Just after I asked the interferometer to go to DC readout at 0:31:37 UTC Oct 30th the watchdogs at EX tripped, including HEPI, ISI ST1+ST2, and TMSX (but not the quad).  

The ISI ETMX ST2 guardian has an ezca connection error, which did not allow the ISI to reisolate.  I stopped the node and went to execute again, and it is now isolating HEPI.  

The attached screenshot from the lockloss tool shows that tidal did not trigger off unitl 0.8 seconds after the lockloss.  The trigger for tidal is based on the sum of the red transmon QPDs, and has been set to 3000 by the guardian in the step DRMI_TO_POP

I added a few lines at the end of the increase power state so that after we increase the input power we reset the tidal off threshold to 8000 counts, which would have helped avoid the HPI trip in this instance.