Fix: n8n webhook error: wrong URL, method or auth

    An n8n webhook error usually comes from the wrong URL, HTTP method, authentication mode, or inactive production endpoint. Test the boundary in order before debugging downstream nodes.

    Last updated 2026-09-21~6 min read
    n8nWebhooksHTTPPostgreSQLGoHighLevel

    Last updated: 2026-06-08 · Tested on self-hosted n8n 1.x with Webhook nodes and HTTP clients · ~5 min read

    Symptoms — what you're seeing

    • The calling application returns an n8n webhook error, timeout, or 404 response
    • The Webhook node works when you click Listen for test event, but the live request does nothing
    • n8n receives the request, but the workflow fails because fields are missing or the JSON shape is different from what the next node expects
    • A GoHighLevel, form, SaaS, or custom application sends data, but no n8n execution appears
    • The webhook responds successfully, but a downstream HTTP Request, database, or Code node fails

    Why this happens

    An n8n webhook has two separate endpoints: a test URL and a production URL. The test URL is available while the Webhook node is listening for a test event. The production URL is available only when the workflow is active. Calling the wrong endpoint produces the appearance of a broken webhook even when the node configuration is correct.

    The next failure points are the HTTP method, authentication mode, request headers, and payload shape. A Webhook node configured for POST does not accept a GET request as the same event. A node configured for header authentication does not accept a token placed in the request body. If the request reaches n8n but the next node reads a field that is not present, the webhook is working and the data contract is the problem.

    I diagnose the webhook boundary before touching downstream nodes. That separates transport errors from workflow errors and prevents a broken URL from being mistaken for an HTTP, credential, or mapping failure.

    The fix, step by step

    1. Open the Webhook node and identify the endpoint. Copy the Test URL and Production URL separately. Do not use the test URL in a live application. The production URL is the one to configure in GoHighLevel, your form provider, or the calling service.
    2. Activate the workflow. The production webhook requires an active workflow. Save the workflow, activate it, and then send a new request to the production URL. Do not use a previous test request as proof that the production endpoint works.
    3. Match the HTTP method exactly. If the Webhook node is configured for POST, send POST. Check the method in the calling application's webhook settings and in any n8n HTTP Request node that calls it. A URL can be correct while the request still fails because the method is wrong.
    4. Test the endpoint directly. Use curl with a clearly marked test value and the same method the integration uses:
      curl -i -X POST "https://n8n.example.com/webhook/replace-with-production-path" \
        -H "Content-Type: application/json" \
        -d '{"source":"manual-webhook-test","reference":"verify-2026-06-08"}'
      Replace the example host and path with your real production webhook URL before running this command — never share a live webhook URL publicly.
    5. Check whether n8n received the request. Open the workflow's executions and inspect the Webhook node input. If no execution exists, the problem is at the URL, network, proxy, method, authentication, or calling-service boundary. If the execution exists, continue with the payload and downstream checks.
    6. Compare the received payload with the next node's expressions. Open the Webhook node output and confirm the exact field paths. For example, a value sent as contact.email must be read from the actual nested structure, not from a flat email field. Use the expression picker against the received item instead of guessing the path.
    7. Check authentication and headers. Confirm that the Webhook node's authentication setting matches the request. Check header names, content type, bearer-token formatting, and any reverse-proxy rules — confirm the exact authentication mode and header names configured in your n8n instance before changing anything.
    8. Check the reverse proxy and public URL. If n8n runs behind Nginx, Traefik, Cloudflare, or another proxy, confirm that the public path forwards POST requests to n8n and that TLS terminates correctly. A proxy that forwards the browser request but blocks or rewrites webhook POSTs can create a 404, 405, or timeout before n8n sees anything.
    9. Inspect the first failing downstream node. If the Webhook node contains the expected payload, the webhook is fixed. Check the next node for an HTTP 401, expired credential, missing expression value, invalid JSON, or provider-specific validation error. For a GoHighLevel boundary, compare the request against the GoHighLevel to n8n webhook setup guide.

    Still stuck after testing the endpoint? Book a 30-min scope call. I'll inspect the webhook request, n8n execution, and downstream failure with you, then either fix the boundary or scope the remaining work. No upsell, no decks.

    Common variations

    • “The n8n webhook works in test but not live.” The calling system is probably using the test URL, or the workflow is not active. Replace the configured URL with the production URL and activate the workflow.
    • “The webhook returns 404.” Check for a copied URL from a deleted or renamed Webhook node, a missing path segment, or a reverse proxy route that does not forward the request to n8n.
    • “The webhook returns 405.” The request method does not match the Webhook node configuration. Check whether the caller sends GET, POST, PUT, or another method.
    • “The webhook returns 401 or 403.” Authentication is enabled and the request does not match the configured credentials, or an upstream proxy is rejecting the request before n8n receives it.
    • “The webhook runs but fields are empty.” Inspect the actual Webhook node output. The sender may be using form encoding, a nested JSON object, or a different field name than the downstream expression expects.
    • “The webhook is slow or times out.” Separate the webhook response from downstream work. Check the Webhook node's response behavior and inspect whether a long-running database, HTTP, or Code node is delaying the response — confirm the response mode configured on the Webhook node before changing it.
    • “The webhook triggers twice.” Check whether the sender retries after a slow response and whether two active workflows use the same integration. Add an idempotency key based on the sender's event ID before processing duplicate requests.

    When this isn't the issue

    If n8n receives the request and the Webhook node shows the expected data, the broken webhook is not the problem. Check these forks in order:

    • The next node cannot find a value. Rebuild the expression from the Webhook node output and confirm that each execution item has the expected field. Do not assume that a payload from one test event represents every production payload.
    • An external API rejects the request. Inspect the HTTP Request node response for the status code and response body. An HTTP 401 means the credential or token is wrong; a 429 means the provider is rate limiting the workflow. For OpenAI 429 errors, use the n8n OpenAI 429 diagnostic.
    • The workflow appears to run at the wrong time. That is a schedule configuration issue, not a webhook transport issue. Check the n8n schedule trigger timezone issue when the failing workflow also contains scheduled logic.

    The boundary is simple: first prove that the correct request reaches the production Webhook node, then validate the received payload, and only then debug downstream nodes. That sequence identifies whether the failure is the URL, method, authentication, proxy, payload mapping, or an external API.

    Need eyes on the deployment? Book a 30-min scope call or email Basit.

    Related pages

    Need this workflow customized or debugged? Book a 30-min scope call →

    Stuck configuring this workflow?

    I build and fix n8n workflows for agencies and SaaS teams. 30-minute scope call — bring the problem, leave with an estimate.

    Related pages