I know, I've been keeping a track of it. I thought I had fixed the issue yesterday.
It's very very hard to replicate locally. I think it's cluster formation related.
I've added another barrier a few minutes ago to make sure the cluster is formed before
the tests are executed.
I'm pretty confident this intermitent failures do not lead to misbehaivours at
runtime.
If you want, you could accept failures in the Hibernate Infinispan module and still build.
I'm pretty much always on top of that module so it shouldn't have an impact on the
overall Hibernate build.
We've had intermitent failures in the Infinispan testsuite that we're now managing
and controlling better (i.e. use memory based cluster discovery mechanisms for example
that are much more reliable than time sensitive standard ping ones) and I've started
implementing some of those in Hibernate 2LC module.
On Jan 11, 2012, at 4:41 PM, Steve Ebersole wrote:
We have been having intermittent failures with the
hibernate-infinispan
module. Yesterday's upgrade to the latest Infinispan version seems to
exacerbate this condition; at least the CI builds are failing routinely now.
I think we either need to spend some time cleaning up whatever leads to
the intermittent-ness or we do "something else" to isolate the build
from it (move this to a separate, dedicated repo; etc).
--
steve(a)hibernate.org
http://hibernate.org
_______________________________________________
hibernate-dev mailing list
hibernate-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/hibernate-dev
--
Galder Zamarreño
Sr. Software Engineer
Infinispan, JBoss Cache