After Evan and Sheila made the first steps towards maintenance recovery I stuck around for a few hours working through the Guardian steps towards resonance.

In alignment, the PRC2 loop is not working in the DRMI state.  I reverted the input matrix from 1.0*REFLA_9I --> PRC2 to the settings we used last night, 0.5*(REFLA_9I + REFLB_9I) + 0.83*(REFLA_45I + REFLB_45I), but this didn't give us a good error signal either.  For this evening I commented out the loop from the DRMI Guardian and tweaked PR2 by hand, it worked well enough to move forward.  Oddly, it's only the yaw loop that looks bad.

The AQ45--> DARM matrix element needs to be re-tuned in the late stages of the CARM offset reduction.  We are losing lock at the 10picometer state -- the optical gain is increasing and DARM becomes unstable.  The DARM loop in the 15pm state is ok (40Hz ugf, 50deg margin).