Mailing-List: contact user-help@hbase.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@hbase.apache.org
Received-SPF: pass (nike.apache.org: domain of nidmgg@gmail.com designates
 209.85.220.170 as permitted sender)
References: 
 <CANx3uAiiez1DKDvpHjnh4Wz3Oy0OG_A=ODBpnY9kWDQ0kHy_LA@mail.gmail.com>
Mime-Version: 1.0 (1.0)
In-Reply-To: 
 <CANx3uAiiez1DKDvpHjnh4Wz3Oy0OG_A=ODBpnY9kWDQ0kHy_LA@mail.gmail.com>
Content-Type: text/plain;
	charset=us-ascii
Content-Transfer-Encoding: quoted-printable
Message-Id: <6DCA836A-ECA6-4A83-A237-81987CDF9347@gmail.com>
Cc: "user@hbase.apache.org" <user@hbase.apache.org>
From: Demai Ni <nidmgg@gmail.com>
Subject: Re: replication
Date: Sat, 8 Mar 2014 11:55:12 -0800
To: "user@hbase.apache.org" <user@hbase.apache.org>

Your understanding is correct and you can actually put/delete to the same ta=
ble on both cluster. It is called master-master setup(aka active-active in R=
DMS).=20

The only issue is about the sequence of transactions. I am not sure how the t=
imestamp works to ensure the same order from client can be ensured by two cl=
uster

Demai on the run

On Mar 8, 2014, at 11:45 AM, Koert Kuipers <koert@tresata.com> wrote:

> do i understand it correctly that it is safe to have 2 hbase clusters
> replicate to each other (so in both directions)?
>=20
> and as long as an update (put/delete) arrives at only one cluster this
> setup will function correctly?
>=20
> not entirely sure about situation where updates get routed to both
> clusters, but i see no immediate harm if the operations are idempotent...
>=20
> thanks! koert