Simple and handy btrfs snapshoting tool. Supports unattended snapshots, tracking, restoring, automatic cleanup and more. Backed with SQLite.

See the docs folder.

• Allows you to specify the origin and destination of snapshots at will of the user.
• Track snapshots using SQLite as backend database.
• Easy setup using small templates instead of confusing long files.
• Ability to create snapshots of the volumes you want simply by using different configuration templates.
• Ability to create read-only or read-write snapshots.
• Ability to use the same SQLite database for everything.
• Ability to specify the prefix of the name for the snapshots for better identification.
• Ability to specify a kind identifier to differentiate them in the database. Useful if you plan to have hourly, weekly, monthly or more "kind" of snapshots of the same subvolume(s).
• Ability to specify the maximum number of snapshots to keep for automatic cleanup.
• Supports restoration of snapshots in the original directory or a specific one.
• Supports machine name identification for better tracking when using the same database in multiple machines.
• Nice CLI output to see the status and details of snapshots.

Due to SQLite limitations to handle concurrent database operations, we need to make use of SQLite's busy_timeout, which was initially implemented in ce44bf6. Without it you will get the error: Error: database is locked (code 5) if you try to make concurrent snapshots exactly at the same time. The default timeout is 5 seconds which is more than enough time so that no problem occurs even in the most demanding scenarios.

In summary, busy_timeout is the maximum time that SQLite will retry the failed transaction. Considering that Rusnapshot's database operations are small and shouldn't take long, 5 seconds is a decent time. If you continue to get the mentioned error, try increasing --timeout, remember that the value must be in milliseconds.