Agreed.  Though we need the patch for the standard / multiplexing
providers.  I'd rather finish the new endpoints (InOut support has not been
written yet), but patches are always welcome.


On 3/20/07, samg <[EMAIL PROTECTED]> wrote:


Hi,
Am I right in thinking that a servicemix-jms provider endpoint can't be
used
in a situation requiring recovery from failure (with a system running on a
single node)?  From looking at the behavior in this kind of test, and the
source of StandardProviderProcessor, it appears that the temporary queue
that is created precludes the possibility of continuing to process the
messages held in the non-temporary queue when the server goes down.  It
seems relatively simple to add a "reply-to" property to the Provider
endpoint, allowing the use of another non-temporary queue- does this
approach make sense?  I'd be happy to do it and contribute a patch.
Sam
--
View this message in context:
http://www.nabble.com/replies-to-JMS-and-recoverability-tf3431435s12049.html#a9565929
Sent from the ServiceMix - User mailing list archive at Nabble.com.




--
Cheers,
Guillaume Nodet
------------------------
Architect, LogicBlaze (http://www.logicblaze.com/)
Blog: http://gnodet.blogspot.com/

Reply via email to