On 08/02/2008, Magnus Heino <[EMAIL PROTECTED]> wrote: > > > > Ah sorry! My bad - I'd missed that you'd not configured the > > JmsComponent with an ID of "jms". DOH! > > > > Well, http://activemq.apache.org/camel/jms.html does say "Basically you can > configure as many JMS component instances as you wish and give them all > unique names."... but I guess I was starring too much on the code examples > to notice that :-P
Me too! :) > Also all examples does use names that makes you think that the uris are > predefined. That jdbc should be jdbc. Atleast that was my impression. Yeah. > > We've tried to minimise the amount of Spring XML configuration > > required in Camel by adding some 'convention over configuration rules' > > - but I guess sometimes this magic trips folks up. > > > > Maybe you should do some more logging? "Endpoint for uri 'jms' not found, > autocreating default JMSComponent instance.."... sounds like something that > deserves a logger.info to me. Yeah. BTW what does your log look like if you use DEBUG logging and don't specify id="jms"? > YAY! We got there in the end! :) > > > > Yes. It does feel nice to have this working when I leave for the weekend. > Thanks!!! :-) Nice end to the week - lets share a virtual beer :) > > So in summary; the main issue was not using the id="jms" for the JMS > > component configuration - and it seems for XA that we cannot use the > > CACHE_CONNECTION / CACHE_CONSUMER. > > > > If this is the case then maybe atleast it should be on > http://activemq.apache.org/camel/jms.html Agreed. I've updated the wiki a bit with a warning of CACHE_NONE http://cwiki.apache.org/CAMEL/jms.html > the covers and injecting the ConnectionFactory itself maybe - which > > could be a bit too much magic maybe? > > > Not if you log the magic imho. Yeah. Lets make sure we fix up the DEBUG logging to show what the magic is doing... -- James ------- http://macstrac.blogspot.com/ Open Source Integration http://open.iona.com
