On: How Completely Messed Up Practices Become Normal

[mod]

> if one build fails and you run it again and it passes, what do you do?

I run it a third time, to start.

[lmm → mod]

What do you do differently based on the outcome of that? The only scenario I can think of where that makes sense as a first step is if you're happy to ignore a 1-out-of-3 failure but feel the need to investigate a 2-out-of-3, which seems like an odd position to take.

[mod → lmm]

I just think it's the best diagnostic step to take first.

If it fails again, perhaps you've got a timing issue, or something that's switching back and forth.

Of course you're not happy to ignore the failure.