d1 state from DISABLING to DISABLING Starting unassign of loadtest_d1,,1429520544378.77dddcd50c22e56bfff133c0e1= f9165b. (offlining), current state: {77dddcd50c22e56bfff133c0e1f9165b state= =3DOPEN, ts=3D1429520545780, server=3Dos-amb-r6-us-1429512014-hbase4-6.no= valocal,16020,1429520535268} bleProcedure$BulkDisabler-0] master.AssignmentManager: Sent CLOSE to os-amb= -r6-us-1429512014-hbase4-6.novalocal,16020,1429520535268 for region loadtes= t_d1,,1429520544378.77dddcd50c22e56bfff133c0e1f9165b. 2015-04-20 09:02:27,414 INFO [AM.ZK.Worker-pool3-t316] master.RegionStates= : Offlined 77dddcd50c22e56bfff133c0e1f9165b from os-amb-r6-us-1429512014-hb= ase4-6.novalocal,16020,1429520535268 {code} On table re-enable, AM does not assign the region:=20 {code} 2015-04-20 09:02:30,415 INFO [ProcedureExecutorThread-3] balancer.BaseLoad= Balancer: Reassigned 25 regions. 25 retained the pre-restart assignment.=C2= =B7 2015-04-20 09:02:30,415 INFO [ProcedureExecutorThread-3] procedure.EnableT= ableProcedure: Bulk assigning 25 region(s) across 5 server(s), retainAssign= ment=3Dtrue l,16000,1429515659726-GeneralBulkAssigner-4] master.RegionStates: Couldn't = reach online server os-amb-r6-us-1429512014-hbase4-6.novalocal,16020,142952= 0535268 l,16000,1429515659726-GeneralBulkAssigner-4] master.AssignmentManager: Upda= ting the state to OFFLINE to allow to be reassigned by SSH nmentManager: Skip assigning loadtest_d1,,1429520544378.77dddcd50c22e56bfff= 133c0e1f9165b., it is on a dead but not processed yet server: os-amb-r6-us-= 1429512014-hbase4-6.novalocal,16020,1429520535268 {code} -- This message was sent by Atlassian JIRA (v6.3.4#6332)