[rabbitmq-discuss] Pathological dead queue behavior in a cluster.

[Matthew Sackman]

On Wed, Mar 30, 2011 at 04:14:16PM -0600, Jason J. W. Williams wrote:
> That would make sense. I was under the (erroneous) impression making the
> exchange durable only mattered if the exchange was lost (e.g. cluster went
> down or a standalone node failed). Thank you for working through it with me.
> I'll see what we can do about getting cancellation notifications into Pika
> and the book.

No probs at all. I agree it would be good for pika to catch up with some
of the features we've been adding to AMQP, but I realise there's likely
a lot of code in flux atm. Please let us know if any of the
documentation regarding our extensions is unclear. If in doubt though,
using the Java Tracer and/or wireshark is probably the most immediate
way for you to see what our clients are sending and receiving from the
broker.

Best wishes,

Matthew