I am testing failure notifications to make sure they work. When setting up a task to fail, so that I can see the failure notification, I find that sometimes the task in VC completes successfully, even though it did not.
For example, I have a task to export the settings for VC; and I want to be notified when it fails. So I deliberately gave it an incorrect path name for testing, so that I could see the failure notification, and I found that the task ran successfully, although of course it did not. What would you suggest in this type of situation, to be able to correctly identify the failure?