Same here. I think there might be a Railway networking issue ongoing? I see nothing on the status page yet. But a few posts about this now. DB network connections are not working.

For color, at 15:30:08 UTC, a single dns.lookup for postgres.railway.internal (→ 10.222.217.136) took 37.1 seconds to resolve. Seeing that spike up and down currently

Mine actually appears to have resolved for the moment (with no action on my part). After a good bit of downtime..

We are aware of an issue affecting Outbound Static IPs in US East. Some users relying on static egress IPs may experience intermittent connectivity issues.

We are actively investigating and working to mitigate the impact. We will provide updates as we learn more.

You can follow along here for updates: https://status.railway.com/incident/66L640AT

This seems to have popped up again at 15:30–15:34 EDT (19:30–19:34 UTC)

After digging further that was a spike (15:30) but seeing DNS issues continually popup today...

Are you on the public proxy or private networking? I'm seeing intermittent 'Connection terminated unexpectedly' issues as well on the public proxy

Private networking

Thanks for the detailed timestamps, they let us confirm this on our side. We verified elevated internal DNS resolution latency on a set of US East hosts during both windows you reported: July 18 around 15:30 UTC (the larger event, which correlated with heavy load on several hosts) and a smaller recurrence on July 19 around 19:30 UTC. This was on Railway's side, not your application, and it was separate from the Outbound Static IPs incident we linked earlier.

DNS latency across the fleet is currently back to normal. Let us know if you need more help.