Langchain Embedding ConnectionError: HTTPSConnectionPool(host='openaipublic.blob.core.windows.net')

[Faranak001]

I get connection error when running

embeddings = OpenAIEmbeddings(openai_api_key = openai.api_key,
                              openai_api_base = openai.api_base,
                              deployment_name = openai.deployment_name,
                              openai_api_version = openai.api_version,
                              openai_api_type = openai.api_type)
db = DocArrayInMemorySearch.from_documents(
    docs, 
    embeddings
)

We do not use any proxy.
I am trying to run the code on corporate server that does not have access to public internet but has access to Azure Openai.

I am not aware of any primate URL to add based on re suggestion in #13499
How can I get thi working?

[dosubot[bot]]

🤖

Hello @Faranak001,

I'm Dosu, a bot here to assist you with any issues, questions, or contributions related to the LangChain repository. I'm here to provide preliminary support while we wait for a human maintainer. Looking forward to assisting you!

Based on the information you've provided and similar issues in the LangChain repository, it seems like the connection error you're experiencing with LangChain's OpenAIEmbeddings might be due to the network configuration of your corporate server.

Since your server has access to Azure OpenAI but not the public internet, you need to set up the environment variables correctly. The environment variables that need to be set are:

• OPENAI_API_TYPE: This should be set to 'azure'.
• OPENAI_API_BASE: This should be set to your Azure OpenAI endpoint.
• OPENAI_API_KEY: This should be set to your Azure OpenAI key.
• OPENAI_API_VERSION: This should be set to the version of your Azure OpenAI API.
• OPENAI_PROXY: This should be set to your corporate proxy.

Here is an example of how you can set these environment variables:

import os

os.environ["OPENAI_API_TYPE"] = "azure"
os.environ["OPENAI_API_BASE"] = "https://<your-endpoint.openai.azure.com/"
os.environ["OPENAI_API_KEY"] = "your AzureOpenAI key"
os.environ["OPENAI_API_VERSION"] = "2023-05-15"
os.environ["OPENAI_PROXY"] = "http://your-corporate-proxy:8080"

After setting these environment variables, you can initialize the OpenAIEmbeddings class as follows:

from langchain.embeddings.openai import OpenAIEmbeddings
embeddings = OpenAIEmbeddings(
    deployment="your-embeddings-deployment-name",
    model="your-embeddings-model-name",
    openai_api_base="https://your-endpoint.openai.azure.com/",
    openai_api_type="azure",
)
text = "This is a test query."
query_result = embeddings.embed_query(text)

This will ensure that the OpenAIEmbeddings class uses the Azure OpenAI endpoint and the corporate proxy for its requests.

You can find more information about this in the OpenAIEmbeddings class in the LangChain repository.

If you're still experiencing issues after trying these steps, could you please provide more details about the exact error message you're receiving? This will help in further diagnosing the issue.

Sources

• OpenAIEmbeddings fails to connect
• Retrying langchain.embeddings.openai.embed_with_retry
• libs/langchain/langchain/embeddings/openai.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[Faranak001]

@dosu-bot
I set up all the following parameters

deployment,
model,
openai_api_base,
openai_api_type

and confirm they are well defined however I get the save connection error is before

[dosubot[bot]]

🤖

Based on the error message you provided, it seems that the connection error is due to the inability of your corporate server to access the public internet. The error message indicates that the connection to 'openaipublic.blob.core.windows.net' is failing, which is a public internet address.

The OpenAIEmbeddings class in LangChain uses the openai_api_base parameter to set the base URL for API requests. If your corporate server has access to Azure OpenAI but not the public internet, you should ensure that the openai_api_base parameter is set to the correct Azure OpenAI endpoint. This can be done by setting the OPENAI_API_BASE environment variable to the Azure OpenAI endpoint. Here is an example of how to do this:

import os
os.environ["OPENAI_API_BASE"] = "https://<your-endpoint>.openai.azure.com/"

Then, you can initialize the OpenAIEmbeddings class as follows:

embeddings = OpenAIEmbeddings(
    deployment="your-embeddings-deployment-name",
    model="your-embeddings-model-name",
    openai_api_base=os.getenv("OPENAI_API_BASE"),
    openai_api_type="azure",
)

If you've already done this and are still experiencing connection errors, it's possible that there's a network issue on your corporate server that's preventing the connection from being established. In this case, you should contact your network administrator for assistance.

If you're still experiencing issues after checking these things, please provide more information about the specific error message you're seeing so that I can provide more targeted assistance.

For more information, you can refer to the OpenAIEmbeddings class in the LangChain codebase and the similar solved issue in the LangChain repository.

Sources

• OpenAIEmbeddings fails to connect
• libs/langchain/langchain/embeddings/openai.py

---

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.

[mahmoud99-cell]

I have a connection Error in general when calling llmchain @dosu-bot