I'm seeing an occasional glitch of DistributedCacheManagerTest. This is on a branch based off 381d4911e8f43e119897eb4fde03a81c7184130e (12-Jan). So it might not be completely relevant anymore. Plus so far no Jenkins appears to be affected. The code itself seems a little odd to me, because there is a ThreadLocal playing ball here which might not have been reset properly. Hmm, maybe I should add a facility to the test runner to show all ThreadLocals at the end of a run. Has anyone else seen this happen? Carlo