My best guest for my case of 502 would be something/someone caused a service crash. It’s not that bad, it restarts and has duplicates, but so some requests will still be lost.
This should trigger some alerting and they’ll fix it server-side eventually.
Why now ? could be because of the update that brought an unexpected and somehow rare case, or overloaded something with too much traffic. or could be something else entirely.
Just retry, if it keeps happening for a few minutes status page will show it, otherwise it’s a case of random bad luck.