confusion with quick back-to-back lock losses

We had a recent event where guardian went through the LOCKLOSS state twice within a very short time (presumably within about a second). This caused the two events to be confused, with resulting mixup in the refined time:

Oddly, the indicator went down after the first lockloss event, which is weird because I thought that IMC_TRANS was the indicator. I'm scratching my head a bit on this one...

We've a little bit committed to resolving lockloss ids to the second of the guardian time, which I guess is too coarse in this case. We should probably just figure out how to ignore the second lock loss and just go with the first.

Assuming I'm interpreting this weird event correctly...