You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@hbase.apache.org by "Duo Zhang (JIRA)" <ji...@apache.org> on 2018/09/18 01:41:00 UTC
[jira] [Commented] (HBASE-21202) Investigate sync-replication
[ https://issues.apache.org/jira/browse/HBASE-21202?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16618366#comment-16618366 ]
Duo Zhang commented on HBASE-21202:
-----------------------------------
I think the correct place to implement the sync replication related wal logic is on top of the WAL interface. We will setup two WAL instances for a region, and write to them all. Now the logic is implemented at the AsyncWriter layer.
But I think the current sync method should be changed to return a CompletableFuture instead of waiting forever until succeeded, otherwise it will hurt the performance of sync replication.
> Investigate sync-replication
> ----------------------------
>
> Key: HBASE-21202
> URL: https://issues.apache.org/jira/browse/HBASE-21202
> Project: HBase
> Issue Type: Sub-task
> Reporter: Josh Elser
> Priority: Major
>
> Need to understand what the synchronous replication feature requires out of a WAL and make sure that the changes made to support the LogService are in line with how the synchronous replication feature works.
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)