hbase-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Enis Soztutar (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HBASE-11569) Flush / Compaction handling from secondary region replicas
Date Wed, 15 Apr 2015 19:10:00 GMT

    [ https://issues.apache.org/jira/browse/HBASE-11569?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14496719#comment-14496719

Enis Soztutar commented on HBASE-11569:

{{writestate.readOnly}} means that the region cannot accept regular writes. {{!writestate.readOnly}}
thus means that the region can accept regular writes and can do flushes. 
In case the region is opened as a secondary replica, readOnly is set to true. The region cannot
accept regular writes, but it can still accept writes coming from replay (from the primary
replica). In this case though, it should still not flush even if there are edits in the memstore.

> Flush / Compaction handling from secondary region replicas
> ----------------------------------------------------------
>                 Key: HBASE-11569
>                 URL: https://issues.apache.org/jira/browse/HBASE-11569
>             Project: HBase
>          Issue Type: Sub-task
>            Reporter: Enis Soztutar
>            Assignee: Enis Soztutar
>             Fix For: 2.0.0, 1.1.0
>         Attachments: hbase-11569-master-v3.patch, hbase-11569-v5.patch
> We should be handling flushes and compactions from the primary region replica being replayed
to the secondary region replica via HBASE-11568. 
> Some initial thoughts for how can this be done is discussed in HBASE-11183. More details
will come together with the patch. 

This message was sent by Atlassian JIRA

View raw message