[
https://issues.apache.org/jira/browse/HBASE-2195?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13095131#comment-13095131
]
Andrew Purtell commented on HBASE-2195:
---------------------------------------
bq. It takes 128bits to represent uuid in bits only. Is that too much per edit
to carry replicating?
Better to group edits and provide the UUID once in the wrapper perhaps?
ReplEdit, like WALEdit? At least would cut down from 128 bits every edit always.
bq. It'd for sure be 'safe' such that cluster ids would never 'clash' because
someone messed up the mappings of clusterid to some short form.
It is reasonable to prefer safety, especially if we are going to be supporting
arbitrary (or at least flexible) replication topologies as a feature.
> 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
