[rabbitmq-discuss] Durable queues for STOMP broker

[Tony Garnock-Jones]

Hi again Sean,

Tony Garnock-Jones wrote:
> Thanks very much for these. They look good. I'll apply them.

The changes to the core broker will appear as part of the next broker 
release. You'll probably have to maintain your patch manually until then.

The changes to the STOMP adapter have been applied and are currently 
live on http://hg.opensource.lshift.net/rabbitmq-stomp/.

> "passive" essentially checks that the queue exists without creating it. 
> If it doesn't exist, an error is signalled. I'll check to make sure the 
> behaviour of the STOMP adapter in this case is sensible.

I ran a couple of tests to make sure that this was working properly, and 
sure enough there were some problems, which I've now fixed as part of 
http://hg.opensource.lshift.net/rabbitmq-stomp/rev/19b67fadda0d.

By the way, is it just me, or does the Ruby STOMP adapter get very 
confused when the socket closes underneath it? I see fairly consistent 
behaviour of 100% CPU usage, with the Ruby program just spinning, if the 
socket's closed by the server for whatever reason. Odd!

Please let me know how the new tip works for you, if you try it out, Sean!

Thanks,
   Tony