[rabbitmq-discuss] amqpcpp | Temp Queue

[Jesper Louis Andersen]

On Wed, Sep 4, 2013 at 7:11 PM, antony <antony.pulicken at gmail.com> wrote:

> We have a scneario where in the consumer will create a temperory queue and
> wait on that queue till it gets a message from the publisher.
>

I am curious... when is such a scenario needed? I don't think creating and
destroying queues like this is going to be efficient in the long run, but
it is a *very* common pattern in certain concurrency settings: You pass a
one-message mailbox and synchronize on the response appearing in that
mailbox. I've done this in several Haskell programs, but there I tend to
reuse the mailbox so I don't run the overhead of creating/destroying it all
the time as well.

Isn't there a clever way you can set up reply-queues and use correlation
id's to achieve the same?


-- 
J.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130905/eb430772/attachment.htm>