[rabbitmq-discuss] Will AMQP 1.0 break current RabbitMQ api

[Cory Showers]

Rabbit team,

I am interested in using RabbitMQ as our messaging broker of choice.  But I am a little concerend with the current AMQP specification.  The ver RabbitMQ supports is 0.9.1.  And since the new 1.0 specification has come out it seems to have changed quite a bit in terms of using the exchanges and bindings.  When RabbitMQ eventually supports ver 1.0 will this break the current api?

[Simon MacMullen]

You are right, 1.0 is a very different thing from 0-9-1.

It's not yet clear when (or indeed if) RabbitMQ will support AMQP 1.0.
But even then we won't be removing 0-9-1 support for a *long* time I
strongly suspect.

For example: We still support 0-8 because some clients still use it.
This is despite the fact that 0-8 and 0-9-1 are conceptually very
similar and really everyone ought to be able to move to 0-9-1.

Cheers, Simon

--
Simon MacMullen
RabbitMQ, VMware
_______________________________________________
rabbitmq-discuss mailing list
rabbitmq...@lists.rabbitmq.com
https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss

[cupdike]

What would be the best way to monitor RabbitMQ's plans regarding a 1.0
adoption decision?

Also, why would RabbitMQ consider not supporting 1.0? Is the thinking
that it may not see industry adoption, or something else?

Just seems a little odd since 1.0 is on it's way to ISO
standardization.

Thanks,
Clark

[Simon MacMullen]

On 24/01/12 14:56, cupdike wrote:
> What would be the best way to monitor RabbitMQ's plans regarding a 1.0
> adoption decision?

Probably to watch this list.

> Also, why would RabbitMQ consider not supporting 1.0? Is the thinking
> that it may not see industry adoption, or something else?

Well, obviously we'd like to support everything and have every feature.
But we live in a world of limited resources.

The thing is, it's such a big difference from 0-9-1 that I view it as a
different protocol really, despite the name. In some ways it's more like
a super-STOMP - giving you less connection with broker internals but
being more focussed on interoperability.

So yes, I'd like to support it, but I'd also like to support MQTT and
SQS and something-cool-over-websockets (and resurrect our support for
XMPP and 0MQ and SMTP, and improve our support for REST and STOMP and...).

And it may not see much industry adoption, the way all those other
things already have. And without wanting to complain too much, it is
rather heavy and complicated, so to support it properly we'd be looking
at *not* doing quite a lot of other stuff we want to.

Of course, if AMQP 1.0 takes off substantially then that could well
change things.

Cheers, Simon

--
Simon MacMullen
RabbitMQ, VMware

[Simone Busoli]

For the little I've read about 1.0 it's really weird that it's called with the same name. I mean, migrating systems working on 0-9-1 to 1.0 sounds nearly impossible given the differences in the protocol.

That's a bit worrying to think that development on 0-9-1 might be halted in case of take off of 1.0, considering the effort you might have put on developing something to target the current version.

_______________________________________________
rabbitmq-discuss mailing list
rabbitmq-discuss@lists.rabbitmq.com
https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss

[Alexis Richardson]

We demonstrated a migration concept at the AMQP conference in New
York, using federation, and two rabbits one of which ran the 1.0
plugin (not prod ready).

About the rest of it.. Lots of things are possible.

>> rabbitmq...@lists.rabbitmq.com
>> https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss

[Simon MacMullen]

On 24/01/12 21:34, Simone Busoli wrote:
> That's a bit worrying to think that development on 0-9-1 might be halted
> in case of take off of 1.0, considering the effort you might have put on
> developing something to target the current version.

Obviously I can't predict the future, but you *really* should not worry
that we're going to leave all our 0-9-1 users high and dry.

Cheers, Simon

--
Simon MacMullen
RabbitMQ, VMware
_______________________________________________
rabbitmq-discuss mailing list

rabbitmq...@lists.rabbitmq.com
https://lists.rabbitmq.com/cgi-bin/mailman/listinfo/rabbitmq-discuss

[Matthew Sackman]

On Wed, Jan 25, 2012 at 11:29:23AM +0000, Simon MacMullen wrote:
> On 24/01/12 21:34, Simone Busoli wrote:
> >That's a bit worrying to think that development on 0-9-1 might be halted
> >in case of take off of 1.0, considering the effort you might have put on
> >developing something to target the current version.
>
> Obviously I can't predict the future, but you *really* should not
> worry that we're going to leave all our 0-9-1 users high and dry.

It's also worth pointing out that we have made a substantial number of
extensions and improvements to 0-9-1 - publisher confirms,
exchange-to-exchange bindings, nacks etc - thus it should be fairly
clear that we're more than happy to extend and improve the existing spec
to meet the needs of our users. This makes it even less likely that
0-9-1 will get abandoned in any way by Rabbit.

Matthew

[Simone Busoli]

Hi Simon, Matthew,

let me clarify that I wasn't trying to imply that work on 0-9-1 would be abandoned in case of success of 1.0, but I think it's fair to assume that in case the latter happens, assuming the same quantity of resources, effort currently being expended on 0-9-1 would be partially stolen away by the work on 1.0. In my opinion users who are not planning to "upgrade" anytime soon would legitimately perceive it as less effort being dedicated to the product on which they have invested. But I'd be glad to be proven wrong.

[Matthew Sackman]

On Wed, Jan 25, 2012 at 11:21:38PM +0100, Simone Busoli wrote:
> let me clarify that I wasn't trying to imply that work on 0-9-1 would be
> abandoned in case of success of 1.0, but I think it's fair to assume that
> in case the latter happens, assuming the same quantity of resources, effort
> currently being expended on 0-9-1 would be partially stolen away by the
> work on 1.0. In my opinion users who are not planning to "upgrade" anytime
> soon would legitimately perceive it as less effort being dedicated to the
> product on which they have invested.

Indeed. If 1.0 shows any sign of being a success then I'm sure it'll be
made clear to all how we're dividing our time and to what purpose.