[rabbitmq-discuss] Persistent messages and durable queue questions

[Matthias Radestock]

Elias,

On 18/07/11 19:30, Elias Levy wrote:
> On Mon, Jul 18, 2011 at 10:02 AM, Matthias Radestock
> <matthias at rabbitmq.com <mailto:matthias at rabbitmq.com>> wrote:
>     We could change things s.t. absent durable queues still participate
>     in routing and that confirms return a 'nack' when messages are
>     routed to such queues.
>
> It seems like that is the correct thing to do.
>
>     I am concerned though that there may be use cases where the current
>     behaviour is the correct one.
>
> I am curious as to what that case may be.

Me too. Alas the behaviour has been like this forever. Even prior to 
confirms, tx.commit behaved in the same way. And to date nobody has 
complained.

It's a lot of work to make this change. So I'd rather be sure that this 
really is the only sensible semantics.


Matthias.