Client-side DNS fixes probably won't help here since this reproduces across multiple different generated domains and projects, it's not something local to your device/network, otherwise you'd expect at least some domains to resolve fine.

NXDOMAIN specifically means the DNS query is resolving successfully but finding no record for that hostname at all different from a timeout or connection refusal, which would point to your app. Since the healthcheck and internal routing are all fine (200 on /health), this looks like the newly generated *.up.railway.app subdomain just hasn't propagated yet on Railway's DNS side.

A couple things worth checking:

How long ago was the domain generated? Sometimes there's a short propagation delay (a few minutes) right after a domain is first created, especially under load.

Try dig yourdomain.up.railway.app or nslookup yourdomain.up.railway.app from a couple different networks/devices if it's NXDOMAIN everywhere, not just your machine, that basically confirms it's not a local DNS/cache issue and needs Railway to check on their end.

Double check the exact domain string wasn't cut off or has a typo when you copied it from the dashboard — small thing, but worth ruling out first.

If dig/nslookup comes back NXDOMAIN from multiple independent networks, this is on Railway's DNS/domain-generation side and needs their team to look at it directly rather than anything fixable from your end.