Unable to access the URL after deploying Discourse

Hey,

Looking at your output, nginx is failing to start. This is a common issue when Discourse is already using ports 80 and 443, which nginx also needs. Here’s how to diagnose and fix this:

Step 1: Check what’s using the ports

sudo netstat -tlnp | grep ':80\|:443'

or

sudo ss -tlnp | grep ':80\|:443'

Step 2: Check nginx error logs

sudo journalctl -u nginx.service -l --no-pager

or

sudo tail -f /var/log/nginx/error.log

Step 3: Check if Discourse is running

sudo docker ps

Most likely solutions:

If Discourse is already running and using ports 80/443: You don’t actually need nginx! Discourse’s built-in reverse proxy handles this. Try accessing your site directly with your domain/IP.

If you still want to use nginx as a reverse proxy:

1. Stop Discourse temporarily:

   cd /var/discourse
   sudo ./launcher stop app

2. Start nginx:

   sudo systemctl start nginx
   sudo systemctl status nginx

3. Configure nginx to proxy to Discourse (running on a different port), then restart Discourse.

Quick test - try accessing your site: Before making changes, try visiting your domain in a browser. If Discourse installed correctly, it might already be working without nginx.

What do you see when you run the diagnostic commands above? This will help determine the exact issue.

If you’re unable to access the Discourse URL after deployment, check the following:

1. DNS – Make sure your domain points to the server IP.

2. Firewall – Open ports 80 and 443.

3. Docker – Ensure the Discourse container is running.

4. SSL – Check for Let’s Encrypt or SSL issues.

5. app.yml – Verify the correct hostname is set.

6. Resources – Ensure the server has enough RAM and disk.

7. Logs – Run ./launcher logs app for error details.

Rebuild with:

cd /var/discourse
sudo ./launcher rebuild app

Heya, @dashan

There’s a misconfiguration or a port conflict.

You can run:

sudo journalctl -xeu nginx

That will show the error message that caused nginx to fail. Often it’s something like “bind() to 0.0.0.0:80 failed” if another process is already using port 80.

Run:

sudo nginx -t

If there’s a typo or bad directive in /etc/nginx/nginx.conf or a site config in /etc/nginx/sites-enabled/, it will tell you.

Port 80 (HTTP) or 443 (HTTPS) might already be in use by Apache, another nginx instance, or something like Discourse’s built-in web server. Check with:

sudo lsof -i :80
sudo lsof -i :443

If Apache is installed, you’ll need to stop and disable it before nginx can bind:

sudo systemctl stop apache2
sudo systemctl disable apache2

Once you resolve the conflict or config error:

sudo systemctl restart nginx
sudo systemctl status nginx

Hope that this helps!