You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "chunhui shen (JIRA)" <ji...@apache.org> on 2012/06/18 05:56:42 UTC
[jira] [Updated] (HBASE-6227) SSH and cluster startup causes data
loss
[ https://issues.apache.org/jira/browse/HBASE-6227?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
chunhui shen updated HBASE-6227:
--------------------------------
Description:
In AssignmentManager#processDeadServersAndRegionsInTransition, if servershutdownhandler is processing and master consider it cluster startup, master will assign all user regions, however, servershutdownhandler has not completed splitting logs.
Detail process as the following( 1 master, 1 regionservers):
1.All the servers are killed
2.master start up, regionserver's zk node exists, so master will record it
3.master won't split any logs
4.regionserver start up, and master assign root and meta
5.the old regionserver is in SSH now, when rebuild user region, no region is added in master's region list
6.master consider it is cluster up, so it will assign all user regions
7.However, some region's data is in logs which are splittting by SSH
was:In AssignmentManager#processDeadServersAndRegionsInTransition, if servershutdownhandler is processing and master consider it cluster startup, master will assign all user regions, however, servershutdownhandler has not completed splitting logs.
> SSH and cluster startup causes data loss
> -----------------------------------------
>
> Key: HBASE-6227
> URL: https://issues.apache.org/jira/browse/HBASE-6227
> Project: HBase
> Issue Type: Bug
> Components: master
> Reporter: chunhui shen
> Assignee: chunhui shen
>
> In AssignmentManager#processDeadServersAndRegionsInTransition, if servershutdownhandler is processing and master consider it cluster startup, master will assign all user regions, however, servershutdownhandler has not completed splitting logs.
> Detail process as the following( 1 master, 1 regionservers):
> 1.All the servers are killed
> 2.master start up, regionserver's zk node exists, so master will record it
> 3.master won't split any logs
> 4.regionserver start up, and master assign root and meta
> 5.the old regionserver is in SSH now, when rebuild user region, no region is added in master's region list
> 6.master consider it is cluster up, so it will assign all user regions
> 7.However, some region's data is in logs which are splittting by SSH
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators: https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira