Segmentation issue

Using Streamlit
1

I have a Github python app that is using streamlit to host and today it started giving this error:
How can I fix this please?

2026-07-10 17:39:56.657 Uvicorn server started on :::8501

/app/scripts/run-streamlit.sh: line 9: 194 Segmentation fault sudo -E -u appuser /home/adminuser/venv/bin/streamlit "$@"

2

Hey there, thanks for reaching out and welcome to the Streamlit community! :partying_face: Segmentation faults like the one you’re seeing (Segmentation fault sudo -E -u appuser /home/adminuser/venv/bin/streamlit “$@”) on Streamlit Community Cloud are often caused by your app exceeding memory/resource limits or by incompatibilities with Python/package versions. This is especially likely if your app was working before and suddenly started crashing, even though it still works locally. Community Cloud has strict resource limits (max ~2.7GB RAM, 2 CPU cores), and memory leaks or large models/files can easily push you over the edge. See the official docs for more on resource limits and common app problems.

My understanding is that the best first steps are: (1) Reboot your app from the workspace or Cloud logs (temporary fix), (2) Optimize your app’s memory usage (use st.cache_data/st.cache_resource, move large data to a database, profile memory), and (3) Check your Python and Streamlit versions—sometimes new releases or dependency conflicts can cause segfaults, so try pinning compatible versions in requirements.txt and avoid using very new Python versions (e.g., 3.13+). If you’re still stuck, please share your requirements.txt and a minimal reproducible example so the community can help debug further! For more troubleshooting, see this FAQ post and the sanity checks guide.

Sources:

3

I have the exact same issue since 2026-07-10. Two apps (production + staging) both crashing.

Environment:

  • Streamlit 1.55.0
  • Python 3.14.6 default (crashes)
  • Python 3.12.13 manually set in Settings (also crashes)
  • pyarrow 25.0.0
  • pandas 2.3.3
  • pillow 12.3.0

Crash pattern:

  • App loads and login works
  • Crashes appear on specific user interactions:
    • Dataframe searches (especially large ones)
    • Navigating between project details
    • Changing project selection
  • Multiple use_container_width deprecation warnings right before Segfault
  • No Python traceback

The bot suggests pinning older versions. Could someone from the Streamlit team confirm which specific versions of streamlit, pyarrow, pandas, and pillow are stable together on Community Cloud right now?

We rolled back our code completely (same as production) and the crash persists. This is clearly a platform-level issue affecting apps that were working before 2026-07-10.

Impact: Production ERP for a construction company, team can’t work reliably.

4

This might be caused by a bug in pyarrow 25.0.0: [C++][Python] SIGSEGV in bundled mimalloc mi_thread_init when libarrow is first loaded on a non-main thread that exits (mimalloc 3.3.x, pyarrow 25.0.0) · Issue #50471 · apache/arrow · GitHub
Make sure to update to Streamlit 1.59.2 or pin pyarrow < 25.0.0 to prevent this crash.