[BUG] nanographrag & ollama file indexing and embeddings error

[scheckley]

Description

installed kotaemon using the Docker method. I've set up the nano_graphrag as described in the readme.

Indexing error uploading pdf files:

INFO:httpx:HTTP Request: POST http://x.x.x.x/v1/embeddings "HTTP/1.1 500 Internal Server Error"
INFO:openai._base_client:Retrying request to /embeddings in 0.480649 seconds

I'm using ollama serving llama 3.2 and nomic-embed-text. All tests return fine, and I can get a response from ollama using curl to the embeddings api. Simple queries are ok, but quality of results is poor. Embeddings don't seem to be working at all. I've had previous installations running better with the same embeddings settings but no graphrag.

I'm just wondering if anybody else has seen this error? Thanks for any help.

OS

Linux

Additional information

Docker container deployed on Openshift.