Your service is deployed to a single US West region, so users in Europe or Asia will experience higher latency due to geographic distance - this is expected behavior for any single-region deployment. We are not aware of any global infrastructure disruptions. To reduce latency for users worldwide, you can enable multi-region replicas in your service settings, which deploys instances in multiple regions and automatically routes traffic to the nearest one.

This is simply not true. Normally servers work anywhere in the world, even if they are slightly slower when a visitor is all the way on the other side of the globe.

What is currently happening is not normal. Connectivity grinds to a halt.

Similar issues reported recently: https://station.railway.com/questions/repeated-server-connectivity-issue-e7134a64

Your service is healthy and responding normally from our side - no dropped connections or errors on the Railway network. Since you're seeing the same behavior on other providers too, this points to an internet routing issue between certain regions rather than anything Railway-specific. To help us dig deeper, could you run the steps at https://docs.railway.com/networking/troubleshooting/network-diagnostics from one of the affected locations and share the report here? Is there particular dates and times that showed problems

Today connections are all good again. Unfortunately I didn't get to investigate while the connection was bad (f.e. with traceroute while VPN'ed to other countries, or the Railway Network Diagnostics tool which I'll try next time).

This seems to be something outside of the control of Galaxy or Railway. Maybe AWS, or maybe domain systems. I'm not sure what yet. I'll debug when it happens again.

It's definitely not the app, as I can always access it just fine from the same region (USA). It's only an intermittent problem from outside the USA, and it has been on/off for weeks. I've tried two different DNS providers too, and the issues have been the same either way.