[rabbitmq-discuss] Usefulness of Channel.Flow with RabbitMQ 2.x.x

[Matthew Sackman]

On Tue, Nov 09, 2010 at 10:31:49AM -0800, A. MacLeod wrote:
> Sounds like a case of "fine in theory, but not so much in practice."
> With such a fuzzy distinction between well-behaved and badly-behaved
> clients, channel.flow starts to look pretty clunky.  From what I can
> tell with the experimental code I've written, the TCP backpressure
> works beautifully.

It does, up to the point that you have one consumer and one producer on
different channels within the same connection, and thus the consumer
gets stopped at the same point as the producer. But then again, channels
themselves are possibly a misguided feature too.

Matthew