[jira] [Commented] (HBASE-24552) Replica region needs to check if primary region directory exists at file system in TransitRegionStateProcedure

["Duo Zhang (Jira)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/HBASE-24552?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17220127#comment-17220127 ] 

Duo Zhang commented on HBASE-24552:
-----------------------------------

Found this when backporting HBASE-25206, do we really need this check? Or it is just for safety? The comment seems to indicate that we still have some bugs which lead to this problem?

In general, if we do this, we will leave a region state node in memory with corrupted state, which means we need to fix it through HBCK2? What is the suggested way to fix this problem? I think we need to add comment about all these things so that it will not make users confusing...

Thanks.

> Replica region needs to check if primary region directory exists at file system  in TransitRegionStateProcedure 
> ----------------------------------------------------------------------------------------------------------------
>
>                 Key: HBASE-24552
>                 URL: https://issues.apache.org/jira/browse/HBASE-24552
>             Project: HBase
>          Issue Type: Bug
>          Components: read replicas
>    Affects Versions: 2.3.0
>            Reporter: Huaxiang Sun
>            Assignee: Huaxiang Sun
>            Priority: Major
>             Fix For: 3.0.0-alpha-1, 2.3.0
>
>
> In hbase-1, it always runs into the situation that primary region has been closed/removed and replica region still stays in master's in-memory db and open at one of the region servers. Balancer can move this replica region to a new region server. During the region open, replica region does not check if primary region has been removed and moves forward. During store open, it will recreates primary region directory at hdfs and caused inconsistency.
>  
> In hbase-2, things get much better. To prevent the above inconsistency from happening, it adds more checks for a replica region, i.e, if primary regions' directory exists and there is a .regioninfo under. 



--
This message was sent by Atlassian Jira
(v8.3.4#803005)