Last night, due to some guardian fast EZCA connection errors (alog 48744 and comment), Georgia and Corey had to manually turn on the BS pitch offloading. Since Ed was unable to load the new version of the guardian that has the fast EZCA parts removed (alog 48747), he ran the old code and seems to have gotten a very similar error. It looks like he accepted the SDF diffs (alog 48753), including the fact that BS offloading of pitch to the top mass is off. So, at least for the entirety of this lock so far, we've been running without any pit offloading for the BS top mass. Things seem okay for now (current lock is 7.75 hours and going), so I'm not going to turn it on just yet.
I believe that I have found and fixed the errors in the non-fez version of ISC_GEN_STATES.py, so I have put Georgia's modifications back in (with the fixes). We just lost lock (I don't know why, but I hadn't reloaded or changed anything in the running code), so I've reloaded the guardian and hopefully next lock we won't be plagued by this problem again.
We got through the acquisition sequence okay, and the BS pit top mass offloading is on as it should be, so I think we can call this one case closed.