Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (nike.apache.org: domain of s5alye@gmail.com designates
 209.85.210.172 as permitted sender)
MIME-Version: 1.0
Date: Tue, 5 Jul 2011 13:07:22 -0400
Message-ID: 
 <CAGXoe2Vv8-JaPDVmoS2uXOiKeLdQRdj=KLv7saSwanJpgb8sZg@mail.gmail.com>
Subject: How to make a node an exact replica of another node ?
From: A J <s5alye@gmail.com>
To: user@cassandra.apache.org
Content-Type: text/plain; charset=ISO-8859-1

Hello,
Let me explain what I am trying to do:
I am prototyping 2 Data centers (DC1 and DC2) with two nodes each. Say
DC1_n1 and DC1_n2 nodes in DC1 and DC2_n1 and DC2_n2 in DC2.
With PropertyFileSnitch and NetworkTopologyStrategy and
'strategy_options of DC1=1 and DC2=1', I am able to ensure that each
DC has ONE AND ONLY ONE copy of ALL the keys.
But how do I ensure that there is one to one mapping between the
nodes? i.e. how to I ensure keys in DC1_n1 = keys in DC2_n1 and keys
in DC1_n2 = keys in DC2_n2 ?
Right now some keys of DC1_n1 are getting replicated to DC2_n1 and
some to DC2_n2 (and the same for DC1_n2 keys).

I am using RPP and calculating tokens to ensure that each node has 25%
of the load (using the python script at
http://www.datastax.com/docs/0.8/operations/clustering to come up with
the tokens)

Thanks for any inputs.