Seeing some 3 -> 2 locklosses
I could have sworn I had already written a solution to this problem, but we seem to be having some locklosses incorrectly labelled as going from 3 -> 2 (LOCKLOSS_DRMI -> LOCKLOSS). I don't believe there is an equivalent problem at LLO (since they don't have a LOCKLOSS_DRMI state), so any change implemented to fix this should be LHO-specific.
https://ldas-jobs.ligo-wa.caltech.edu/~lockloss/index.cgi?event=1264528901
https://ldas-jobs.ligo-wa.caltech.edu/~lockloss/index.cgi?event=1264438976
https://ldas-jobs.ligo-wa.caltech.edu/~lockloss/index.cgi?event=1264252273