[
https://issues.apache.org/jira/browse/HBASE-16342?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Joseph updated HBASE-16342:
---------------------------
Description:
HBASE-16342 Added Replication WAL offset tracking out of ZooKeeper and into an
HBase table. Provided new implementation of ReplicationQueues and
ReplicationQueuesClient that implement this. Peer tracking still remains inside
of ZooKeeper though. Includes H-Base issues: HBASE-15883, HBASE-15958,
HBASE-15974, HBASE-16036, HBASE-16083, HBASE-16080, HBASE-1620, and HBASE-15937
Tested these changes on Branch-1 with IntegrationTestReplication and it appears
stable. Can also handle late master startups and replication table regions
going down as long as only system tables are hosted on Master and
hbase:replication is hosted on Master.
was:HBASE-16342 Added Replication WAL offset tracking out of ZooKeeper and
into an HBase table. Provided new implementation of ReplicationQueues and
ReplicationQueuesClient that implement this. Peer tracking still remains inside
of ZooKeeper though. Includes H-Base issues: HBASE-15883, HBASE-15958,
HBASE-15974, HBASE-16036, HBASE-16083, HBASE-16080, HBASE-1620, and HBASE-15937
> Backport replication tracking via system table to Branch-1
> ----------------------------------------------------------
>
> Key: HBASE-16342
> URL: https://issues.apache.org/jira/browse/HBASE-16342
> Project: HBase
> Issue Type: Sub-task
> Components: Replication
> Reporter: Joseph
> Assignee: Joseph
> Attachments: HBASE-16342-branch-1.patch
>
>
> HBASE-16342 Added Replication WAL offset tracking out of ZooKeeper and into
> an HBase table. Provided new implementation of ReplicationQueues and
> ReplicationQueuesClient that implement this. Peer tracking still remains
> inside of ZooKeeper though. Includes H-Base issues: HBASE-15883, HBASE-15958,
> HBASE-15974, HBASE-16036, HBASE-16083, HBASE-16080, HBASE-1620, and
> HBASE-15937
> Tested these changes on Branch-1 with IntegrationTestReplication and it
> appears stable. Can also handle late master startups and replication table
> regions going down as long as only system tables are hosted on Master and
> hbase:replication is hosted on Master.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
