hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Lars Hofhansl (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-2195) Support cyclic replication
Date Thu, 01 Sep 2011 05:42:10 GMT

    [ https://issues.apache.org/jira/browse/HBASE-2195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13095133#comment-13095133
] 

Lars Hofhansl commented on HBASE-2195:
--------------------------------------

The ClusterId as byte part actually comes from the replication code.
( See ReplicationSource.clusterId and HLogKey.{get|set}ClusterId(...) )

You "label" the peer cluster when you add it with ReplicationAdmin.addPeer. And the "ClusterId"
here is just the ZK identifier for the cluster for replication purposes.
It just means we never have a replication cycle > 256 clusters, which seems OK.

Since this ends up in every log entry, it should be kept small. And since HLogKey already
has this byte, it wouldn't be a format change either.

Maybe J-D can chime in?


> Support cyclic replication
> --------------------------
>
>                 Key: HBASE-2195
>                 URL: https://issues.apache.org/jira/browse/HBASE-2195
>             Project: HBase
>          Issue Type: Sub-task
>          Components: replication
>            Reporter: Jean-Daniel Cryans
>
> We need to support cyclic replication by using the cluster id of each HlogKey and stop
replicating when it goes back to the original cluster.

--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira

        

Mime
View raw message