Interesting theory by DC Rainmaker

Interesting, wonder if it could be related to decommissioming sat. SVN45 (PRN21) yesterday and an invalid CPE having been generated:

www.navcen.uscg.gov/.../2025010.nnu

This is actually the second time the EPO file somehow is wrong/corrupt. The Solution is probably going to be that the dev team, will try to prevent this in the future by validating the EPO-datafile and sign it with a HASH so if data is transferred wrongly to a device the EPO-file will be ignored and secondly a delay of the release of EPO-files (where they are being validated) before they are "shipped" to GPS-devices.

   

Can you share the details on how you fixed this?

The astonishing fact about this whole problem is, that there is no error handling implemented for a file which is not even crucial for the functionality of the systems.

Hopefully it is now in the top 5 of the todo list 

But no matter if it is called Garmin, Apple or Microsoft (you can extend this list) - every company consist of normal people like you and I - and mistakes are happening.

This is now fixed for me after resyncing with Garmin Connect.

I'm not sure if that's true. I have watch version 20.32 (I stopped updating some time ago due to more and more bugs in subsequent versions). My EPO is up to date. Today I did 2 activities with GPS and my watch is not affected by the restart loop.

I also was not affected and was able to do a run in the "blackout" time. Maybe it also depends where you have been (in my case Germany). So I did not notice anything before I read all those posts here.

Edit: So the idea is that not the whole file itself was broken but some specific satellite information. And if this satellite was not in your visible hemisphere you did not notice anything. Because I assume that the at least check that they do not provide completly broken files for all users. But it is hard so check if some specific information is corrupt or not handeled.

The reason for that was that your EPO file was not due for an update and then you did not get the "triangle of death"...

So this means this A-GNSS file do not expire at the same time but in a rolling interval. Meaning that only a part of the users was affected by syncing a broken file.