Hi Nisbus,
Thanks for sharing this issue.
Am I correct in my understanding that this test is executing code across multiple threads?
I'm afraid that I have no knowledge of the how this framework behaves, although I suspect that this is a problem with the way the test is designed, rather than NCrunch itself. Because NCrunch executes the test much more frequently (and in a very different environment) to other test runners, there is a much high probability that race conditions within the test will be noticeable with NCrunch but not other test runners.
Intermittent failure of tests that execute code across multiple threads is highly likely to be the result of race conditions within either the test, or the code under test.
If this message bus listens for messages on a background thread, then there is a suspect race condition between the dispatch of the message and where it is received, as this test is checking for the existence of a received message on a different thread to that response for receiving the message.
If you aren't executing this code across multiple threads, check that the test is not somehow sequence dependent (i.e. can be confused by leftover state from previous execution runs of itself or other tests).
Relevant documentation links that you might want to look through:
Multi-threaded tests,
Test atomicity.
Cheers,
Remco