[
https://issues.apache.org/jira/browse/HBASE-11393?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14986421#comment-14986421
]
Enis Soztutar commented on HBASE-11393:
---------------------------------------
Can we change the TableCF to be something like:
{code}
+message TableCF {
+ optional TableName table_name = 1;
+ repeated bytes family = 2;
+}
{code}
This is more in-line with the rest of the protos.
Did you handle the case where there is existing replication peers? We should
allow (and maybe auto-convert) existing replication peers with the old format.
Historically, we have written PB structures to disk using a magic prefix to
differentiate between old-writable formats and new PB format. Maybe we can do a
similar thing here as well.
> Replication TableCfs should be a PB object rather than a string
> ---------------------------------------------------------------
>
> Key: HBASE-11393
> URL: https://issues.apache.org/jira/browse/HBASE-11393
> Project: HBase
> Issue Type: Sub-task
> Reporter: Enis Soztutar
> Fix For: 2.0.0
>
> Attachments: HBASE-11393.patch, HBASE-11393_v1.patch,
> HBASE-11393_v2.patch, HBASE-11393_v3.patch, HBASE-11393_v4.patch,
> HBASE-11393_v5.patch
>
>
> We concatenate the list of tables and column families in format
> "table1:cf1,cf2;table2:cfA,cfB" in zookeeper for table-cf to replication peer
> mapping.
> This results in ugly parsing code. We should do this a PB object.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
