infiniflow / ragflow

RAGFlow is an open-source RAG (Retrieval-Augmented Generation) engine based on deep document understanding.
https://ragflow.io
Apache License 2.0
16.44k stars 1.68k forks source link

[Question]: Is it possible to store all the knowledge base data into single bucket? #1098

Open surya9teja opened 3 months ago

surya9teja commented 3 months ago

Describe your problem

Hi,

Is it possible to leverage Docker Compose's .env file to centralize knowledge base data storage within a single bucket, eliminating the creation of individual buckets per knowledge base? If this necessitates source code modifications, could you point me towards the specific file(s) requiring changes?

KevinHuSh commented 3 months ago

Could you tell me the specific reason.

surya9teja commented 3 months ago

I am planning to use AWS S3 as my backend storage (replacing minio host with aws s3 host url didn't even know whether it works) so I want ragflow to store all the data into single bucket something like "sample.example.bucket" .

If you have any better approach for this replacing minio with aws s3 as a backend storage via access and secret key. Please suggest.

KevinHuSh commented 2 months ago

You could have a look at rag/util/minio_conn.py. If you want to centrolize or something like that, just change this file.