The way you do this is by having an exact duplicate of the remote system so you can set up a test with conditions as close to those under which the remote system is currently operating. You can then do a series of carefully controlled test solutions to determine the optimum prior to trying it on the "live" system.
Which is, of course, exactly what they did. My son Charlie (class of 2020, MIT or CalTech, he hasn't decided yet :-) and I have been following the Mars Rovers since sol 1, and as soon as MER-A stopped sending, I googled up all available articles. One of them mentioned that there was an exact duplicate of the rover in Pasadena, and that JPL engineers were duplicating the sequence of events to figure out the best solution.
Charlie hit upon a pretty fair solution himself: send up an astronaut, have them delete the files and reinstall new software. I reminded him that the send-an-astronaut part would mean at least a 14-month round trip, and pointed out (to help him get a sense of the time involved) that 14 months ago he had just had his sixth birthday.
I think he was a little disappointed, as I'm sure he was ready to volunteer... :-)
