Return-Path: X-Original-To: apmail-hbase-issues-archive@www.apache.org Delivered-To: apmail-hbase-issues-archive@www.apache.org Received: from mail.apache.org (hermes.apache.org [140.211.11.3]) by minotaur.apache.org (Postfix) with SMTP id 6F8CCE2FE for ; Sat, 8 Dec 2012 17:17:22 +0000 (UTC) Received: (qmail 18675 invoked by uid 500); 8 Dec 2012 17:17:22 -0000 Delivered-To: apmail-hbase-issues-archive@hbase.apache.org Received: (qmail 18496 invoked by uid 500); 8 Dec 2012 17:17:21 -0000 Mailing-List: contact issues-help@hbase.apache.org; run by ezmlm Precedence: bulk List-Help: List-Unsubscribe: List-Post: List-Id: Delivered-To: mailing list issues@hbase.apache.org Received: (qmail 18484 invoked by uid 99); 8 Dec 2012 17:17:21 -0000 Received: from arcas.apache.org (HELO arcas.apache.org) (140.211.11.28) by apache.org (qpsmtpd/0.29) with ESMTP; Sat, 08 Dec 2012 17:17:21 +0000 Date: Sat, 8 Dec 2012 17:17:21 +0000 (UTC) From: "ramkrishna.s.vasudevan (JIRA)" To: issues@hbase.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Commented] (HBASE-7307) MetaReader.tableExists should not return false if the specified table regions has been split MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-JIRA-FingerPrint: 30527f35849b9dde25b450d4833f0394 [ https://issues.apache.org/jira/browse/HBASE-7307?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13527198#comment-13527198 ] ramkrishna.s.vasudevan commented on HBASE-7307: ----------------------------------------------- I think this is a very good find.. Rajesh, i have a question here.. In your case tableExists() was called before the split had done the DaughterOpener part. Which means that the META is going to have only one entry and and daughter entries are yet to be added, correct... Nice one.. I think it should be easy to add up a testcase to this.. I think even if we find a region to be splitParent still add the region to the visitors result.. So that i have a size > 1. > MetaReader.tableExists should not return false if the specified table regions has been split > -------------------------------------------------------------------------------------------- > > Key: HBASE-7307 > URL: https://issues.apache.org/jira/browse/HBASE-7307 > Project: HBase > Issue Type: Bug > Affects Versions: 0.92.3, 0.96.0, 0.94.4 > Reporter: rajeshbabu > Assignee: rajeshbabu > Fix For: 0.92.2, 0.94.3 > > > If a region is split parent we are not adding it to META scan results during full scan. > {code} > if (!isInsideTable(this.current, tableNameBytes)) return false; > if (this.current.isSplitParent()) return true; > // Else call super and add this Result to the collection. > super.visit(r); > {code} > If all regions of a table has been split then result size will be zero and returning false. > {code} > fullScan(catalogTracker, visitor, getTableStartRowForMeta(tableNameBytes)); > // If visitor has results >= 1 then table exists. > return visitor.getResults().size() >= 1; > {code} > Even table is present we are returning false which is not correct(its highly possible in case of tables with one region). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira