n8n: The requested webhook "..." is not registered
Last verified
The requested webhook "POST my-path" is not registered.What it means
n8n returns this with HTTP 404 when a request arrives at a webhook URL that no workflow is currently listening on. Every Webhook node has two URLs. The Test URL (containing /webhook-test/) only listens after you click to execute or listen for a test event in the editor; n8n's docs say it listens for 120 seconds, and community reports confirm it accepts a single call. The Production URL (containing /webhook/) listens only while the workflow is active (published).
So the error almost always means one of three things: the caller uses the test URL outside a test session, the workflow is not active, or the path or HTTP method does not match. n8n allows only one webhook per path and method, so copied workflows can collide.
What is not wrong: n8n is running and reachable; it answered with a 404.
What it looks like
$ curl -X POST https://n8n.example.com/webhook-test/orders
# HTTP 404: The requested webhook "POST orders" is not registered.
# production (workflow must be active):
$ curl -X POST https://n8n.example.com/webhook/ordersWhy it happens
- An external service is configured with the Test URL instead of the Production URL.
- The workflow is not active, so the production webhook is not registered.
- The HTTP method differs (the node expects POST, the caller sends GET).
- The path is wrong, or two workflows use the same path and method.
- Behind a reverse proxy, the public URL n8n generates does not match the one callers use.
How to fix it
- Compare the URL in the error with the Webhook node's Production URL, including method and path.
- Activate (publish) the workflow and use the Production URL in the external service.
- For testing, click to listen for a test event first, then send exactly one request within the listening window.
- If two workflows share a path, change one of them.
- On self-hosted instances behind a proxy, set the public webhook URL configuration so generated URLs match.
How to stop it recurring
Keep test URLs out of external services entirely, give each webhook a unique descriptive path, and add a monitoring call to production webhooks. On n8n Cloud, webhooks must respond within 100 seconds or callers get a 524 from Cloudflare.
Official documentation: docs.n8n.io/integrations/builtin/core-nodes/n8n-nodes-base.webhook/common-issues. See all n8n error messages or the n8n limits and quotas.
Related errors
Definitions
Quick answers
What does "The requested webhook "..." is not registered" mean in n8n?
n8n returns this with HTTP 404 when a request arrives at a webhook URL that no workflow is currently listening on. Every Webhook node has two URLs. The Test URL (containing /webhook-test/ ) only listens after you click to execute or listen for a test event in the editor; n8n's docs say it listens for 120 seconds, and community reports confirm it accepts a single call.
How do I fix "The requested webhook "..." is not registered"?
1. Compare the URL in the error with the Webhook node's Production URL, including method and path. 2. Activate (publish) the workflow and use the Production URL in the external service. 3. For testing, click to listen for a test event first, then send exactly one request within the listening window. 4. If two workflows share a path, change one of them. 5. On self-hosted instances behind a proxy, set the public webhook URL configuration so generated URLs match.
Error handling you can see in the workflow
Autonoly workflows have per-node retries, timeouts, continue-on-fail and error-handler branches, so a failing step can be retried or routed instead of stopping the whole run. Saved workflows have no loop node; lists are worked through by the AI agent or one run per trigger item.
Also see Visual workflow builder.