Qdrant Vector Store Timing Out

Questions
1

Describe the problem/error/question

I’m trying to implement a simple RAG agent for some company docs. I’ve set up an N8N form and have uploaded 47 .txt files (eventually I want 1000+ of many formats, but I figured this was a good starting point). Qdrant seems to chew over this for about 5-10 minutes, but then suddenly stops and says that the operation was aborted. I have tried changing the upload method and file type.

What is the error message (if any)?

Problem in node ‘Qdrant Vector Store’
This operation was aborted

Please share your workflow

Share the output returned by the last node

Problem in node ‘Qdrant Vector Store’
This operation was aborted

Information on your n8n setup

  • n8n version:
    -1.102.3 (self-hosted)
  • Database (default: SQLite):
  • SQLite
  • n8n EXECUTIONS_PROCESS setting (default: own, main):
  • Default
  • Running n8n via (Docker, npm, n8n cloud, desktop app):
  • npm in a conda env
  • Operating system:
  • Windows
2

Hey @ste_f hope all is well. Welcome to the community.

How large is this set of documents in total?

1 Like
3

Hi there!
In total it’s 169KB. Eventually, I would like to go up to a 0.4GB dataset.

4

Yeah, 169KB is not much… It should be processed almost instantly. Out of curiosity - does it work for one or three files?

1 Like
5

No, it doesn’t. Weirdly, the exact same node setup worked for me yesterday – but I don’t think I’ve made any changes.

6

No, it doesn’t.

7

Have you been trying with other embedding model?

1 Like
8

Yeah, I’ve tried with mxbai-embed-large(successful but pretty bad performance) and switched to BGE-M3 which also worked until this error emerged.

9

So, I think the problem isn’t coming from the Vector Store.
I’m aware that coming from the AI models (?)