[
https://issues.apache.org/jira/browse/PHOENIX-4442?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16282690#comment-16282690
]
Ethan Wang commented on PHOENIX-4442:
-------------------------------------
[~vincentpoon] I see.
bq. That way queued replication edits could continue to drain at the HBase
level on the destination, after which the table could be dropped.
regarding that, I think we may also use hbase pre-disable table hook. Actually,
if I understand right, even you immediately drop a table from HBase, it still
has to go through "this draining process", which will slowly replay the
remaining WAL(since already ack-ed) and flush them. Is that already the case?
> Rename system catalog INDEX_STATE to STATE
> ------------------------------------------
>
> Key: PHOENIX-4442
> URL: https://issues.apache.org/jira/browse/PHOENIX-4442
> Project: Phoenix
> Issue Type: Improvement
> Reporter: Ethan Wang
>
> Rename system catalog INDEX_STATE to STATE so that this column can be used
> for table as well
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
