Simplicity was the key!

That's it.

Here's what I was thinking before you mentioned checking the affected rows: In the "undesirable scenario" I wouldn't have checked the number of rows affected by the update and both scripts would have gone on with the 60 seconds of processing.

You're very correct, though. If I simply make sure that the update affected one row, I'm fine. If it didn't affect one row, then I will just keep doing queries until I get a row that I can successfully update.

Perfect! I haven't coded it yet, but I understand the concept.

Thanks so much szlamany. RobDog and Penagate told me you were the SQL master. That seems to be the case.