genai-stack icon indicating copy to clipboard operation
genai-stack copied to clipboard

No returns from requests

Open tobibuchmann opened this issue 2 years ago • 4 comments

Everything seems to start smoothly, but I don't get any returns wehen I use the different apps. It says for instance "Model: llama2:13b RAG: Disabled" but no further response when I ask "How can I calculate age from date of birth in Cypher?" what could be the issue here?

tobibuchmann avatar Nov 03 '23 15:11 tobibuchmann

I have had the same issue. I downed with docker compose and rebuilt. This is how it got solved. Haven't investigated the logs so maybe it's a good moment to take a look at yours? It happened a few times and I haven't caught a pattern. Using Docker Desktop 4.25 on WSL. Issue was present with 4.24 too.

zooninja avatar Nov 03 '23 15:11 zooninja

Yes you need to use 4.24.1 at least, there was a bug in Docker in between that broke pytorch.

If you can enable langsmith you should be able to see what's happening behind the scenes or use verbose=True for the LLM or chains.

jexp avatar Nov 03 '23 15:11 jexp

Thanks for your ideas. Unfortunately I was't able to solve the issue yet. I upgrade from Docker 4.23 to 4.25 but that didn't solve it. I don't have access to langsmith. Where would be the best way to use verbose = True? Or what is the best way to debug? Which logs should I check? Thanks BTW, this is from my VS Code terminal: genai-stack-api-1 | INFO: 127.0.0.1:52148 - "GET / HTTP/1.1" 200 OK genai-stack-bot-1 | 2023-11-06 15:07:32.688 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:07:32.701 LLM: Using Ollama: llama2:13b genai-stack-api-1 | INFO: 127.0.0.1:48336 - "GET / HTTP/1.1" 200 OK genai-stack-bot-1 | 2023-11-06 15:07:50.575 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:07:50.590 LLM: Using Ollama: llama2:13b genai-stack-api-1 | INFO: 127.0.0.1:54948 - "GET / HTTP/1.1" 200 OK genai-stack-bot-1 | 2023-11-06 15:08:16.286 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:08:16.299 LLM: Using Ollama: llama2:13b genai-stack-api-1 | INFO: 127.0.0.1:39352 - "GET / HTTP/1.1" 200 OK genai-stack-bot-1 | 2023-11-06 15:08:40.502 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:08:40.516 LLM: Using Ollama: llama2:13b genai-stack-bot-1 | 2023-11-06 15:08:44.372 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:08:44.386 LLM: Using Ollama: llama2:13b genai-stack-api-1 | INFO: 127.0.0.1:37368 - "GET / HTTP/1.1" 200 OK genai-stack-api-1 | INFO: 127.0.0.1:60360 - "GET / HTTP/1.1" 200 OK genai-stack-api-1 | INFO: 127.0.0.1:60272 - "GET / HTTP/1.1" 200 OK genai-stack-bot-1 | 2023-11-06 15:22:55.763 Embedding: Using SentenceTransformer genai-stack-bot-1 | 2023-11-06 15:22:55.797 LLM: Using Ollama: llama2:13b genai-stack-api-1 | INFO: 127.0.0.1:49086 - "GET / HTTP/1.1" 200 OK

tobibuchmann avatar Nov 06 '23 15:11 tobibuchmann

@tobibuchmann is this one resolved for you now?

slimslenderslacks avatar Dec 22 '23 22:12 slimslenderslacks