Questionabout Supabase Vector Connection

Been having issues w/ connecting to my supabase vector table. I have created a chat input workflow connecting to my Supabase Vector DB. The whole flow doesn’t produce any errors, however, when asking specific questions regarding details of the DB it is unable to retrieve them. I properly connected Supabase, however, in checking the log it produces an empty output from Supabase.

When I created the documents table in Supabase I did so from this doc - LangChain | Supabase Docs. I then embedded my data using openAI.

Unsure of what might be causing this and seeing if anybody had any ideas of what I can do?



as for n8n version i am running it locally, and set it up within the last few days, unsure of exact version

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:

Hi @Tony_bree

Thanks for posting here and welcome to the community! :tada:

Looking at the documentation you’ve provided, it looks like the table is being queried correctly with the given function (match_documents).

Have you checked that these columns actually contain data in your table?
id, content, metadata

I’m not an expert on Supabase, but it looks like there’s no issue on the n8n side. Especially since you’re not getting any errors while executing.

Hi Ria, thank you for the response and verifying things look good on the n8n side! The issue was in the way I embedded my db so it was unable to query it properly

2 Likes

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