Does DataSpell’s Local History preserve Jupyter kernel state after notebook restoration?

Question

Goal
To understand whether DataSpell’s Local History feature can reliably restore a Jupyter notebook to a runnable state, including re‑establishing the kernel connection and any associated environment settings.
Current Behavior
Local History records changes to notebook files (.ipynb) and allows a user to revert to any prior snapshot via the IDE’s context menu. After restoring a notebook, the kernel connection is lost; the user must manually reconnect and re‑execute cells to return to a runnable state. Database connection configurations are not backed up, and there is no public API to trigger a revert programmatically.
Unresolved Questions

Is there a built‑in or future‑planned mechanism that automatically reconnects the Jupyter kernel or restores the execution environment when a notebook is reverted from Local History?
Can the snapshot retention policy be configured to keep all historical versions regardless of the configured size limit, ensuring critical changes are never purged?
Does DataSpell expose any API or command‑line option to programmatically trigger a Local History revert, enabling automated backup verification workflows?

Tasadduq Burney · Answer

Answer
DataSpell’s Local History does **not** preserve the Jupyter kernel state or any associated execution environment when a notebook is reverted to an earlier snapshot.
Confirmed facts

Local History records only changes to the notebook file (.ipynb) content.
It does not store IDE runtime data such as kernel connections, environment variables, or executed cell outputs.
After a revert, the kernel indicator shows “No kernel” and the user must manually reconnect and re‑run cells to regain a runnable state.

Likely explanation
The feature is designed as a file‑level version control; snapshotting the live kernel would require capturing process state, which is outside its scope and would introduce significant complexity and potential security concerns.
Steps to verify

Open a notebook in DataSpell, attach a kernel, and execute a few cells to produce output.
Open Local History → Show History, select an earlier snapshot, and choose Revert.
Observe that the kernel status changes to disconnected (“No kernel”).
Re‑connect the kernel via the kernel selector and re‑execute the previously run cells to confirm that outputs are regenerated.

Does DataSpell’s Local History preserve Jupyter kernel state after notebook restoration?

Goal

Current Behavior

Unresolved Questions

1 answer

Answer

Confirmed facts

Likely explanation

Steps to verify

Your perspective belongs here.

0 question comments