Briefing
Checkout started failing just after the overnight maintenance window, and every request to /api/ on the edge now comes back as nginx's 502 page.
The edge itself is up, and its proxy configuration has not changed in weeks.
The first message in the channel blames the nginx package update that went out in the same window.
The orders team is asleep, so whoever takes this page owns both tiers until morning.
The fix is yours to find: hints, the recovery checks and the debrief unlock inside the incident.