Find answers from the community

Updated 9 months ago

I'm trying to upload embeddings to a

I'm trying to upload embeddings to a local vector store , but i'm facing issues with timeout. How to debug it?
T
w
L
15 comments
Do you have logging enabled? It should give a pretty good idea

Plain Text
logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))


https://docs.llamaindex.ai/en/stable/understanding/tracing_and_debugging/tracing_and_debugging.html#basic-logging
I think OpenAI call fails
uit's trying to retry
Do you have logging enabled?

Plain Text
logging.basicConfig(stream=sys.stdout, level=logging.DEBUG)
logging.getLogger().addHandler(logging.StreamHandler(stream=sys.stdout))


https://docs.llamaindex.ai/en/stable/understanding/tracing_and_debugging/tracing_and_debugging.html#basic-logging
httpcore.ConnectTimeout: timed out
so interestingly langchain works fine with this setup
I assume there is a networking issue - but hard to debug OpenAI calls ( i assume it's using OpenAI under the hood for embeddings calling)
Using llama-index==0.9.48 and openai==1.1.0
You can try testing the embed model directly

Plain Text
embed_model = OpenAIEmbedding()
embed = embed_model.get_text_embedding("test")
How are you setting up your embeddings? Or are you leaving it to defaults?
Does Base url works in LLAMA index?
i use this variable for my langchain
You should explcitly set the api base

OpenAI(...., api_base="...")
Add a reply
Sign up and join the conversation on Discord