Yeah, looks like the issue started at roughly midnight Eastern / 4am UTC. It’s the same issue as June 2, but I think I actually figured it out this time.
The core issue is that too many Challenge processing jobs were queued, saturating my job database memory, bringing that server down. I thought I found where I was doing that after the June 2 issue, but there were more places that I was reprocessing Challenge progress (e.g. activity deletions).
@gijs.smets try a single-activity sync, from your settings page, if it hasn’t arrived yet. This may fail - I’m actively working with Garmin support on the exact failure, so if it does fail know that I’m already working on it.
