[
https://issues.apache.org/jira/browse/HBASE-5583?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13616178#comment-13616178
]
ramkrishna.s.vasudevan commented on HBASE-5583:
-----------------------------------------------
Thinking on the problems of having two znodes is that
-> Table is ENABLED but before the status znode is cleared the Master goes
down. Now the new master may see that the znode is ENABLED but the status
znode will remain there. We now have to check for every table if the status
znode is present and if the table is in ENABLED state we need to delete the
znode.
-> The other case is - while the table creation was going on theere was a
failure in table creation and so we go and try to delete the status and znode
and also the table znode. Now assume the master goes down just after the
status znode deletion. Now the new master knows it is a partially created
table but it will not know what action to take if the status znode is not
available. According to me this is the only scenario where having 2 status
znodes will be problematic.
-> The advantage of having only one znode is, our entire recovery just depends
on the state present in the table znode. If the table is in any of the state
other than ENABLED/ENABLING, DISABLING,DISABLED we go ahead with the existing
code. If the znode is any of the the new states that is introduced by this
patch then master would be clear in knowing what action to take.
But introducing new state to the new existing table znode - am not sure if
others will accept to it. So some feedback on this part will help me to
proceed with the patch.
> Master restart on create table with splitkeys does not recreate table with
> all the splitkey regions
> ---------------------------------------------------------------------------------------------------
>
> Key: HBASE-5583
> URL: https://issues.apache.org/jira/browse/HBASE-5583
> Project: HBase
> Issue Type: Bug
> Reporter: ramkrishna.s.vasudevan
> Assignee: ramkrishna.s.vasudevan
> Fix For: 0.95.0
>
> Attachments: HBASE-5583_new_1.patch, HBASE-5583_new_2.patch,
> HBASE-5583_new_4_WIP.patch
>
>
> -> Create table using splitkeys
> -> MAster goes down before all regions are added to meta
> -> On master restart the table is again enabled but with less number of
> regions than specified in splitkeys
> Anyway client will get an exception if i had called sync create table. But
> table exists or not check will say table exists.
> Is this scenario to be handled by client only or can we have some mechanism
> on the master side for this? Pls suggest.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
