Connecting streamlit with the snowflake

OperationalError: 250001: Could not connect to Snowflake backend after 1 attempt(s).Aborting

I was attempting to build an LLM-powered chatbot called FROSTY. I am following a resource on the official Snowflake website. In that resource, it is mentioned to install Miniconda from the Anaconda website. However, due to our company policies, we do not have access to that website. Therefore, I tried setting up Streamlit using pip commands instead of conda commands, and I was able to run three out of the four codes in the Frosty chatbot application. Only the main chatbot code, Frosty_app.py, is not running, and I am encountering the error mentioned above. I have verified all my credentials, and I have a working OpenAI API key. I have also installed all the required libraries using pip. Please investigate this issue.

Hi @Anand,

There seems to be an issue with the connection to the Snowflake backend.

It’s difficult to provide a definitive solution without specific code and setup details. However, I wanted to check if you’re able to connect to Snowflake outside of the Streamlit app?

You’d need to double-check the Snowflake account name, username, password, warehouse, database, and schema you are using to connect. You can use any Snowflake client or Snowflake’s UI to validate this.

Best,
Charly

Hey @Anand ,

Could you please provide a snippet of your code here please?

Specifically, your snowflake connection snippet. If you’re not aware, Streamlit has docs for connecting to Snowflake here: Connect Streamlit to Snowflake - Streamlit Docs

Make sure you follow all the steps.

Again, please let me know if you have any code snippet to help out with.

Let me know if it works!

This is my snowflake connection screenshot and i have followed all the steps mentioned in the documentation, But still getting that error.

This is the screenshot of the code, While running this code only i am getting the error.

Thanks, @Anand.

Would it be possible for you to share the repo, please?

Thanks,
Charly

Hi @Anand

There’s also a tutorial YouTube video on building this LLM chatbot at:

Hope this helps!

I have followed this video and followed the same guidelines, But for setting up the streamlit using miniconda, I was not able to download the miniconda because of our company policies, I have used pip commands to setup the streamlit. In this vedio if you notice, he will run 4 codes in total, The first 3 are running for me also but the last code frosty_app.py is not running. I am not knowing where it is going wrong.

This is the sample github repository, Mine is almost the same.

And @Charly_Wargnier , If possible can we have a zoom call regarding this. I have tried all the ways but facing the same error.

@Anand, I am getting the same error. Is this error resolved?