[ 
https://issues.apache.org/jira/browse/HBASE-1462?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Jean-Daniel Cryans updated HBASE-1462:
--------------------------------------

    Attachment: hbase-1462-v1.patch

Patch that removes a bunch of getMaster() in HCM (keeping the important ones). 
I tested this patch on a 4 nodes cluster with PE sequentialWrite 5, I killed 
the master once the table was created and waited 1 min before bringing it back 
with no problem. It also passes the tests.

> hclient still seems to depend on master
> ---------------------------------------
>
>                 Key: HBASE-1462
>                 URL: https://issues.apache.org/jira/browse/HBASE-1462
>             Project: Hadoop HBase
>          Issue Type: Bug
>    Affects Versions: 0.20.0
>            Reporter: ryan rawson
>            Assignee: Jean-Daniel Cryans
>             Fix For: 0.20.0
>
>         Attachments: hbase-1462-v1.patch
>
>
> during a master down, but cluster up event, my clients seem to not work.
> clients shouldnt need to talk to the master anymore in 0.20.  We should 
> double check this.

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.

Reply via email to