Streamlit Cloud : Deployed App crashes without any error after 30-20 minutes

Hello, I’ve been using streamlit for proof of concept purposes. Currently I’m working on an application that creates a pandas dataframe based on some input data that comes from different widgets. When I’m on my test environment (localhost) it works correctly and the session state that stores the df keeps the data. But when I’m sharing the app on streamlit cloud, for some reason after a while it erases all the data inside the dataframe. Could this issue be related with a bad handled cache procedure?


Does anyone knows what is happening?

Without seeing your code, it’s hard to say what the issue might be. The first thing I would check is that you are using the newest version of Streamlit, as we’ve patched a few memory leaks in recent releases.

Beyond that, depending on how much data you are caching, you may be exceeding RAM. Unlike a local machine, running on Streamlit Cloud has strict limits. Meaning, there’s no overflow…if you use all the RAM, the container will crash and restart.