OpenAI node 400 error


I was building a conversational agent using the conversational AI node with data being webhooked across the user’s chat ‘endpoint’ (just a conversation management platform/CRM). I’d done something like this on a different n8n instance in the past, but I keep getting a 400 error msg for some reason. The thing is, I cannot find the directory that the message refers to. I tried every means I knew how to and scanned the entire file system but I have not got such a directory that the err msg stems from. Does anyone have any clue how I should troubleshoot this? I’m really at a loss here.

Here’s a screenshot of the error message:

Here’s a screenshot of the workflow:

For context, this is on a self-hosted n8n; version 1.47.3. It’s hosted on an Ubuntu system and run via Docker. DB and EXECUTIONS_PROCESS are default

It looks like your topic is missing some important information. Could you provide the following if applicable.

  • n8n version:
  • Database (default: SQLite):
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • Operating system:

Hey @paulugwuanyi,

Looking at that error it looks like you are referencing a value that is empty, It would be handy if you could share the workflow json and an example payload so we can test it and see if we can reproduce the issue.

Hi Jon,

Thanks for getting back to me. I’ll be sending these to you privately if you do not mind

Hey @paulugwuanyi,

That sounds ok but it is normally better if you can share here and remove anything sensitive so that you are not waiting for just me, There are a few members of the team and helpful community members that may also be able to provide an answer.