You are viewing a plain text version of this content. The canonical link for it is here.

Posted to issues@hbase.apache.org by "Viraj Jasani (Jira)" <ji...@apache.org> on 2020/12/22 08:52:00 UTC

[jira] [Commented] (HBASE-25429) After droped the table, using snapshot recovery to cause data consistency problems

    [ https://issues.apache.org/jira/browse/HBASE-25429?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17253342#comment-17253342 ] 

Viraj Jasani commented on HBASE-25429:
--------------------------------------

This seems duplicate of HBASE-25428. Let me close this. Please feel free to reopen if the purpose of this Jira is different than HBASE-25428.

Thanks

> After droped the table, using snapshot recovery to cause data consistency problems
> ----------------------------------------------------------------------------------
>
>                 Key: HBASE-25429
>                 URL: https://issues.apache.org/jira/browse/HBASE-25429
>             Project: HBase
>          Issue Type: Bug
>            Reporter: wxzkenny
>            Priority: Major
>
> We discovered a consistency problem during use HBase. The cause of the problem should be that after the table is dropped, when the snapshot is restored, in some cases the wal log will be applied, thereby the wal log of the dropped table restored to the restored table, because their region id are the same.
> Steps to reproduce
> 1. Create a tablea
> 2. Make a snapshot1
> 3. Insert some records r1
> 4. Drop the table
> 5. Restore snapshot 1
> 6. Down RS or move the region, etc., wal is reapplied
> You may see the data r1 in the snapshot1 table. The problem is not a stable recurrence, but there is a certain probability that it can recur.



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