[rabbitmq-discuss] Flow control behavior

[Matthias Radestock]

On 21/04/14 19:07, Jason McIntosh wrote:
> Hrmm, I guess it sorta does.  I thought though, at least, I recall
> testing about a year ago, where if a consumer was able to consume at
> only 15 messages a second, then publishing would be restricted (via flow
> control) to 15 messages a second.  While testing today, this doesn't
> appear to be the case - that publishing is only restricted by the
> capacities of the rabbit server (or cluster) locally.

It's always been like that. What has changed a few times though is how 
busy queues prioritise work relating to producers vs consumers. See 
http://www.rabbitmq.com/blog/2014/04/10/consumer-bias-in-rabbitmq-3-3/

Matthias.