**Agent stopped due to iteration limit or time limit**

I have a langchain application with a streamlit frontend that runs seamlessly in the local environment. But when I deploy it the response returned by the langchain agent executor is
Agent stopped due to iteration limit or time limit. This is for for the first query/prompt. But I can view the SQL query output in the logs.

Then for the second query it responds with:
FileNotFoundError: [Errno 2] No such file or directory: 'output.csv'. But I see that the current directory actually exists.

I fear like missing something, but unable to figure out the actual cause. Would like to get quick help on the same to deliver it to the client.

Thanks in advance.

What do you mean by โ€œcurrentโ€ here? The csv file is in streamlit-app/output.csv.

Thanks @Goyo , I tried the one as suggested which sorted the earlier issue.
I actually referred to current as the same directory equivalent to './'
And now facing a issue is ChromaDB.

Please visit                     https://docs.trychroma.com/troubleshooting#sqlite to learn how                     to upgrade.

Any help to upgrade is package is appreciated.

Please visit ๐Ÿ” Troubleshooting | Chroma to learn how to upgrade.

I checked this and was looking for a solution specific to streamlit because I couldnโ€™t directly hardcode the version of pysqlite3-binary.

Also, not quite sure how compatible is ChromaDB in Streamlit cloud. Any specific help on these steps could be of great help.

The first suggestion is โ€œInstall the latest version of Python 3.10โ€. Why donโ€™t you just try it and tell us how it goes?

I have updated to Python version 3.10.x as suggested and executed the application. Works well in localhost but fails in Streamlit cloud.
I have attached a video recording showing the same. Hope it helps to understand the issue better.

Thanks in advance.

This topic was automatically closed 180 days after the last reply. New replies are no longer allowed.