- Test webhook (with
testin URL) works, production URL 404s - External service shows delivery failure or timeout
- Workflow shows no execution, or only manual executions
1. Confirm you are calling the right URL
- Test URL contains
/webhook-test/, production is/webhook/. Many zaps paste the test URL into production. - Path is case-sensitive and must match the Webhook node path exactly, including trailing slash handling.
- Re-copy the production URL from the Webhook node after activating the workflow.
2. Is the workflow actually active?
- Toggle Active on. In queue mode, check that the main and worker both see the same activation state.
- Repro: trigger with
curl -i https://YOUR_HOST/webhook/YOUR-PATHand watchExecutions. No execution = not reaching n8n.
3. Auth, headers, and method mismatch
- Webhook node method must match caller (GET vs POST). Check
Allow GETif the caller uses GET. - If auth is on, the caller must send the exact header or basic auth. A silent 401 often looks like "not firing".
- Inspect response headers with
curl -i— a 308 or 404 usually reveals the proxy rewriting the path.
4. Reverse proxy and Nginx path rewriting
- Behind Nginx,
location /webhook/withproxy_pass http://localhost:5678/must not strip the prefix your workflow expects. - Check
X-Forwarded-Proto— mixed http/https causes webhook URL mismatch in the UI.
5. Quick fix sequence
1) Deactivate, re-activate, re-copy production URL 2) curl -i production URL from outside the VPS 3) If 404 → fix Nginx location. If 401 → fix auth. If 200 but no execution → check workflow activation and queue mode 4) Add a Respond to Webhook node so the caller gets 200 immediately and you can inspect the execution
When to escalate
If the webhook reaches n8n but the downstream API times out or the queue leaves executions/temp behind, see the API timeout and Postgres SSH guides. The hosted instance at myn8n-automation.org proxies / to port 5678 and keeps /services/ and /preview/ as exact static routes — so this site's funnel is isolated from n8n proxy misconfig.
Grab the free diagnostic checklist — no email wall
Copy this pre-filled request, paste it wherever you asked for help, and you will get a targeted answer without sharing secrets. Nothing is tracked on this page.
Subject: n8n webhook — diagnostic request Describe: - Webhook or HTTP Request node + method/path - Test URL works? Production URL status (curl -i)? - Activation + queue mode? Postgres SSH in use? Paste sanitized config (redact secrets) and I will reply with the exact fix. — myn8n-automation.org/guides/n8n-webhook-not-firing/
Claude Weekly Dev Summary for n8n ($29)
Want a production-tested automation ready to import today? Automatically turns any GitHub repo into an executive Friday 5pm summary delivered to Slack or Discord. 8 nodes, zero code, verified execution proof included.
Want a hands-on fix?
The diagnostic above is free to run. If you want me to diagnose and repair the workflow for you with test evidence and handoff notes, direct checkout is available instantly.
Instant Diagnostic ($15) — Direct CheckoutTruth disclosure: This guide shows owned capability and a demo eight-node GitHub activity → Claude-assisted weekly summary → Slack/Discord workflow with local JSON/README/image evidence. No client result, sale, review, accepted bounty, earnings, or production customer outcome is claimed. The named preview businesses did not commission this work.