hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Hudson (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-17658) Bookkeeping issue in BaseLoadBalancer for Table Skew
Date Fri, 17 Feb 2017 19:59:41 GMT

    [ https://issues.apache.org/jira/browse/HBASE-17658?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15872422#comment-15872422
] 

Hudson commented on HBASE-17658:
--------------------------------

SUCCESS: Integrated in Jenkins build HBase-Trunk_matrix #2522 (See [https://builds.apache.org/job/HBase-Trunk_matrix/2522/])
HBASE-17658 Fix bookkeeping error with max regions for a table (tedyu: rev 938aef772d82f7a52228436b141e62af66e11e76)
* (edit) hbase-server/src/test/java/org/apache/hadoop/hbase/master/balancer/TestBaseLoadBalancer.java
* (edit) hbase-server/src/main/java/org/apache/hadoop/hbase/master/balancer/BaseLoadBalancer.java


> Bookkeeping issue in BaseLoadBalancer for Table Skew
> ----------------------------------------------------
>
>                 Key: HBASE-17658
>                 URL: https://issues.apache.org/jira/browse/HBASE-17658
>             Project: HBase
>          Issue Type: Bug
>          Components: Balancer
>            Reporter: Tim Brown
>            Assignee: Tim Brown
>             Fix For: 2.0.0, 1.4.0
>
>         Attachments: JIRA-17658.master.001.patch
>
>
> Currently when numMaxRegionsPerTable of a given table increases above the current maximum,
the value is not properly updated. This means that the cost for Table Skew cannot get worse
for a given region move during the balancer process leading to an imbalanced cluster with
respect to Table Skew.



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Mime
View raw message