LightRAG icon indicating copy to clipboard operation
LightRAG copied to clipboard

Production version not working anymiore after last updates

Open ParisNeo opened this issue 9 months ago • 5 comments

Do you need to file an issue?

  • [ ] I have searched the existing issues and this bug is not already filed.
  • [ ] I believe this is a legitimate bug, not just a question or feature request.

Describe the bug

When I run lightrag-gunicorn, I get this:

================================================================================
MAIN PROCESS INITIALIZATION
Process ID: 803610
Workers setting: 1
================================================================================


Error: unsupported operand type(s) for *: 'NoneType' and 'int'

The lightrag-server works but not the gunicorn version. I use .env to store my configurations in my launch folder as always. @danielaskdd do you know this issue.

Sorry for being abscent lately.

Steps to reproduce

No response

Expected Behavior

No response

LightRAG Config Used

Paste your config here

Logs and screenshots

No response

Additional Information

  • LightRAG Version:
  • Operating System:
  • Python Version:
  • Related Issues:

ParisNeo avatar Mar 21 '25 13:03 ParisNeo

no answer when a question is made to the rag system.

i can see the post to embeddings to "map" the query vectors

DEBUG:HTTP Request: POST https://api.openai.com/v1/embeddings "200 OK" DEBUG:close.started DEBUG:close.started DEBUG:close.started DEBUG:close.started DEBUG:close.complete DEBUG:close.complete DEBUG:close.complete DEBUG:close.complete DEBUG:close.started DEBUG:close.complete

and after that.. nothing there is no search of the vectors anywhere

bzImage avatar Mar 21 '25 16:03 bzImage

LightRAG is using preload mode of Gunicorn which Windows is not supported.

danielaskdd avatar Mar 22 '25 02:03 danielaskdd

I'm not using windows. I'm on linux ubuntu: And this error doesn't seem to be linked to the system:

Error: unsupported operand type(s) for *: 'NoneType' and 'int'

ParisNeo avatar Mar 25 '25 10:03 ParisNeo

maybe we need a debug mode to log more infos

ParisNeo avatar Mar 25 '25 10:03 ParisNeo

Just find out that.. if you are using postgres backend.. something its wrong.. and .. it takes up to 20-30 seconds to timeout from the "query" and fail.. meanwhile your cpu goes to 100% ..

bzImage avatar Mar 25 '25 15:03 bzImage