Skip to main content

​
Before restoring

A snapshot belongs to a workspace and captures its root disk. It does not rewind persistent volumes, application data stored on those volumes, or running processes. Keep application-level backups for databases. Restoring replaces the current root disk. Do not create a replacement snapshot just to preserve the current state unless you intend to replace the existing checkpoint: this implementation retains one snapshot per workspace.

​
Restore and verify

Run locally using IDs from the list command:
Check workspace status, app health, and mounted data. Older application code may not understand a database schema that changed after the snapshot; restore compatible application data separately when needed. If the restore fails, preserve the error and inspect storage troubleshooting. Do not remove the workspace or its volumes to retry.