From issues-return-482390-archive-asf-public=cust-asf.ponee.io@hbase.apache.org Sun Aug 29 13:48:02 2021 Return-Path: X-Original-To: archive-asf-public@cust-asf.ponee.io Delivered-To: archive-asf-public@cust-asf.ponee.io Received: from mxout1-ec2-va.apache.org (mxout1-ec2-va.apache.org [3.227.148.255]) by mx-eu-01.ponee.io (Postfix) with ESMTPS id 49F9F180652 for ; Sun, 29 Aug 2021 15:48:02 +0200 (CEST) Received: from mail.apache.org (mailroute1-lw-us.apache.org [207.244.88.153]) by mxout1-ec2-va.apache.org (ASF Mail Server at mxout1-ec2-va.apache.org) with SMTP id 6CB983F0AC for ; Sun, 29 Aug 2021 13:48:01 +0000 (UTC) Received: (qmail 97856 invoked by uid 500); 29 Aug 2021 13:48:01 -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 97844 invoked by uid 99); 29 Aug 2021 13:48:01 -0000 Received: from mailrelay1-he-de.apache.org (HELO mailrelay1-he-de.apache.org) (116.203.21.61) by apache.org (qpsmtpd/0.29) with ESMTP; Sun, 29 Aug 2021 13:48:01 +0000 Received: from jira2-he-de.apache.org (jira2-he-de.apache.org [168.119.33.54]) by mailrelay1-he-de.apache.org (ASF Mail Server at mailrelay1-he-de.apache.org) with ESMTPS id 2272D3E8AE for ; Sun, 29 Aug 2021 13:48:00 +0000 (UTC) Received: from jira2-he-de.apache.org (localhost.localdomain [127.0.0.1]) by jira2-he-de.apache.org (ASF Mail Server at jira2-he-de.apache.org) with ESMTP id 080EFC800FB for ; Sun, 29 Aug 2021 13:48:00 +0000 (UTC) Date: Sun, 29 Aug 2021 13:48:00 +0000 (UTC) From: "Duo Zhang (Jira)" To: issues@hbase.apache.org Message-ID: In-Reply-To: References: Subject: [jira] [Updated] (HBASE-26235) We could start RegionServerTracker before becoming active master 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-26235?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ] Duo Zhang updated HBASE-26235: ------------------------------ Description: The direct reason is that, after decoupling, HMaster does not extends HRegionServer, so we do not have RegionServerAddressTracker now, so for backup masters we do not have a way to get all the region server which means it can not serve as connection registry any more, so we need to find a way to let backup masters to also have a way to get the region server list. But in general, since we already have a class in HMaster which is used to track region servers, we'd better reuse it. And after this change, the backup masters could also be used as the fallback servers for getting region server list in HBASE-26220, which could also reduce the load of active master. And this could also speed up the active master start process as we could save one zookeeper scan when becoming active master. was: The direct reason is that, after decoupling, HMaster does not extends HRegionServer, so we do not have RegionServerAddressTracker now, so for backup masters we do not have a way to get all the region server which means it can not serve as connection registry any more, so we need to find a way to let backup masters to also have a way to get the region server list. But in general, since we already have a class in HMaster which is used to track region servers, we'd better reuse it. And after this change, the backup masters could also be used as the fallback servers for getting region server list in HBASE-26220, which could also reduce the load of active master. > We could start RegionServerTracker before becoming active master > ---------------------------------------------------------------- > > Key: HBASE-26235 > URL: https://issues.apache.org/jira/browse/HBASE-26235 > Project: HBase > Issue Type: Sub-task > Components: master, Zookeeper > Reporter: Duo Zhang > Assignee: Duo Zhang > Priority: Major > > The direct reason is that, after decoupling, HMaster does not extends HRegionServer, so we do not have RegionServerAddressTracker now, so for backup masters we do not have a way to get all the region server which means it can not serve as connection registry any more, so we need to find a way to let backup masters to also have a way to get the region server list. > But in general, since we already have a class in HMaster which is used to track region servers, we'd better reuse it. And after this change, the backup masters could also be used as the fallback servers for getting region server list in HBASE-26220, which could also reduce the load of active master. > And this could also speed up the active master start process as we could save one zookeeper scan when becoming active master. -- This message was sent by Atlassian Jira (v8.3.4#803005)