Snapshot being taken each time there is a Timemachine backup running

Hi, running TrueNAS Scale and using it for Timeachine backup of MAC´s among other things. I recently discovered that a snapshot is taken every time a timemachine backup is running (as it seems). This mean a snapshot is taken once an hour or more. Additionally, these snapshot is market as “Will not be destroyed automatically” so it seems they will exist forever. Is this really the way it is supposed to work? I thought TimeMachine had an own internal way of taking snapshots and keep data over time or are the TrueNAS snapshots part of this in some way?

Thanks for all help from the large knowledge base out there.

Those snapshots are internally managed by the smbd service. Basically time machine backups are a sparsebundle volume (directory on SMB server) that internally manages snapshots, but those internal snapshots don’t do anything if the whole volume gets corrupted due to network issues, bugs in apple code, or accidental deletion by user. Depending on how share is configured, the smbd service takes a ZFS snapshot after each successful time machine backup, and auto-deletes its snapshots after a certain amount of time.

These snapshots do not appear in any place obvious to me in the web interface. I would love to know more about them.

My goal — I think — is to replicate backups to another off-site server as a second-tier backup for my clients.

It seems best to avoid pushing data which is incomplete — from a Mac’s perspective — to the remote server.

Or maybe, depending on the relationship between these snapshots and regular ones, that doesn’t matter at all.

In any case, it seems an important thing to understand.

I’m running CORE, for whatever that’s worth.

Have you tried looking in the snapshots list?

(Yes, scale pic)

In a word… yes. :smile:

It’s possible none of these snapshots has been created yet, though the Macs have made several backup runs over the last few days.

Or it’s possible that somehow CORE works differently.

1 Like