The dashboard can browse your databases without a public TCP proxy, so you can keep them private. I confirmed the connection attempts were timing out, though I was able to open your dev Postgres database and load its tables successfully in a fresh session.

Could you try a hard refresh or a private browsing window? If it still spins, please share the browser console errors and any failed WebSocket requests from the Network tab so we can narrow down where the connection is failing.

it still fails even in incognito mode:

Console errors are

"Failed to load resource: the server responded iwth a status of 403 ()"

"Uncaught InvalidStateError: Failed to execute 'send' on 'WebSocket': Still in CONNECTING state."

"WebSocket connection to 'wss://ssh.railway.com:2226/ws/tcpip' failed:"

here is a snapshot of what shows up under the invalid state error

Thanks, that narrows it down. The Database tab opens its connection through a WebSocket to ssh.railway.com on port 2226, and your console shows that socket never gets past connecting. The other errors follow from that. When we opened the same database from our side, the connection completed.

Since the failure is at the WebSocket itself, the next thing to rule out is the network between your browser and that endpoint. Could you try the Database tab from a different network, such as a phone hotspot, with any VPN, proxy, or corporate firewall turned off? Port 2226 is not a standard web port, and some networks block outbound traffic on it. If it works there, the network you're normally on is blocking that port.

no go on two other wireless networks including my phone hotspot.

Could it be an internal firewall issue on my laptop i need to review?

All signs do point to something in that regard, yes, since our monitoring for that domain, port, and path combination is all green on our end, and our signals show no other users reporting this error at this time.

Ok, thanks for the help, it is not internal on my laptop, checked firewall and profiles and rules, and they are clear, so it must be our ZScalar, i'll take it to IT. Thank you for helping me confirm where the issue is