search

FlowiseAI / Flowise

Drag & drop UI to build your customized LLM flow
https://flowiseai.com
Apache License 2.0
31.26k stars 16.28k forks source link

Document-store -> vectorstore -> insert -> Error: documentStoreServices.insertIntoVectorStore - Error: documentStoreServices._insertIntoVectorStoreWorkerThread - TypeError: fetch failed #3417

Open dev3py opened 1 week ago

dev3py commented 1 week ago

Describe the bug While upserting a new Document Store getting a 500 error

To Reproduce Steps to reproduce the behavior:

  1. Adding an ollama embeddings nomic-embed-text:latest
  2. Adding a Faiss as a Vector Store
  3. Upsert
  4. See error

Expected behavior The success upsert message is expected, but the 500 error is coming Status: 500 Error: documentStoreServices.insertIntoVectorStore - Error: documentStoreServices._insertIntoVectorStoreWorkerThread - TypeError: fetch failed

Screenshots image image

Flow If applicable, add exported flow in order to help replicating the problem.

Setup

AlamokCR commented 1 week ago

I have the exact issue, it doesn't mater if I use docker or install all manually.

HenryHengZJ commented 6 days ago

is your FAISS dimension compatible with nomic embed text embedding dimension?

shighcloud-ea commented 5 days ago

same issue for me

Status: 500 Error: documentStoreServices.insertIntoVectorStore - Error: documentStoreServices._insertIntoVectorStoreWorkerThread - TypeError: fetch failed

suresh-jbt commented 1 day ago

@shighcloud-ea @HenryHengZJ @AlamokCR @dev3py I found the solution and its worked In the Ollama node configuration, use http://host.docker.internal:11434 instead of http://localhost:11434.

Since Ollama is installed on the host machine and not within Docker, Docker containers cannot connect to the host’s services via localhost. By using host.docker.internal, you enable Docker to communicate with services on the host.