Since I changed the ramp time to 2 seconds for the second time, there hasn't been a change in the rate of these locklosses, there have been 8 per nln lockloss since March 18th at 21 UTC.  I've now changed the ramp time back just to avoid an unnecessary change to ALS.

Oli used data from the lockloss tool to make this useful plot showing how many locklosses we've had from the LOCKING_ALS state (state 15 for ISC_LOCK), blue shows the total number of state 15 locklosses for that day and orange shows those that were tagged as high wind or EQ.  This shows the problem starting on the 22nd or 23rd UTC time, which might line up in time with this change to the ALS demod phase and locking gain: https://alog.ligo-wa.caltech.edu/aLOG/index.php?callRep=82388

Because of the large EQ right now, we can't lock the Y arm to measure the TF.  I've reverted the changes, and accepted the changes in SDF safe.snap.  They will need to be accepted in observe.  It would also be a good idea to measure the OLG when we can.