If you do have an "outgoing inspection" (as opposed to testing) procedure, that's where you can verify that the voltages are correct. If all your components are supposed to work within a 5% supply voltage range, then the fact that it fails at 4.89 is a clear indication that something isn't working properly.

This is undoubtedly a hardware problem and no amount of firmware effort is likely to fix it. If there's any part of the board that doesn't receive the correct voltage at Vcc, then it's time to find out why.

If you find a part of the circuit that IS receiving Vcc within the specified limits, yet doesn't function properly, but which DOES work at 5.00 Vcc, then it's time to call the lawyers, because the circuit malfunction then seems to be caused by improper specification of the supply voltage tolerance, and the cost for such should be borne by the manufacturer of those parts. Nost likely, however, that's not the case.

If this were my predicament, I'd be looking very closely at the supply voltage, possibly with the idea in mind that, somehow, the supply voltage is being driven out of the 5% tolerance. It's possible that a change in the supply routing or in the supply bypass can fix the problem.

Have you tried measuring the Vcc path to each Vcc connection with a milliohm-meter? There might be an "overetching" issue with the PCB. Maybe a close look at the supply distribution net on the artwork will shed some light.

RE