Your Caddy proxy started up without HTTPS/TLS enabled. This means it's only serving plain HTTP on port 8080. Depending on how your Railway project is configured, the public URL may be expecting HTTPS, so requests never actually reach your app — hence "this site can't be reached."

How to Fix It

Here are the steps to try, in order:

Check your Caddyfile — Open your Caddyfile in your project repo. It likely has something like http:// hardcoded, or is missing a proper domain block. It should look like:

your-domain.com {

   reverse_proxy localhost:2001

}

Not:

http://your-domain.com { ← this disables TLS

   reverse_proxy localhost:2001

}

Remove the http:// prefix from your Caddyfile if it's there — that single prefix is what tells Caddy to disable HTTPS entirely.

Redeploy after fixing the Caddyfile — once TLS is no longer being skipped, Caddy will handle HTTPS automatically and your admin page should be reachable again.

If you don't control the Caddyfile — check Railway's environment variables for anything like RAILWAY_PUBLIC_DOMAIN or custom domain settings that may need updating.

The good news is your actual app (Image 1) started perfectly fine on port 2001 — so the app itself is healthy, it's just the proxy in front of it blocking access

if this helped please mark as solved

cheers

Hey,

Thanks so much for your help!

So sorry I'm not the most computer literate.

I'm struggling to find the caddyfile. I have attached a few screenshots below. Would our repo be that 'Softaims'?

No worries at all! Looking at your screenshots, I can see something important — there's actually a bigger issue here that explains everything.

The Real Problem

In your Settings screenshot, it shows "GitHub Repo not found" in red. This means Railway has lost its connection to your GitHub repository (Softaims/hostelguest-admin). This is likely why redeploying isn't fixing things — Railway can't actually pull your latest code.

Step-by-Step Guide

Here's exactly what to do, nice and simple:

Step 1 — Reconnect your GitHub repo

In that same Settings page you're on

Click the "Disconnect" button next to Softaims/hostelguest-admin

Then reconnect it by selecting the repo again from the list

This should clear the "GitHub Repo not found" error

Step 2 — Redeploy

Once reconnected, go to the Deployments tab

Click "Deploy" to trigger a fresh deployment

Step 3 — Check if it works

Visit your admin page URL again after the deployment finishes (usually takes 2-3 minutes)

About the Caddyfile — don't worry about finding it for now. The disconnected GitHub repo is almost certainly the root cause. Fix that first and there's a good chance everything comes back to life without needing to touch any config files.

How to Find Your Caddyfile

The Caddyfile lives in your GitHub repository. Here's how to get to it:

Go to github.com and sign in

Click on your profile/organisation — look for "Softaims"

Find the repository called hostelguest-admin

Browse the files — the Caddyfile will usually be in the root folder (the main/top level of the project), or sometimes inside a folder called config/ or docker/

It will literally be a file named just Caddyfile with no extension

Ah thank you for your help honestly! The next problem for us it seems is that we don't have access to Github. I disconnected it and then tried signing in through Github with the same email we use to sign into Railway, but it doesn't recognise the email. I think its the software developer Softaims that has access to this... So I just messaged them and the owner of the business.

Your backend itself actually looks healthy from the logs:

• the Node app starts successfully
• environment variables load correctly
• the server is listening on localhost:2001

The Caddy HTTPS/TLS messages are likely warnings rather than the root problem. Railway usually handles external TLS/HTTPS at the platform layer already. (Railway)

The bigger issue in your screenshots is probably this:

“GitHub Repo not found”

If Railway lost access to the repository, redeploys can become inconsistent or fail to update the proxy/admin configuration correctly.

Since you mentioned you no longer have GitHub access and the software developer (“Softaims”) controls the repository, I think the next step is:

• regain GitHub repo access
• reconnect the repository inside Railway
• trigger a completely fresh deployment
• verify custom domain/admin endpoint mappings afterward

At the moment this looks more like a broken repository/deployment linkage issue than an actual backend crash.