error: [/files] Error processing file: Request failed with status code 400

[mpl-ont]

What is your question?

I cant seem to get some file Uploads to work. Images, PDFs, XLSX, DOCX can be uploaded, but for other files like mp3s i am getting "An error occurred while processing the file"

I tried searching for people who had similar issues, but Threads like #2779 only covered different issues.

In addition, my Container Logs are giving me the following error:

rag_api-1         | 2025-01-24 16:21:48,816 - root - ERROR - Error during file processing: Could not detect encoding for ./uploads/67939b739511e6014c5b1fc0/testfile.mp3
rag_api-1         | Traceback: Traceback (most recent call last):
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_community/document_loaders/text.py", line 43, in lazy_load
rag_api-1         |     text = f.read()
rag_api-1         |   File "/usr/local/lib/python3.10/codecs.py", line 322, in decode
rag_api-1         |     (result, consumed) = self._buffer_decode(data, self.errors, final)
rag_api-1         | UnicodeDecodeError: 'utf-8' codec can't decode byte 0xfe in position 121: invalid start byte
rag_api-1         |
rag_api-1         | During handling of the above exception, another exception occurred:
rag_api-1         |
rag_api-1         | Traceback (most recent call last):
rag_api-1         |   File "/app/main.py", line 476, in embed_file
rag_api-1         |     data = loader.load()
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_core/document_loaders/base.py", line 31, in load
rag_api-1         |     return list(self.lazy_load())
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_community/document_loaders/text.py", line 46, in lazy_load
rag_api-1         |     detected_encodings = detect_file_encodings(self.file_path)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_community/document_loaders/helpers.py", line 50, in detect_file_encodings
rag_api-1         |     raise RuntimeError(f"Could not detect encoding for {file_path}")
rag_api-1         | RuntimeError: Could not detect encoding for ./uploads/67939b739511e6014c5b1fc0/testfile.mp3
rag_api-1         |
rag_api-1         | 2025-01-24 16:21:48,817 - root - INFO - Request POST http://rag_api:8000/embed - 400
LibreChat-API     | 2025-01-24 16:21:48 error: Error uploading vectors The request was made and the server responded with a status code that falls out of the range of 2xx: Request failed with status code 400. Error response data:
LibreChat-API     |
LibreChat-API     | 2025-01-24 16:21:48 error: [/files] Error processing file: Request failed with status code 400

Also, yes, the file is really a mp3 file that plays fine on my computer. It is just a couple of kb big.

I am using an external Apache as reverse Proxy, but also tried directly accessing the App at ip:3080, which results in the same error, as does using the bundled NGINX Proxy.

I am accessing the openai API trough a squid proxy.

I really dont know what else to try. I believe that i followed the Instructions for setting up rag_api, but maybe im missing something. Any Help is appreciated.

More Details

cat librechat.yaml
# For more information, see the Configuration Guide:
# https://docs.librechat.ai/install/configuration/custom_config.html

# Configuration version (required)
version: 1.2.1
# This setting caches the config file for faster loading across app lifecycle
cache: true

fileConfig:
  endpoints:
    default:
      fileLimit: 5
      fileSizeLimit: 10
      totalSizeLimit: 50
      supportedMimeTypes:
        - "image/.*"
        - "text/.*"
        - "application/.*"
        - "audio/.*"
        - "video/.*"
  serverFileSizeLimit: 1000
  avatarSizeLimit: 2

.env:

[...]
RAG_API_URL=http://host.docker.internal:8000
RAG_OPENAI_API_KEY=sk-abcd
[...]

I am using a mostly unchanged docker-conpose.yml, i just removed the ports from the client since i need to use Apache on the default ports. (and overriding does not work on lists, just adding onto them)

grep -Pv "^(\s*\#|^$)" docker-compose.override.yml
services:
  client:
    ports:
      - 8001:80
      - 4430:443
  api:
    image: ghcr.io/danny-avila/librechat-dev-api:b8b7f40e98c73b9dc85bb5f980d1e3e67dfe4e10
  mongodb:
    image: mongo:8.0.4
  rag_api:
    image: ghcr.io/danny-avila/librechat-rag-api-dev:9e4bb52e15d97856e3b69653c88d2cf1bb34324f
    extra_hosts:
      - "host.docker.internal:host-gateway"
  vectordb:
    image: pgvector/pgvector:0.8.0-pg17 # i also tried the bundled version, same error

What is the main subject of your question?

rag_api file upload

Screenshots

[danny-avila]

mp3 is not supported for embedding

[mpl-ont]

Thank you! Which file types are officially supported? I was not able to find a good answer in the documentation.