Just for example, doesn't the satellite box have a current monitor ports? The best solution would be to attach a LED readback but it is impossible now...

Abstract:
To confirm the place having problems is inside vacuum or outside, I checked signals in several situation and concluded the problem is inside the vacuum.
Also, I checked whether these glitches are scattered light or ghost beams of OpLevs, I turned off all MN and PF OpLev but the glitches are sitll there.

Details:
I swapped the cables for MNV1(glitchy signals) and MNV3() at several points outside the chamber and checked the signals.
Position of the replacement and the glitch channels are summarized as follows:

Wherever we swapped cables outside the chamber, glitchy channel was swapped: that means the signal from chamber have glitches.

After confirming the glitches are comeing from chambers, I checked MNV1 signal without MN/PF OpLev injection beam to confirm OpLev light makes glitches or not. 
The signal has still glitches when the OpLev light was turned OFF: that means at least one of the source of glitches is not OpLev light (I don't say OpLev light is not guilty at this moment). 
We will do further investigation tomorrow (onece clamping the suspension and swapping the cables at BF and PF, checking connection of in-vac D-sub and legs of LED and PDs, and so on).

Note:
During the measurement, the suspension is SAFE state and there was no output from DAC.
So, electrical coupling between control signals and sensor signals are not the source of this glitches.