dosubot[bot]
commented
5 months ago 🤖
Hey there @jkndrkn! 👋 Great to see you diving into the depths of LangChain again. Looks like you've hit a snag, but don't worry, we'll get to the bottom of it together!
The error you're encountering, NameResolutionError, is typically related to DNS resolution issues. This error suggests that the Python interpreter is unable to resolve the hostname answers-dev-jde-test-REDACTED.svc.us-east-1-aws.pinecone.io to an IP address.
Here are a few things you can try to resolve this issue:
-
Check your internet connection: Ensure that your internet connection is stable and you're able to access other websites or services.
-
Check DNS settings: If you're on a corporate network, there might be specific DNS settings or proxies that you need to use. You might need to consult with your network administrator for this.
-
Try a different network: If possible, try to run your code on a different network. This can help determine if the issue is related to your current network's configuration.
-
Check the Pinecone service status: It's also possible that the Pinecone service is temporarily unavailable. You can check the status of Pinecone's services or reach out to their support for more information.
-
Update your dependencies: Although you're using the latest version of LangChain, you might want to check if there are updates available for other dependencies, especially those related to network connections and HTTP requests (like
requestsorurllib3).
Unfortunately, I can't access the repository you've linked to, so I can't review your code or the full error traceback. If none of the above suggestions work, could you please provide more details about your environment? Specifically, information about your operating system, Python version, and network configuration could be helpful.
Also, as mentioned in the context, the langchain_community.vectorstores.Pinecone class is deprecated. It's recommended to use langchain_pinecone.Pinecone instead. If possible, try updating your code to use the new class and see if the issue persists.
For more information about the langchain_pinecone.Pinecone class, you can refer to the LangChain documentation. If you need help updating your code, feel free to ask.
I hope this helps! Let me know if you have any other questions or if there's anything else I can do to assist you.
Sources
- libs/community/langchain_community/vectorstores/pinecone.py
- libs/core/langchain_core/vectorstores.py
jkndrkn
efriis
Checked other resources
Example Code
The following code is copied from a small repository I created that helps reproduce the issue.
Please see the project README.md for instructions for how to configure and run this code.
Error Message and Stack Trace (if applicable)
Description
I am trying to use from
langchain_community.vectorstores.Pineconeto upsert embeddings to a Pinecone index usingPinecone.from_documents(). When I start a script that callsfrom_documents()once for each document it will succeed for the first few runs but then fail and return this error:This is an unexpected error. The index does exist. I have tried this with many different Pinecone indexes of both 1024 dimensions with Cohere embeddings and 768 dimensions with HuggingFace embeddings. I have also tried various document sizes.
System Info
LangChain dependencies
Here is my environment.yml
Here is the entire output of
pip freeze: