hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "stack (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-12035) Client does an RPC to master everytime a region is relocated
Date Sun, 21 Sep 2014 22:56:33 GMT

    [ https://issues.apache.org/jira/browse/HBASE-12035?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14142761#comment-14142761

stack commented on HBASE-12035:

bq ...or because it puts additional requirement on master to "always" be up?

I'm thinking because of the above [~mantonov] Being able to operate -- r/w -- w/o M for some
period of time makes for a less complication all around.

On all you say about colocated meta and master, agree, down to I no longer think we should
have master serve meta nor any regions at all.

Cache of table state in client sounds sensible, better than a trip to the master on each NSRE.

Over in HBASE-11974, was suggested that table state be kept in a 'table' table.  Would make
it so we did not do a trip to the master on each NSRE to figure table state enabled/disabled.
 That'd be better than going to master.  But even this seems wrong asking if table is disable
everytime region moves? Client will go back to meta to figure if region has moved.  Could
meta have a mark in it that says region is enable/disabled?  If disable, means table is disabled.
 This breaks down though when you have 1M regions in a table.  Would be crazy writing 'disabled'
into 1M regions.  Mark one region only, the first one?

> Client does an RPC to master everytime a region is relocated
> ------------------------------------------------------------
>                 Key: HBASE-12035
>                 URL: https://issues.apache.org/jira/browse/HBASE-12035
>             Project: HBase
>          Issue Type: Improvement
>            Reporter: Enis Soztutar
>            Priority: Critical
>             Fix For: 2.0.0
> HBASE-7767 moved table enabled|disabled state to be kept in hdfs instead of zookeeper.
isTableDisabled() which is used in HConnectionImplementation.relocateRegion() now became a
master RPC call rather than a zookeeper client call. Since we do relocateRegion() calls everytime
we want to relocate a region (region moved, RS down, etc) this implies that when the master
is down, the some of the clients for uncached regions will be affected. 
> See HBASE-7767 and HBASE-11974 for some more background. 

This message was sent by Atlassian JIRA

View raw message