Self Hosted n8n Form Trigger won’t load page

Questions
1

Describe the problem/error/question

I’m trying to start a workflow in my self hosted n8n instance and the form simply doesn’t load. Neither the test url or the production url. A popular comes up, but it stays empty or fails to load.

What is the error message (if any)?

No explicit error message

Please share your

  • n8n version: 1.70.3
  • Database (default: SQLite): Postgres
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app): Docker Desktop on macOS
  • Operating system: macOS Monterey
2

Hey @ToHo Welcome to community!
Could you please elaborate more on your setup, did you connect Fully Qualified Domain Names (FQDNs) to your instance?

3

Can you explain to a low level user?

I have hosted it on a sub-domain of our website and I am accessing it through this subdomain. The ssl configuration is placed in my docker-compose.yml. Reverse Proxy is configured in our hardware firewall.

I can access the instance via https://n8n.oursite.de an I now can see, that the form urls in the node use http instead of https. Does that point to the problems root?

4

Yes it should point to your https://n8n.oursite.de
Here is docs Webhook URL

5

Ah perfect, thanks for the approach. I’ll try it tomorrow.

1 Like
6

I’m still getting a timeout in the forms.

n8n:
<<: *service-n8n
container_name: n8n
restart: unless-stopped
ports:
- 5678:5678
environment:
- N8N_HOST=n8n.oursite.de
- N8N_PROTOCOL=http
- N8N_PORT=5678
- N8N_WEBHOOK_URL=https://n8n.oursite.de
- N8N_SKIP_WEBHOOK_SSL_VERIFICATION=true
- N8N_ENFORCE_SETTINGS_FILE_PERMISSIONS=true
- NODE_ENV=production
- N8N_TRUSTED_PROXY_ADDRESSES=true
- N8N_EXTERNALHOOKS_WAITING_TIMEOUT=5000

This is the beginning of my docker-compose.yml in the n8n service section. I have also tried port 443 and https protocol following Claude’s instructions, but i’m getting 502 proxy error in that case.

7

There’s a mismatch in your configuration. You’re accessing n8n via HTTPS but configuring it internally with HTTP. If changing to HTTPS/443 causes a 502 proxy error, this indicates your reverse proxy can’t reach n8n on that port. You need to ensure your reverse proxy is correctly forwarding requests to the container’s port 5678.

8

Does this also apply if I run self-hosting on my own network? There I do not access n8n via 443, but via 5678 if i understand it correctly from what my it support says.

Strangely enough, the form link works if I change it to https:// and remove the port from the link.

9

@ToHo What reverse proxy are you using?

It is normal behaviour since your https://n8n.oursite.de link already pointed to http://your_ip:5678

Look at this video about nginx proxy manager NPM

10

I can see in the documentation, that I may be using the wrong variable. Documentation says
„WEBHOOK_URL“ and I’m using „N8N_WEBHOOK_URL“

Trying that on Monday.

2 Likes
11

Okay, so this has worked. The form opens with the correct URL. Thank you for the hint with the wrong webhook configuration.

But my workflow is still having problems. I’m trying to send a GET-request to one of our servers. When i open the URL manually in my browser i can send the request for a job id for example and i will get the job id information. With n8n i want to send the job id via the form input as an expression in the request-url. The problem is, that the test doesn’t really respond. It’s still on “waiting for trigger event” even though i sent the form. When i look into the executions i can see, that the http-request cancels with the following message:

The connection cannot be established, this usually occurs due to an incorrect host (domain) value

ENOTFOUND

Bildschirmfoto 2025-03-03 um 11.26.56
Bildschirmfoto 2025-03-03 um 11.26.561883×877 160 KB

This also accurs when i try to trigger the request manually. It seems like i have configured my get request wrong, but i can’t find the problem.

12

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.