Let’s all remember to check out the simplest, easiest to check causes first anytime we’re debugging a testing or production defect. If nothing else, knock the simple explanations off the list of possibles quickly before you consider anything more complex. We had a case today where a data validation message in the test environment was wrong, but the FitNesse test in our build environment was correct. Simplest explanation? The deployment to testing wasn’t correct. Actual explanation? The deployment to testing was incorrect. That should have been the very first thing to check, but it wasn’t.
We had a manual post-step in a build process that was missed. That’ll be automated by the end of day tomorrow to stop that problem in its tracks. Grrr.