Have to have a deeper look into that to avoid such problems in the future.
For now. n8n normally deletes all Webhooks when it gets turned off. It seems that this did not happen for some unknown reason. The fix should be to simply delete the already existing webhook.
There should be no need to do that via Postman you can do it directly via the settings page (Settings → Webhooks) of your Github repository. The URL will be in this format: https://github.com/n8n-io/n8n/settings/hooks
(The above will not work as you are not part of n8n, is just an example)
Once it is deleted you should be able to simply deactivate and activate the workflow again.
Okay the problem was that I had also broken the webhooks’ endpoint. I’m moving to a production server right now. I guess it was just giving a misleading error…
I’m running n8n on a home mac mini and then ssh-ing localhost from the mac mini to my server. I’ve managed to use nginx to expose the UI on the server with https, and that’s all working.
But now I don’t know how to expose the webhooks endpoint on the server? I’m not even sure where n8n expects the webhooks url to be either. Any pointers? I’m new to a lot of this.
I got the webhooks working on my server once I hardcoded the base url for the webhooks in WebhookHelpers.js. I think it’s not currently picking up the WEBHOOK_TUNNEL_URL env var. Here’s a post I made about it.