[
https://issues.apache.org/jira/browse/HBASE-18444?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16099284#comment-16099284
]
Ajay Jadhav commented on HBASE-18444:
-------------------------------------
[~ndimiduk] Yes, only one cluster (primary) is allowed to perform write
operations while all others (replica) will be in read-only mode. This way
replica cannot corrupt the data present on storage (S3).
The specific use case for this is to provide high availability i.e. in case an
availability zone goes down, you still have access to data through another
cluster present in different availability zone.
More information is available here-
https://aws.amazon.com/blogs/big-data/setting-up-read-replica-clusters-with-hbase-on-amazon-s3/
> Add support for specifying custom meta table suffix
> ---------------------------------------------------
>
> Key: HBASE-18444
> URL: https://issues.apache.org/jira/browse/HBASE-18444
> Project: HBase
> Issue Type: Improvement
> Affects Versions: 1.0.0, 2.0.0
> Reporter: Ajay Jadhav
> Assignee: Ajay Jadhav
> Priority: Minor
> Labels: patch
> Fix For: 1.4.0
>
> Attachments: HBASE-18444.branch-1.001.patch
>
>
> This patch provides user an ability to have a custom meta table suffix.
> In order to share a common rootDir amongst multiple HBase clusters, we need
> separate meta table for each cluster.
> With custom meta table suffix, the meta table is named in the form of
> meta_{suffix} making it specific to a cluster.
> Users can specify the meta table suffix as part of hbase-site.xml.
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
