I'm not sure I know what you mean by "I was running two outs ..."--were you splitting the signals at the outputs of your preamp? Or does the preamp have two parallel sets of outputs? Mainly, does this mean that you were making two recordings, and that they both cut in and out at the same times? Sorry for the confusion but I want to get the whole picture in my mind if I can.
Do you have editing software that would let you look at the waveforms at the points where the signals cut out? If so, can you see whether there's a smooth decay (even if it's less than a second), which would be consistent with a loss of powering to the microphones, or whether there's a click or crackle or other transient noise, which would be more consistent with an interruption in the cabling while the microphones are still putting out their signals?
You might learn something similar by looking at the point at which the recording resumes.
Also, if the recorded waveform is dead silent/constant during the outage, then you have a defective recorder, since you were using analog inputs and there is always some noise there (the M10 has enough residual noise in its analog inputs that its 24-bit mode seems pretty worthless based on what I've measured with mine, but that's a whole other discussion ...).
I know you say it wasn't the cabling, but loooong experience says that the most likely cause is either the cabling or a connector somewhere. Murphy's Law dictates that the cheaper parts of the setup have a nasty tendency to undermine the more expensive parts.
--best regards