[rabbitmq-discuss] Consumers from slave queues on different nodes and duplicate messages

[Simon MacMullen]

On 12/01/12 16:12, Dave Stevens wrote:
> I am thinking specifically about the case when consumer A is consuming
> from slave queue on node X and consumer B is consuming from slave
> queue on node Y.

This doesn't really happen.

A can consume from node X, and B from node Y, *but* internally to the 
cluster both are connected to the master node. That's pretty much what 
makes it the master. The reason for this is exactly the concern you 
raise :-)

So publishes go to all nodes, consumes just come from the master (which 
then informs the slaves so they drop the messages).

Cheers, Simon

-- 
Simon MacMullen
RabbitMQ, VMware