On Tue, Apr 26, 2011 at 5:24 PM, Heath Kesler <[email protected]> wrote: > Thanks Claus, > > I will take a look at this, I have done a few things in JMS lately and look > forward to the improved performance.... I also noticed a couple of failures I > will try and fix. >
The optimization in AMQ is among others about acknowledgement so AMQ can do that in batches/bulk. However I think I have seen one test which failed due message2 arrived before message1. Running the test many times thereafter didnt show this error. Of course I would expect AMQ to consume messages in FIFO order when 1 consumer only. So anyway, just wanted to share. If we start see "odd" failures. Then we may have to report to AMQ and/or disable some of the optimizations. PS: The big boost would be to not fork the JVM per test. And also to reduce the number of Spring XML based test, as Spring Framework is becoming dead slow at booting up its XML files. (Spring 2.0 or older was much faster at this). > Cheers > Heath > > On Apr 26, 2011, at 1:35 AM, Claus Ibsen wrote: > >> Hi >> >> I recently commited a change to camel-jms with some optimize >> instructions to AMQ. >> This cut down unit test time a bit. I see roughly about 1min or better >> on my system. >> >> But in case some unit tests start to mis behave on your system. Then >> please report back so we can take a look. >> >> >> -- >> Claus Ibsen >> ----------------- >> FuseSource >> Email: [email protected] >> Web: http://fusesource.com >> CamelOne 2011: http://fusesource.com/camelone2011/ >> Twitter: davsclaus >> Blog: http://davsclaus.blogspot.com/ >> Author of Camel in Action: http://www.manning.com/ibsen/ > > -- Claus Ibsen ----------------- FuseSource Email: [email protected] Web: http://fusesource.com CamelOne 2011: http://fusesource.com/camelone2011/ Twitter: davsclaus Blog: http://davsclaus.blogspot.com/ Author of Camel in Action: http://www.manning.com/ibsen/
