[rabbitmq-discuss] After server notifies of canceled consume, it doesn't acknowledge a channel close
rev.chip at gmail.com
Fri May 3 00:23:32 BST 2013
Look closer and the inconsistency is clear. In both cases the only problem
is that the client wants to consume from a queue, but the queue's
nonexistence makes that impossible. Yet in one case the channel gets
closed; in the other, it doesn't. Why should nonexistent queue at start be
a channel exception, but nonexistent queue later not be one? Makes no
sense to me. IMO neither case should be a channel exception. Too late to
change it of course.
BTW the client code isn't general-purpose, so you need not worry that any
(other) users will be inconvenienced by my idiosyncrasy.
On Wed, May 1, 2013 at 10:16 PM, Michael Klishin <
michael.s.klishin at gmail.com> wrote:
> 2013/5/2 Chip Salzenberg <rev.chip at gmail.com>
>> The decision to close the channel is for symmetry with the "consume"
>> method, which closes the channel if it fails.
> When basic.consume fails, it is because you try to consume from a queue
> that does not exist or the queue is exclusive
> and owned by another connection. This raises a channel exception and that
> causes the channel to be closed.
> This is not specific to basic.consume but all channel exceptions. I
> personally don't think making cancellation notifications
> close a channel improves consistency. I'm not aware of other clients doing
> rabbitmq-discuss mailing list
> rabbitmq-discuss at lists.rabbitmq.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the rabbitmq-discuss