hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ted Yu (JIRA)" <j...@apache.org>
Subject [jira] [Updated] (HBASE-15486) Avoid multiple disable/enable balancer calls while running rolling-restart.sh --graceful
Date Thu, 24 Mar 2016 13:56:25 GMT

     [ https://issues.apache.org/jira/browse/HBASE-15486?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]

Ted Yu updated HBASE-15486:
---------------------------
      Resolution: Fixed
    Hadoop Flags: Reviewed
          Status: Resolved  (was: Patch Available)

Thanks for the patch, Samir.

> Avoid multiple disable/enable balancer calls while running rolling-restart.sh --graceful

> -----------------------------------------------------------------------------------------
>
>                 Key: HBASE-15486
>                 URL: https://issues.apache.org/jira/browse/HBASE-15486
>             Project: HBase
>          Issue Type: Improvement
>          Components: scripts
>    Affects Versions: 2.0.0, 1.1.4, 1.0.4
>            Reporter: Samir Ahmic
>            Assignee: Samir Ahmic
>            Priority: Minor
>             Fix For: 2.0.0
>
>         Attachments: HBASE-15486_v0.patch, HBASE-15486_v1.patch, HBASE-15486_v2.patch
>
>
> When running rolling-restart.sh --graceful we are using graceful_stop.sh to unload/restart/load
regionserver. For every restarted  regionserver  we are making two calls in graceful-stop.sh
to hbase shell to disable/enable balancer. This means if we have 100 regionservers to restart
by rolling-restart.sh --graceful we will make 200 calls to hbase shell. Instead i propose
that we add optional parameter in graceful_stop.sh that will tell script not to manage balancer
and manage balancer states in rolling-restart.sh when running with --graceful parameter. 



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message