What value is timeout?
My experience of Hudson is that timing tests are prone to sporadic failure, due. I'm
not sure if that is due to large gc's kicking in, or because it's virtualized and
over-stretched so the non real time nature of Java thread scheduling is massively
exacerbated, but we've found we've had to increase timeouts to high values
(several seconds) to "guarantee" receiving messages.
BTW I've only seen this problem on Hudson.
For JBM 2.0 we've extracted all our timing dependent tests into a single smaller test
suite, which we're going to run on a non virtualized server. I believe Hudson already
has one server group which is non virtualized (?)
View the original post :
http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4176034#...
Reply to the post :
http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&a...