One of the interesting things about testing is how you interpret the results. e.g., you have to run three test cases with passing results to pass the overall t

One of the interesting things about testing is how you interpret the results. e.g., you have to run three test cases with passing results to pass the overall test and certify the system. So, you run the test. All three test cases pass with flying colors, but during test #3, something that you hadn't thought of came up and it could be a problem. What do you do now? You've reached your stated qualification for passing the test but now there's this wrinkle. Which one should take precedence in certifying the system for use?