I get an error message when I view my app

Community Cloud
1

I get this error message when I view my app. I deployed it a few months ago with no problems but when I viewed it again, it had an error. I’ve tried deleting and redeploying but to no avail.
Screenshot 2024-10-28 103448

  1. https://sentiment-analyser-ai.streamlit.app/
  2. GitHub - Froxzen/Sentiment-Analyser: An application which uses NLTK's VADER model to classify text as positive, negative or neutral
  3. Oh no.
    Error running app. If you need help, try the Streamlit Docs and forums
  4. Streamlit 1.38.0
  5. Python 3.12.4
2

Generally, this is in indication that an app has gone over its resource limits. In this case, I’d recommend rebooting the app. I know you said you deleted it and redeployed it, but can you try rebooting and watching the logs as it’s rebooting? (You’ll need to reboot it from your dashboard since apps in this state don’t have the Cloud overlay in their lower-right corner.) Just reboot it from the dashboard and immediately view it. You should see it switch to “In the oven” in a few seconds so you can watch the logs.

If it goes to “Oh no” before it gets up an running, I’d be curious if it’s got more information that it’s reading in now vs when you created it, or some other reason it might be holding more in memory now.

3

It’s been like this for an hour. I’m not sure what this error means.

Screenshot 2024-10-28 123658
Screenshot 2024-10-28 1236581909×855 112 KB

4

There is a little problem with Streamlit versions <= 1.19 as it relates to Altair. Streamlit didn’t pin the version of Altair (for Streamlit <=1.19) but needs Altair version <5. As such, Community Cloud tries to address this, but it basically resinstalls Altair after Streamlit, which can change all the versions of Altair’s dependencies in the process.

Since you’ve pinned streamlit==1.14.0, try also pinning altair<5. The error looks a little different than I’m used to seeing it, though, so not sure if there’s something else going, too. If you can use a newer version of Streamlit, that should unlock some of the pip restrictions that happen.

5

This topic was automatically closed 2 days after the last reply. New replies are no longer allowed.