Mailing-List: contact user-help@cassandra.apache.org; run by ezmlm
Precedence: bulk
Reply-To: user@cassandra.apache.org
Received-SPF: pass (athena.apache.org: domain of dnd1066@gmail.com designates
 209.85.215.172 as permitted sender)
Message-ID: <50044FE3.9060207@gmail.com>
Date: Mon, 16 Jul 2012 18:31:15 +0100
From: Guy Incognito <dnd1066@gmail.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; WOW64;
 rv:13.0) Gecko/20120614 Thunderbird/13.0.1
MIME-Version: 1.0
To: user@cassandra.apache.org
Subject: Truncate failing with 1.0 client against 0.7 cluster
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

i'm doing an upgrade of Cassandra 0.7 to 1.0 at the moment, and as part 
of the preparation i'm upgrading to 1.0 client libraries (we use Hector 
1.0-5) prior to upgrading the cluster itself.  I'm seeing some of our 
integration tests against the dev 0.7 cluster fail as they get 
UnavailableExceptions when trying to truncate the test column families.  
This is new behaviour with the 1.0 client libraries, it doesn't happen 
with the 0.7 libraries.

It seems to fail immediately, it doesn't eg wait for eg the 10 second 
RPC timeout, it fails straight away.  Anyone have any ideas as to what 
may be happening?  Interestingly I seem to be able to get around it if i 
only tell Hector about one of the nodes (we have 4). If I give it all 
four then it throws the UnavailableException.