[
https://issues.apache.org/jira/browse/HBASE-11467?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14117874#comment-14117874
]
stack commented on HBASE-11467:
-------------------------------
[~mantonov] is latest patch up on rb?
Why a setRegistry in ClusterConnection? Why not just read from config or on
construction?
On the higher-level question, this patch makes a radical change -- disruptive
-- in folks expectations undoing zk as first port-of-call connecting to a
cluster instead putting in place a cluster of masters as the cluster id string.
Needs surfacing on dev list for discussion. 2.0 hbase at least. How would we
migrate a zk-based cluster to a master-based one?
Could we take the bulk of this patch and keep zk string as the minimum to
connect to a cluster until we put stake in the ground for topology in 'scaling
1M regions' and while ruminating on implications of going from zk to master
quorum?
> New impl of Registry interface not using ZK + new RPCs on master protocol
> -------------------------------------------------------------------------
>
> Key: HBASE-11467
> URL: https://issues.apache.org/jira/browse/HBASE-11467
> Project: HBase
> Issue Type: Sub-task
> Components: Client, Consensus, Zookeeper
> Affects Versions: 2.0.0
> Reporter: Mikhail Antonov
> Assignee: Mikhail Antonov
> Fix For: 2.0.0
>
> Attachments: HBASE-11467.patch, HBASE-11467.patch, HBASE-11467.patch
>
>
> Currently there' only one implementation of Registry interface, which is
> using ZK to get info about meta. Need to create implementation which will be
> using RPC calls to master the client is connected to.
> Review of early version of patch is here: https://reviews.apache.org/r/24296/
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
