[rabbitmq-discuss] Using Load balancer(HA Proxy) while Federating of clusters.

[Tim Watson]

Hi,

On 9 May 2013, at 21:10, k.madnani84 wrote:

> Is the question whether the connections federation makes to upstream brokers
> need to be handled 'specially' if they access the upstream via HAProxy?
> Yes exactly i want to know how federation communicates with upstreams and
> presence of HAProxy in between federation and upstream will make any
> difference or we need to communicate as it is?
> 

Federation uses ordinary AMQP connections to communicate with upstream nodes. For each upstream node, there will be a connection. Load balancing should not be a problem in theory, because RabbitMQ clusters share information about queues and federation uses a queue internally (on the upstream node) known as the upstream queue, which the downstream connections are consuming from. As per the reliability guide (http://www.rabbitmq.com/reliability.html):

"Federation will automatically distribute links across the downstream cluster and fail them over on failure of a downstream node. In order to connect to a new upstream when an upstream node fails you can specify multiple redundant URIs for an upstream, or connect via a TCP load balancer."

HTH

Cheers,
Tim
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20130509/b3635c1a/attachment.htm>