Webhook Test Mode 500 – Cannot read properties of undefined (reading 'name') (Cloud 2.6.3)

[NOMOS]

Warning: I used ai to format and write this post. My english is not great, from Brazil.

Environment

• n8n Cloud

• Version: 2.6.3

• Webhook path: whatsapp-webhook

• Method: POST

• Using /webhook-test/whatsapp-webhook for manual testing

---

Issue

When Twilio sends a POST request to the test webhook, it returns HTTP 500.

Twilio error:

11200 – Twilio’s servers were unable to fetch a non-error response from the designated URL.

Response body from n8n:

{"code":0,"message":"Workflow Webhook Error: Workflow could not be started!"}

---

n8n Stack Trace

TypeError: Cannot read properties of undefined (reading 'name')
    at WebhookContext.getChildNodes
    at checkResponseModeConfiguration
    at Webhook.webhook
    at WebhookService.runWebhook
    at executeWebhook
    at test-webhooks.ts

---

What’s happening

• The workflow never starts

• The error occurs before execution

• It fails inside checkResponseModeConfiguration()

• This only happens when hitting /webhook-test/...

• The workflow is in listening mode

Has anyone encountered this in Cloud 2.6.3?
Is this a known issue related to webhook response mode validation in test mode?

Any guidance would be appreciated.

[NOMOS]

Addendum: When I switch to the production webhook in twilio, everything works fine. This only happens when using the test webhook.

[achamm]

Hi @NOMOS !

Known bug — #13836 has your exact stack trace, closed as a dupe of #13086 which is still open.

What’s happening is checkResponseModeConfiguration tries to walk the child nodes from your Webhook and hits an undefined reference. Test mode runs this check more strictly than production does, which is why production works fine.

Most likely your Webhook’s “Respond With” is set to “Using ‘Respond to Webhook’ Node” but that node is either missing or not wired up properly. Could also be an orphaned reference from a node you deleted.

Check that setting first — if you don’t actually need a Respond to Webhook node, switch it to “Immediately” or “When Last Node Finishes”. If it still acts up, export the workflow JSON, delete it, reimport. That tends to clean up stale references.

Or honestly since production works, just point Twilio at the production URL and skip test mode until they fix this.

I hope this helps, and feel free to ask if you have any questions!

[NOMOS]

First off, thanks for the response, I really appreciate it.

This is so weird, because my webhook node was set to immediately already. I tried switching things up and duplicating the workflow to clear cache to see if that helped but still didn’t work. At a dead end, and debugging through the production url is not an option. Do you have any other suggestions?

[NOMOS]

Update: Duplicating didn’t work, but deleting and reimporting did work. Thank you so much for the help!

[achamm]

Your welcome! Anytime!