[rabbitmq-discuss] Question on Rabbit MQ

Tim Watson watson.timothy at gmail.com
Mon Oct 15 08:35:53 BST 2012


On 13 Oct 2012, at 19:52, Chinmay Sahoo wrote:

> Hi,
> I like erlang and thus the interest in RabbitMQ.
> 1. Can I use any other database(apart from Mnesia) for persistence ?

Mnesia is used to persist schema information (e.g., vhosts, queues, exchanges, bindings, users, etc) and no, you cannot use another database for that at this time. You should note that Mnesia is *not* used to persist messages - instead rabbit has its own disk based persistence mechanism, and you *can* replace this, although the only alternative store that I know of currently is Tokyo Cabinet.  

> 2. Does Queue Data replication exists ?

Yes - please see http://www.rabbitmq.com/distributed.html and http://www.rabbitmq.com/ha.html for more details on some of these options.

> 3. How well does RabbitMQ confirms to the ONCE-ONLY philosophy. Will
> RabbitMQ send a same message to the consumer twice ? Cause this can be
> a problem in some areas such as payment processing where the message
> would be processed twice unless the scenario is somehow handled in
> consumer.

Yes indeed. See http://rabbitmq.1065348.n5.nabble.com/Exactly-Once-Delivery-td16826.html for a good explanation of why in our view, handling this in the consumer is *necessary* in most cases.

> 4. I will test RabbitMQ on Windows but would like to have it on Unix in Prod.
> Are there any tools to help migrate the instance of RabbitMQ from
> Windows to Unix ?

I certainly haven't heard of any. Are you talking about migrating the definitions of queues/exchanges/bindings/etc, or persistent messages, or both?

More information about the rabbitmq-discuss mailing list