Cloned snapshot. Now what?

[freedombacon]

My Time Machine messed up again (different issue). I cloned the snapshot from when it was working and changed the share to mount that dataset. Is this the right way to use the snapshots to recover? The clone doesn't have the quota settings from the original, and the periodic snapshots are still taking snapshots of the original, now broken, dataset. Also, the size of the clone doesn't seem right. The size looks like the amount of data written to it since making, but I would expect the size to be the same as the original plus the differences.

 

[fracai]

I would usually stop the TimeMachine process on the Mac, make sure the volume isn't mounted on your Mac (maybe even turn off the FreeNAS AFP service), then rollback the dataset to that working snapshot, and restart TimeMachine. Your method should be valid as well.

Regarding the size, it's a clone so the data prior to the clone is shared by both datasets. You also won't be able to delete the original dataset unless you first delete the clone, or "promote" the clone to reverse the direction of dependencies.

 

[freedombacon]

How do I promote a clone? I think that will be ideal for how I'm using this. Rolling back to the working snapshots means TM will need to do its back up again.

 

[fracai]

Rolling back wouldn't reset time machine any further than your clone. Unless you're talking about rolling back at this point and losing the backups you've made recently.

Promoting is done with: "zfs promote dataset"

See also: http://docs.oracle.com/cd/E19253-01/819-5461/gcvfl/index.html

 

[freedombacon]

Thanks! Looks like that will do it.