self hosting
N8N SSL Setup
For public n8n deployments, use HTTPS either through a reverse proxy or n8n's SSL settings, then make sure generated editor and webhook URLs also use HTTPS.
Match your incident first
Start with the symptom you can prove
Test URL works, production URL returns 404
First check: Open the Webhook node and compare the active production URL with the external app callback URL.
Wrong fix to avoid: Do not rebuild the whole workflow before proving whether the active production endpoint exists.
Verify: Activate the workflow, send a POST to the production URL, and confirm one new production execution appears.
Production URL shows localhost, container hostname, or http instead of public HTTPS
First check: Check the public URL shown in n8n and compare it with WEBHOOK_URL and reverse proxy forwarded headers.
Wrong fix to avoid: Do not paste the internal Docker service name into external SaaS callback settings.
Verify: The Webhook node displays the public HTTPS domain and an external curl reaches n8n without redirect loops.
External app times out but n8n logs show no execution
First check: Run an external curl smoke test and inspect reverse proxy access logs for the same timestamp.
Wrong fix to avoid: Do not rotate credentials when the request never reaches n8n.
Verify: Proxy access log and n8n execution log both show the same request.
- Use when
- self-hosted, HTTPS, SSL
- First check
- Confirm workflows and credentials survive a restart before sending real traffic.
- Time to check
- 5-10 minutes
- Next step
- Match the symptom, then run the verification checks.
Independent third-party notes. n8n is a trademark of its owner and is referenced only for compatibility and troubleshooting context.
Quick Answer
For public n8n deployments, use HTTPS either through a reverse proxy or n8n's SSL settings, then make sure generated editor and webhook URLs also use HTTPS.
Does this match your symptom?
Self-hosted Docker setup looks risky
The instance runs, but persistence, .env files, public URLs, HTTPS, Postgres, or backups are unclear.
First check: Confirm workflows and credentials survive a restart before sending real traffic.
Version awareness
Last reviewed 2026-05-21
Key Facts
- Security variable group
- n8n documents SSL-related environment variables under security settings.
- Public URL alignment
- N8N_PROTOCOL and WEBHOOK_URL should reflect HTTPS when public access uses HTTPS.
- Common production approach
- Terminate TLS at a reverse proxy and forward traffic to n8n internally.
- Webhook dependency
- External apps often require HTTPS webhook targets.
Production Diagnostic Matrix
Turn checks into a brief| Exact symptom or log | Likely cause | First check | Wrong fix to avoid | Verification |
|---|---|---|---|---|
| Test URL works, production URL returns 404 | Workflow is not active, caller is using the test URL, or the production path was changed after activation. | Open the Webhook node and compare the active production URL with the external app callback URL. | Do not rebuild the whole workflow before proving whether the active production endpoint exists. | Activate the workflow, send a POST to the production URL, and confirm one new production execution appears. |
| Production URL shows localhost, container hostname, or http instead of public HTTPS | WEBHOOK_URL, N8N_HOST, N8N_PROTOCOL, or proxy headers do not match the public domain. | Check the public URL shown in n8n and compare it with WEBHOOK_URL and reverse proxy forwarded headers. | Do not paste the internal Docker service name into external SaaS callback settings. | The Webhook node displays the public HTTPS domain and an external curl reaches n8n without redirect loops. |
| External app times out but n8n logs show no execution | DNS, firewall, Cloudflare, reverse proxy, or path routing blocks the request before it reaches n8n. | Run an external curl smoke test and inspect reverse proxy access logs for the same timestamp. | Do not rotate credentials when the request never reaches n8n. | Proxy access log and n8n execution log both show the same request. |
| Workflow receives payload but caller gets wrong response or hangs | Respond to Webhook is missing, configured for the wrong response mode, or waits for a branch that does not finish. | Check the Webhook node response mode and the Respond to Webhook node path. | Do not add arbitrary Wait nodes to hide timeout behavior. | Caller receives the intended status code and body within the provider timeout window. |
| Webhook path works internally but fails through reverse proxy | Proxy location, base path, body size, TLS termination, or forwarded proto/host headers are wrong. | Compare direct container access with the public reverse proxy route and check forwarded headers. | Do not disable HTTPS verification as a proxy troubleshooting shortcut. | Public HTTPS request preserves method, path, body, and host all the way to n8n. |
Recommended Steps
- Choose whether TLS terminates at a reverse proxy or n8n itself.
- Install and renew certificates through the chosen method.
- Set N8N_PROTOCOL=https for public HTTPS deployments.
- Set WEBHOOK_URL to the HTTPS public base URL.
- Test the editor and a production webhook over HTTPS.
Verification
- The browser shows a valid HTTPS connection.
- Generated webhook URLs start with https://.
- External webhook providers can deliver requests without TLS errors.
First Commands / Checks
curl -i -X POST https://automation.example.com/webhook/example-path \
-H "content-type: application/json" \
-d '{"smokeTest":true}' docker compose ps docker compose logs n8n --tail=100 docker compose exec n8n printenv WEBHOOK_URL N8N_HOST N8N_PROTOCOL N8N_EDITOR_BASE_URL Safe Copyable Config
location / {
proxy_pass http://n8n:5678;
proxy_http_version 1.1;
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 Upgrade $http_upgrade;
proxy_set_header Connection "upgrade";
} Warnings
- Do not expose credentials or webhook payloads over plain HTTP on public networks.
- Certificate renewal failures can break webhook integrations.
Best For
- Public self-hosted n8n instances.
- Webhook integrations from third-party services that require HTTPS.
- Deployments that need secure browser access to credentials and workflow data.
Not For
- Local-only disposable testing on localhost.
- Situations where a managed platform already handles TLS completely.
- Cases where the domain or proxy route is still not configured.
Common Mistakes
- Serving the editor over HTTPS while generated webhook URLs still use http://.
- Letting certificates expire without monitoring.
- Exposing n8n publicly over plain HTTP.
- Confusing TLS termination at the proxy with SSL settings inside n8n.
Examples
Browser URL:
https://automation.example.com
Generated webhook URL:
https://automation.example.com/webhook/order-created
Mismatch to fix:
http://automation.example.com/webhook/order-created Valid certificate: yes
Auto-renewal: yes
N8N_PROTOCOL=https: yes
WEBHOOK_URL=https://...: yes
External webhook smoke test: yes FAQ
Should TLS terminate at the reverse proxy or n8n?
Many production deployments terminate TLS at a reverse proxy. The key is that public URLs and n8n endpoint settings still generate HTTPS links.
Why do webhooks show http:// after I enabled HTTPS?
The public endpoint variables may still indicate HTTP or the old base URL. Check N8N_PROTOCOL and WEBHOOK_URL.
Is HTTPS required?
For public deployments handling credentials, webhook payloads, or browser sessions, HTTPS should be treated as required.