Notebooks in Workspaces limitations¶

Notebook services and runtime¶

• Notebook services are subject to an account limit of 200 active services.

• Notebooks in different workspaces cannot share a service.

• Notebooks in the same workspace connect to a shared service by default.

• Users may create multiple services within a workspace and assign notebooks to different services as needed.

• Notebook services may be restarted over the weekend for container service maintenance. After a restart, you must rerun notebooks and reinstall any packages to restore variables and packages.

• Custom container images and the artifact repository are not yet supported in Notebooks in Workspaces.

• Package installation and listing behavior differs between uv and standard pip. Snowflake supports installing packages using uv pip install, and uv pip freeze lists only packages installed using uv pip install. To view a complete list of all packages, including those in the base image and those installed with standard pip install, use pip freeze.

• Communicating with external stages is not supported.

Using notebooks in Workspaces¶

• Renaming notebook files, folders, or the workspace can cause unexpected behavior, including service disconnection, clearing the notebook’s output cache, or delays in updating referenced files.

• If you are disconnected, try reconnecting the notebook. If you renamed the workspace, create and use a new service.

• Secondary roles are no longer required to use personal workspaces.

  • If secondary roles aren’t set to ALL, the user must select a role with OWNERSHIP or USAGE on compute pools and EAIs to create a service. Other personal workspace features, such as SQL files and Git integration, remain available.

  • If account session policies block the use of secondary roles, notebooks cannot run in personal workspaces.

Editing and running notebooks¶

• iPywidgets are not yet supported.

• Embedding images in Markdown cells or using remote images via URLs is not yet supported.

  To embed an image in your notebook, upload it to your workspace and display it using a Python cell, for example:

  from IPython.display import Image, display
  display(Image(filename="path/to/example_image.png"))
  

  Copy

  For a cleaner presentation, you can collapse the code cell to show only the image result.

• Cell outputs are not saved to the .ipynb file and cannot be exported.

• SQL cells cannot run EXECUTE NOTEBOOK PROJECT (non-interactive execution). To chain notebooks, use Jupyter magic such as %run, which executes another notebook in the same Python process.

  Currently, the notebooks invoked by %run can only support pure Python processes.

• Interactive data grids do not render in Notebooks in Workspaces.

• Cell referencing is not supported.

• SQL cells can be converted to Python cells that start with %%sql.

• Queuing cells is not supported, except when Run all executes all the cells in your notebook sequentially.

Using the original Snowflake Notebooks in Workspaces¶

If you have an original Snowflake Notebook and would like to use it in Workspaces, you must first export the .ipynb file and any supporting files, then upload them to your workspace.

If you have questions about availability timelines for specific features, ask your account representative to contact the Notebooks product team.