[rabbitmq-discuss] redelivered count

Thu Dec 9 01:48:43 GMT 2010

Honestly I wasn't under the impression that the message ordering could be
relied on at all. I mean a basic.reject with a redeliver puts the message to
the end of the queue, right?

On Wed, Dec 8, 2010 at 6:14 PM, John DeTreville <jdetreville at vmware.com>wrote:

> Well, I suppose, but I doubt that all clients would be happy with
> arbitrarily reordered messages. If the point is just that messages can
> already get reordered, and so reordering a few more in various corner cases
> can't hurt, well, maybe, but I'd prefer to be able to set some limits.
>
> That's just my own preference, of course!
>
> Cheers,
> John
>
> On Dec 8, 2010, at 10:58 AM, Jason J. W. Williams wrote:
>
> > I would imagine so. But there are no real ordering guarantees anyway,
> > so I don't think that's a problem.
> >
> > -J
> >
> > On Wed, Dec 8, 2010 at 11:56 AM, John DeTreville <jdetreville at vmware.com>
> wrote:
> >> Well, I understand that part.
> >>
> >> But if suspected bad messages are to be delayed for a while, am I right
> to imagine that other messages would not be delayed? That is, messages would
> be reordered?
> >>
> >> Cheers,
> >> John
> >>
> >> On Dec 7, 2010, at 6:05 PM, Jason J. W. Williams wrote:
> >>
> >>> Hey John,
> >>>
> >>> I think the intent is to keep a message that's crashing a consumer
> >>> from being endlessly retried.
> >>>
> >>> -J
> >>>
> >>> On Tue, Dec 7, 2010 at 6:15 PM, John DeTreville <
> jdetreville at vmware.com> wrote:
> >>>> I'm a little confused. Would messages be reordered? Is the idea to
> move from message queues to something else?
> >>>>
> >>>> Cheers,
> >>>> John
> >>>>
> >>>> On Dec 7, 2010, at 2:39 PM, Justin Etheredge wrote:
> >>>>
> >>>>> It would interesting and useful if there was some way to see a
> redelivered count on a message. And as a little icing on the cake, it would
> be useful to be able to specify how long before a message is redelivered.
> That way if I specify that a message can be redelivered at most 10 times, it
> won't happen in a thousandth of a second. Any thoughts on this?
> >>>> _______________________________________________
> >>>> rabbitmq-discuss mailing list
> >>>> rabbitmq-discuss at lists.rabbitmq.com
> >>>> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss
> >>>>
> >>
> >>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.rabbitmq.com/pipermail/rabbitmq-discuss/attachments/20101208/bc5b8cb4/attachment.htm>