What happened is that the roof closed and PHD2 lost the guide star, but since it can take up to 60 seconds for the safety monitor to get the Unsafe notification, and NINA doesn't issue the park instruction until then, PHD2 was already reporting the guide star was lost. If the park/slew took place at that point (it was a significant slew to the park position), is PHD2 supposed to stop looping, or is it too late since PHD2 has already lost the guide star? The NINA sequence does have a Stop Guiding instruction, but the NINA log shows it was skipped since PHD2 reported as looping rather than guiding. In the NINA log at 22:47:56.4266 there's a note that the guide star was lost. The unsafe flag popped up at 22:47:58.7025 and the first attempt to stop guiding was at 22:48:00.3330.
Thanks for your help. Stefan said he's going to add into NINA's Stop Guiding instruction the ability to stop PHD2 even if it's looping, so that may solve this.