test_rag.py is failing on document inserts #112
Description
First noticed this in #111 but upon investigation this has been happening in the Container test for a few weeks now: https://github.com/containers/ramalama-stack/actions/workflows/test-container.yml
Example error:
===> starting 'test-rag'...
Using existing RamaLama and Llama Stack servers
===> test_rag_functionality: start
INFO:httpx:HTTP Request: GET http://localhost:8321/v1/models "HTTP/1.1 200 OK"
INFO:httpx:HTTP Request: POST http://localhost:8321/v1/vector-dbs "HTTP/1.1 200 OK"
INFO:httpx:HTTP Request: POST http://localhost:8321/v1/tool-runtime/rag-tool/insert "HTTP/1.1 500 Internal Server Error"
INFO:llama_stack_client._base_client:Retrying request to /v1/tool-runtime/rag-tool/insert in 0.492332 seconds
INFO:httpx:HTTP Request: POST http://localhost:8321/v1/tool-runtime/rag-tool/insert "HTTP/1.1 500 Internal Server Error"
INFO:llama_stack_client._base_client:Retrying request to /v1/tool-runtime/rag-tool/insert in 0.817746 seconds
INFO:httpx:HTTP Request: POST http://localhost:8321/v1/tool-runtime/rag-tool/insert "HTTP/1.1 500 Internal Server Error"
=== Llama Stack RAG Test ===
Connected to Llama Stack server at http://localhost:8321
Inference Parameters:
Model: granite3.2:2b
Sampling Parameters: {'strategy': {'type': 'greedy'}, 'max_tokens': 4096}
Stream: False
Using embedding model: all-MiniLM-L6-v2
Ingesting 3 test documents into vector database...
Error: Error code: 500 - {'detail': 'Internal server error: An unexpected error occurred.'}
===> test_rag_functionality: fail
RAG test script output above shows the failure details
===> 'test-rag' failed!