I've been fooling with microcircuits for quite a long time, yet, in all that time, since the 8008 was a pup, I've only had two processors fail. I've had bad memory from time to time (and my own memory isn't doing as well as it once did, either!) and a few MC1489's and '244's, '245's, 373's, 374's have "croaked" along with a handful of programmable devices, mostly fairly large FPGA's. MCU's aren't that easy to "zap" unless, perhaps, you put them in the wrong programming configuration.

Moreover, I've yet to see even one problem with RC reset when the Vcc supply comes up briskly. I have seen issues when Vcc takes too long decaying, though, including that one case I mentioned, wherein I actually observed activity from the MCU while RESET was true and Vcc above 4.5 volts, which was the trip-voltage for my NVRAM's.

I'd look for something else, were I in your place. I wouldn't discourage you from using a supervisor, but, just for grins ... how about taking it out and going back to the RC, just for a trial, once you've fixed the real problem and see if it actually fixed something. Who knows? It might actually show us something.

RE