[rabbitmq-discuss] txAMQP with RabbitMQ Cluster

[Alex Clemesha]

On Wed, Mar 18, 2009 at 3:39 PM, Terry Jones <terry at jon.es> wrote:
>>>>>> "Alex" == Alex Clemesha <clemesha at gmail.com> writes:
>
> Hi Alex
>
> With respect to your second issue:
>
> Alex> Secondly, and this is where I'm a bit more lost (i.e, the problem may
> Alex> or may not lie in the client), can txAMQP handle redirects from
> Alex> brokers?  Here is my situation: I have a Rabbit cluster going on ec2
> Alex> (see this thread [2]), and right now I'm only connecting to the
> Alex> Cluster from another ec2 instance in the same Availability Zone
> Alex> (LAN).
>
> Alex> The interesting behavior is this: producing and consuming through the
> Alex> Cluster node that I performed the "rabbitmqctl {stop_app, reset,
> Alex> cluster nodeA nodeB, start_app}" on *works*.  Trying to connect to
> Alex> the *other* cluster node results (in /var/log/rabbitmq/rabbit.log)
> Alex> *redirect* messages like so:
>
> Alex> =INFO REPORT==== 18-Mar-2009::21:15:49 ===
> Alex> connection <0.804.0> redirecting to
> Alex> <<"domU-12-31-39-00-A1-E3.compute-1.internal:5672">>
>
> Alex> and at this point my txAMQP producer client just "hangs", i.e. it
> Alex> does not successfully send its message.
>
> Is your client using the internal IP address of the broker server? I.e.,
> the one you fetch as the private DNS entry from
> http://169.254.169.254/user-data when the instance starts up.
Yes, it is. Right now I'm restricting my tests to the availability
zone "us-east-1a",
so they all have the form: domU-***.internal.


thanks,
Alex

>
> Terry
>



-- 
Alex Clemesha
clemesha.org