Problem running workflow Unrecognized node type: @n8n/n8n-nodes-langchain.chatResponse

Describe the problem/error/question

Since update of my container, I can’t run a simple chat node

What is the error message (if any)?

Problem running workflow Unrecognized node type: @n8n/n8n-nodes-langchain.chatResponse

Please share your workflow

Share the output returned by the last node

none as the workflow is stopped at its starts

Information on your n8n setup

  • n8n version: 2.9.3
  • Database (default: SQLite): postgresql
  • n8n EXECUTIONS_PROCESS setting (default: own, main): do not have any environment variable of this name on that container
  • Running n8n via (Docker, npm, n8n cloud, desktop app): docker
  • Operating system: standard docker image

Hi @Cris_R

n8n has all the AI stuff built-in natively now. if you have the old @n8n/n8n-nodes-langchain package sitting in your community nodes list, it conflicts with the core system. it completely breaks the nodes and throws that exact unrecognized error.

check settings > community nodes in your n8n workspace. if it’s in there, just uninstall it and restart your docker container.

if that list is empty, might just be a corrupted docker pull. forcing a fresh pull of the 2.9.3 image usually sorts it out from what i’ve seen.

Hi,

Thanks a lot for this very quick answer :slight_smile: , I am using portainer to deploy the stacks and I repulled so many different images… it might as you said a file corrupted. I will retry this evening and will post the result.

Belle journée

C

1 Like

Hey, check your settings > community nodes and see if you have the old @n8n/n8n-nodes-langchain package installed there. The AI/langchain nodes are built into n8n core now so if you still have that old community package hanging around it conflicts with the native nodes and causes exactly this error. Uninstall it from community nodes, restart your container, and it should work. If community nodes is already empty then try doing a fresh pull of the 2.9.3 image with docker pull n8nio/n8n:2.9.3 before recreating the container, sometimes Docker caches old layers.

Hey, this usually happens when there’s an old @n8n/n8n-nodes-langchain community package hanging around that conflicts with the built-in AI nodes in newer versions. Go to Settings > Community Nodes and check if langchain is listed there, if so uninstall it and restart your container. The AI/langchain nodes are built into n8n now so you don’t need the separate package anymore, and having both causes exactly this kind of “unrecognized node type” error. If community nodes is already empty try doing a clean pull of the 2.9.3 image since pulling multiple times through Portainer can sometimes leave stale layers.

The chatResponse node got renamed/replaced in a recent update, the current node type is @n8n/n8n-nodes-langchain.respondToChat (shows up as “Respond to Chat” in the editor). Delete the old node from your workflow and add the new Respond to Chat node in its place, that should fix it right away.

Hi,

I did that it does not worked, file corruption in upgrade process must be a good reason.

Thanks a lot for the input and the help.

1 Like

It was indeed the node type, it was solved by applying in the json directly the new naming : @n8n/n8n-nodes-langchain.chatTrigger as respondToChat was generating the error.

At least it’s how the node is now called in the 2.10.2 version used here. What confused me is the fact that deleting the node and recreating a new one from the list of available nodes was not sufficient, I endeed up with pasting the error workflow into a notepad page and then replace manually the name of the node type (“type”: “@n8n/n8n-nodes-langchain.chatTrigger”,).

Happy to put this post in “solved mode” :grinning_face: and again thanks to all who helped !

1 Like

This topic was automatically closed 7 days after the last reply. New replies are no longer allowed.