Hey there! We've found the following might help you get unblocked faster:

• 🧵 502 Connection Dial Timeout

• 🧵 Traffic spike errors http 502

• 🧵 TCP Proxy for PostgreSQL not accepting connections - Connection timeout

If you find the answer from one of these, please let us know by solving the thread!

what's definitely happening: ipv4 connections timing out (10.240.116.48:5432 - 2+ seconds) , and ipv6 connections working instantly (fd12:... - 0 seconds) , postgres itself is running fine

so if your environment was created before oct 16 2025 , it's a "legacy" ipv6-only environment, your go app tries ipv4 first which times out. proven fix from railway docs:

go

config, err := pgxpool.ParseConfig(os.Getenv("DATABASE_URL"))
config.ConnConfig.DialFunc = func(ctx context.Context, network, addr string) (net.Conn, error) {
    return (&net.Dialer{}).DialContext(ctx, "tcp6", addr)
}
pool, err := pgxpool.NewWithConfig(context.Background(), config)

a guaranteed workaround is to use DATABASE_PUBLIC_URL temporarily to confirm postgres works, then you know it's the private network issue

check when your railway environment was created. if before oct 16 2025, option 1 will fix it. the tcp6 forces ipv6-only connections which your logs show work perfectly

i hope this help you, let me know the result

ilyass

Thank you ilyass for your quick replies on this. It is great to see a community that works!

I tried putting your code fix to force ipv6 connection (I had tried this before but I was up for trying anything) and it worked! I checked the logs and could see all connections, both ipv6 and ipv4, were working again. Odd, I thought, so I removed the new code forcing the ipv6 connection and it still works.

I may be wrong, but this seems to have been an environment issue. I'm not sure if anyone else was affected but nothing changed in my project other than the passage of time, and it all now works again.

Odd.

Just a note that the environment was created in the last few days so the ipv6 issue shouldn't have affected it. Shouldn't have...

glad it's working now , since your environment was created in the last few days , it should have dual-stack ipv4/ipv6 support by default according to railway's docs. so the tcp6 fix shouldn't have been necessary for your setup.

whatever caused the temporary connection issues seems to have resolved. keep monitoring it to make sure it stays stable.

happy it works for you