hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Jerry He (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-14453) HBaseAdmin#deleteTable should relocate META when cached location is stale
Date Sat, 19 Sep 2015 01:56:04 GMT

    [ https://issues.apache.org/jira/browse/HBASE-14453?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14876824#comment-14876824
] 

Jerry He commented on HBASE-14453:
----------------------------------

Patch looks good.
As you said,  getFirstMetaServerForTable() does nothing to refresh cached meta location. 
This private method is not used anywhere else, and probably can be removed altogether.
The MetaScanner.metaScan(conf, connection, visitor, tableName) uses the usual HTable to scan
meta. I would assume this route would do all the right things.  Correct?

There are so many ways to do the same thing (get meta and scan meta).  I wouldn't be surprised
there is another place that is mis-used.

> HBaseAdmin#deleteTable should relocate META when cached location is stale
> -------------------------------------------------------------------------
>
>                 Key: HBASE-14453
>                 URL: https://issues.apache.org/jira/browse/HBASE-14453
>             Project: HBase
>          Issue Type: Bug
>    Affects Versions: 0.98.14
>            Reporter: Andrew Purtell
>            Assignee: Andrew Purtell
>             Fix For: 0.98.15
>
>         Attachments: HBASE-14453-0.98.patch, HBASE-14453-0.98.patch
>
>
> After HBASE-14275, in HBaseAdmin#deleteTable, when using MetaReader to wait until all
regions are deleted, we won't attempt to relocate META should its cached location be stale.



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

Mime
View raw message