hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] Updated: (HBASE-3574) Move ServerLoad off the heartbeat; prereq of hbase-1502
Date Tue, 01 Mar 2011 07:05:37 GMT

     [ https://issues.apache.org/jira/browse/HBASE-3574?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel

stack updated HBASE-3574:

    Attachment: 3574.txt

This patch serializes HServerLoad into the regionserver ephemeral node instead of passing
the load over the heartbeat.  It starts a Chore inside regionserver to update the znode every
30 seconds.

I'm not going to finish this patch.  To complete, I'd have to make investment in translating
znode content into HServerInfo, a class deprecated by patch over in hbase-1502.  Instead,
I'm going to close this issue and roll it into hbase-1502.

> Move ServerLoad off the heartbeat; prereq of hbase-1502
> -------------------------------------------------------
>                 Key: HBASE-3574
>                 URL: https://issues.apache.org/jira/browse/HBASE-3574
>             Project: HBase
>          Issue Type: Task
>            Reporter: stack
>         Attachments: 3574.txt
> The master ui and shell report on cluster load by reading the serverload delivered by
each RS on heartbeat.  HBASE-1502 is removing heartbeat.  Need to do serverload otherwise.
 I'm thinking of writing it instead into the RS ephemeral node and having the master UI, etc.
read it from zk instead.

This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira


View raw message