Using nginx as reverse proxy and 404 assets problems

[ericsala]

I’m using Nginx as reverse proxy but I have a problem with it. It’s not serving any asset (JS, CSS, image, etc). When I go to my subdomain I can only see a blank page. I can see asset 404 errors if I use inspector in my browser.

I’m using this in docker:

docker run -it --rm \
    --name n8n \
    --add-host host.docker.internal:host-gateway \
    -e GENERIC_TIMEZONE="Europe/Madrid" \
    -e TZ="Europe/Madrid" \
    -e NODE_ENV=production \
    -e N8N_HOST=subdomain.example.net \
    -e N8N_PROTOCOL=http \
    -e N8N_PORT=5678 \
    -e N8N_BASIC_AUTH_ACTIVE=true \
    -e N8N_BASIC_AUTH_USER=user \
    -e N8N_BASIC_AUTH_PASSWORD=password \
    -e DB_TYPE=mariadb \
    -e DB_MYSQLDB_DATABASE=databse \
    -e DB_MYSQLDB_HOST=host.docker.internal \
    -e DB_MYSQLDB_USER=n8n_workflows \
    -e DB_MYSQLDB_PASSWORD=password \
    -p 5678:5678 \
    -v ~/.n8n:/home/node/.n8n \
    n8nio/n8n  

And this in Nginx:

proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header Host $host;
proxy_pass http://127.0.0.1:5678;
proxy_http_version 1.1;
proxy_set_header Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";

Any similar problem? Is there any flag to use in n8n to avoid any domain or cookies restriction?

[Jon]

Hey @ericsala,

Welcome to the community

Try setting WEBHOOK_URL to match your protocol and domain so for HTTPS on the standard port it would just be WEBHOOK_URL=https://subdomain.example.net.

I would also add the 2 lines below to your nginx config as well.

proxy_buffering off;
proxy_cache off;

[ericsala]

Still not working

I’m using this nginx configuration:

proxy_pass http://127.0.0.1:5678;
proxy_set_header Host $host;
proxy_set_header X-Real-IP $remote_addr;
proxy_set_header X-Forwarded-For $proxy_add_x_forwarded_for;
proxy_set_header X-Forwarded-Proto $scheme;

proxy_set_header Connection '';
proxy_http_version 1.1;
chunked_transfer_encoding off;
proxy_buffering off;
proxy_cache off;

Now I’m using docker-compose with same results:

version: "3"

services:
  n8n:
    image: n8nio/n8n
    restart: always
    extra_hosts:
      - "host.docker.internal:host-gateway"
    ports:
      - "127.0.0.1:5678:5678"
    environment:
      - GENERIC_TIMEZONE="Europe/Madrid"
      - TZ="Europe/Madrid"
      - NODE_ENV=production
      - VUE_APP_URL_BASE_API=https://subdomain.domain.net/
      - N8N_HOST=subdomain.domain.net
      - N8N_PROTOCOL=https
      - N8N_PORT=5678
      - N8N_BASIC_AUTH_ACTIVE=true
      - N8N_BASIC_AUTH_USER=user
      - N8N_BASIC_AUTH_PASSWORD=password
      - DB_TYPE=mariadb
      - DB_MYSQLDB_DATABASE=n8n_workflows
      - DB_MYSQLDB_HOST=host.docker.internal
      - DB_MYSQLDB_USER=n8n_workflows
      - DB_MYSQLDB_PASSWORD=passowrd
      - N8N_SMTP_HOST=smtp.example.com
      - [email protected]
      - N8N_SMTP_PASS="password"
      - WEBHOOK_URL=https://subdomain.domain.net/
    volumes:
      - /home/n8n/.n8n:/home/node/.n8n

[Jon]

Hey @ericsala,

Is that with the webhook_url set as well? If you click on one of those failed assets does it show the correct URL?

For my setup I always use

server {
  listen 443 ssl;
  server_name n8n.my_domain.tld;
  location / {
    proxy_pass http://127.0.0.1:5678;
    proxy_set_header Connection '';
    proxy_http_version 1.1;
    chunked_transfer_encoding off;
    proxy_buffering off;
    proxy_cache off;
  }
  ssl_certificate # removed
  ssl_certificate_key # removed
}

Where the ssl side of things is normally managed by certbot.

[ericsala]

Hey @Jon I’ve just edited my last message with my docker-compose configuration.

[Jon]

Just spotted it, I can only assume it is still an issue on the nginx side somewhere assuming the URL you see in the network requests is correct.

[ericsala]

Maybe the problem is related with Runcloud (the gui I use to create webaplications).

The current configuration is inside: location.root

[Jon]

That I don’t know, I have not heard of runcloud before and I have only ever really configured nginx from the config files by dropping a proxy file into the sites-enabled folder.

[astroflash]

Hello, i faced the same issue and the solution was changing the stack on runcloud from NGINX to Native NGINX + Custom config.