In order to create the test case that you would automate, you first must create the repro scenario. IOW, automation has nothing to do with this until the bug is found in the first place. Arguably, one could create a test model that might have found this but raise your hand if you even know what I'm talking about when I say "test model".
The only mitigation that automation would bring is if the bug was found in earlier versions, and test case was subsequently written. IOW, and very much a generalization, automation is to find regressions. But if the bug is new...
(To be clear, this bug still should have been found. But automation is unlikely to have found it.)
Respectfully disagree. "User cannot log in as root if root user is disabled" is absolutely a test case that should be written regardless of previously seeing the bug.
Meh, you're probably right. If nothing else, I'd want to verify the result of trying to use a disabled account (text in the dialog is localized, et. al.) Run through the scenario before I formally write the case and...WTF? Yeah, I could see that.
