Can't upload most files; Error 413 Request Entity Too Large #5042

kekkoudesu · 2024-12-19T02:00:12Z

kekkoudesu
Dec 19, 2024

What happened?

I am able to successfully upload images under 1MB. However, I can't upload images over 1MB, and I can't upload any other files whether they be a JPG, PDF, or anything else. For files over 1MB, it fails with 413 Request Entity Too Large:

response: "<html>\r\n<head><title>413 Request Entity Too Large</title></head>\r\n<body>\r\n<center><h1>413 Request Entity Too Large</h1></center>\r\n<hr><center>nginx/1.18.0 (Ubuntu)</center>\r\n</body>\r\n</html>\r\n"

This is strange, because in my nginx configuration file (which runs on the host), I have set:

client_max_body_size 100M;

This should give me plenty of headroom.

I have this set in my .env for RAG:

RAG_API_URL=http://host.docker.internal:8000
RAG_OPENAI_API_KEY=user_provide

Each user provides their own API key when they login.

I can't find any respective error in docker compose logs when the file is "too large". I only see the error in the client and in the javascript console.

With a .txt file that is 10 bytes large, it fails with error 500:

status: 500

statusText: "Internal Server Error"

The respective server error for this is under "Relevant log output".

It would seem to me that the issue is not the file size at all, but something else entirely.

Steps to Reproduce

Try to upload an image above 1MB or any other file.
Fail.

What browsers are you seeing the problem on?

No response

Relevant log output

rag_api-1         | USER_AGENT environment variable not set, consider setting it to identify your requests.
rag_api-1         | 2024-12-19 01:37:55,704 - root - INFO - Initialized embeddings of type: <class 'langchain_openai.embeddings.base.OpenAIEmbeddings'>
rag_api-1         | /app/store_factory.py:22: LangChainPendingDeprecationWarning: Please use JSONB instead of JSON for metadata. This change will allow for more efficient querying that involves filtering based on metadata. Please note that filtering operators have been changed when using JSONB metadata to be prefixed with a $ sign to avoid name collisions with columns. If you're using an existing database, you will need to create a db migration for your metadata column to be JSONB and update your queries to use the new operators.
rag_api-1         |   return AsyncPgVector(
rag_api-1         | 2024-12-19 01:37:55,933 - uvicorn.error - INFO - Started server process [1]
rag_api-1         | 2024-12-19 01:37:55,939 - uvicorn.error - INFO - Waiting for application startup.
rag_api-1         | 2024-12-19 01:37:55,939 - uvicorn.error - INFO - Application startup complete.
rag_api-1         | 2024-12-19 01:37:55,940 - uvicorn.error - INFO - Uvicorn running on http://0.0.0.0:8000 (Press CTRL+C to quit)
rag_api-1         | 2024-12-19 01:51:06,442 - httpx - INFO - HTTP Request: POST https://api.openai.com/v1/embeddings "HTTP/1.1 401 Unauthorized"
rag_api-1         | 2024-12-19 01:51:06,449 - root - ERROR - Failed to store data in vector DB | File ID: 6b55332e-8a40-46dd-a5fa-1f4b73f4d137 | User ID: 6762afa06610c7ae8ecff8f6 | Error: Error code: 401 - {'error': {'message': 'Incorrect API key provided: user_pro*ided. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}} | Traceback: Traceback (most recent call last):
rag_api-1         |   File "/app/main.py", line 326, in store_data_in_vector_db
rag_api-1         |     ids = await vector_store.aadd_documents(
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_core/vectorstores/base.py", line 324, in aadd_documents
rag_api-1         |     return await run_in_executor(None, self.add_documents, documents, **kwargs)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_core/runnables/config.py", line 588, in run_in_executor
rag_api-1         |     return await asyncio.get_running_loop().run_in_executor(
rag_api-1         |   File "/usr/local/lib/python3.10/concurrent/futures/thread.py", line 58, in run
rag_api-1         |     result = self.fn(*self.args, **self.kwargs)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_core/runnables/config.py", line 579, in wrapper
rag_api-1         |     return func(*args, **kwargs)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_core/vectorstores/base.py", line 287, in add_documents
rag_api-1         |     return self.add_texts(texts, metadatas, **kwargs)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_community/vectorstores/pgvector.py", line 561, in add_texts
rag_api-1         |     embeddings = self.embedding_function.embed_documents(list(texts))
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_openai/embeddings/base.py", line 588, in embed_documents
rag_api-1         |     return self._get_len_safe_embeddings(texts, engine=engine)
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/langchain_openai/embeddings/base.py", line 483, in _get_len_safe_embeddings
rag_api-1         |     response = self.client.create(
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/openai/resources/embeddings.py", line 124, in create
rag_api-1         |     return self._post(
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/openai/_base_client.py", line 1280, in post
rag_api-1         |     return cast(ResponseT, self.request(cast_to, opts, stream=stream, stream_cls=stream_cls))
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/openai/_base_client.py", line 957, in request
rag_api-1         |     return self._request(
rag_api-1         |   File "/usr/local/lib/python3.10/site-packages/openai/_base_client.py", line 1061, in _request
rag_api-1         |     raise self._make_status_error_from_response(err.response) from None
rag_api-1         | openai.AuthenticationError: Error code: 401 - {'error': {'message': 'Incorrect API key provided: user_pro*ided. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}
rag_api-1         |
rag_api-1         | 2024-12-19 01:51:06,451 - root - INFO - Request POST http://rag_api:8000/embed - 20

But the API key is correct. This can be verified by the fact that images under 1MB can be uploaded and OpenAI responds to them.

Screenshots

No response

Code of Conduct

I agree to follow this project's Code of Conduct

Answered by kekkoudesu

Dec 19, 2024

File upload works now. Don't use Nginx; use Traefik: https://www.librechat.ai/docs/remote/traefik

View full answer

danny-avila · 2024-12-19T02:03:08Z

danny-avila
Dec 19, 2024
Maintainer

From your logs:

rag_api-1 | openai.AuthenticationError: Error code: 401 - {'error': {'message': 'Incorrect API key provided: user_pro*ided. You can find your API key at https://platform.openai.com/account/api-keys.', 'type': 'invalid_request_error', 'param': None, 'code': 'invalid_api_key'}}

Please read this for more information on configuring the RAG API, which is used for non-image files:
https://www.librechat.ai/docs/configuration/rag_api#docker-setup---quick-start

2 replies

kekkoudesu Dec 19, 2024
Author

I asked here after following the documentation on that page. I filled out the two environment variables like so:

RAG_API_URL=http://host.docker.internal:8000
RAG_OPENAI_API_KEY=user_provided

kekkoudesu Dec 19, 2024
Author

Instead of user_provided, I tried using my own API key in RAG_OPENAI_API_KEY. Suddenly file upload started working for the 10 byte text file, and PDFs under 1MB, etc.

However, it still fails with Error 413 for files larger than 1MB.

kekkoudesu · 2024-12-19T03:07:04Z

kekkoudesu
Dec 19, 2024
Author

File upload works now. Don't use Nginx; use Traefik: https://www.librechat.ai/docs/remote/traefik

0 replies

roman-doubrava · 2025-06-12T15:33:30Z

roman-doubrava
Jun 12, 2025

@danny-avila i had the same issue 513, but mitigated the NGINX limit of 1MB in nano /etc/nginx/nginx.conf


        # Increase client body buffer size
        client_body_buffer_size 50M;

        # Increase timeouts
        client_body_timeout 600s;
        client_header_timeout 600s;
        send_timeout 600s;

        client_max_body_size 20M;
        proxy_read_timeout 500s;
        proxy_send_timeout 500s;
        proxy_connect_timeout 600s;

However now the upload hangs and never finishes for larger files 1 MB

1 reply

kekkoudesu Jun 12, 2025
Author

I ended up just using Traefik; it didn't seem like there was a way to get Nginx/Caddy to work.

josephgimenez · 2025-08-11T17:05:34Z

josephgimenez
Aug 11, 2025

In case it helps anyone else, we were able to work around this in kubernetes by annotating the librechat ingress controller with:

nginx.ingress.kubernetes.io/proxy-body-size: Xm

where X is the maximum size in megabytes

0 replies

Uh oh!

Can't upload most files; Error 413 Request Entity Too Large #5042

Uh oh!

Uh oh!

kekkoudesu Dec 19, 2024

What happened?

Steps to Reproduce

What browsers are you seeing the problem on?

Relevant log output

Screenshots

Code of Conduct

Replies: 4 comments · 3 replies

Uh oh!

danny-avila Dec 19, 2024 Maintainer

Uh oh!

kekkoudesu Dec 19, 2024 Author

Uh oh!

Uh oh!

kekkoudesu Dec 19, 2024 Author

Uh oh!

kekkoudesu Dec 19, 2024 Author

Uh oh!

Uh oh!

roman-doubrava Jun 12, 2025

Uh oh!

kekkoudesu Jun 12, 2025 Author

Uh oh!

josephgimenez Aug 11, 2025

kekkoudesu
Dec 19, 2024

Replies: 4 comments 3 replies

danny-avila
Dec 19, 2024
Maintainer

kekkoudesu Dec 19, 2024
Author

kekkoudesu Dec 19, 2024
Author

kekkoudesu
Dec 19, 2024
Author

roman-doubrava
Jun 12, 2025

kekkoudesu Jun 12, 2025
Author

josephgimenez
Aug 11, 2025