hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Ted Yu (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-2195) Support cyclic replication
Date Thu, 08 Sep 2011 05:41:09 GMT

    [ https://issues.apache.org/jira/browse/HBASE-2195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13100088#comment-13100088

Ted Yu commented on HBASE-2195:

bq. Hopefully Replication{Source|Sink} never have to know about that.

So far our test implicitly assumes that all the clusters participating in replication are
not older than version -1 (i.e. version 0).

See ReplicationSource.readAllEntriesToReplicateOrNextFile():
+          if (HConstants.DEFAULT_CLUSTER_ID == logKey.getClusterId()) {
+            logKey.setClusterId(this.clusterId);
+          }
I think there might be a problem if version of logKey is 0 in the above case because logKey
might not be local.

Having getVersion() allows us to make better judgment.

I will publish version 13 of the patch.

BTW I noticed there is an int version inside readFields() while VERSION is declared final
in HLogKey - this happens when two people work on different parts of the same file :-)

> 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
>            Assignee: Lars Hofhansl
>         Attachments: 2195-v10.txt, 2195-v12.txt, 2195-v5.txt, 2195-v6.txt, 2195.txt
> 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


View raw message