did I just inadvertently promote a clone over the current dataset?

[guermantes]

Hi,
I have nightly snapshots of my homeshare. I wanted to retrieve a file from the last shapshot since I botched it in the current dataset, so I cloned the snapshot to a new dataset. I then shared that dataset to be able to fetch the file. But the new share was empty all the time. I could see the files via the GUI shell, but in windows file explorer the share was empty. I changed acls to my user, restarted smb service, but no success. I was in a hurry and started to stress out. So I clicked "Promote dataset" in the GUI for the clone. That did not help, but I managed to recover the file from an rsync backup I should have gone to in the first place.

But now I wonder, by promoting the clone have I messed up my home dataset and home share? Because I have the impression that the 94,5 GB used jumped from "home" to "home-auto...-clone" as a result of my clicking promote, and the few KB jumped from the clone to my home dataset. I should have read up on promoting but I was getting stressed and still am (deadline approaching). The original home share still has the files I am used to seeing and the cloneshare has none, but do the homeshare files actually reside in the clone dataset now?

As you can see I am superconfused. Now the home dataset has a promote option. Is the solution to click it and that way go back to the way things were a few hours ago? And then delete the clone dataset in order to get back to normal? Or should I go to the home snapshot and rollback?

 

[guermantes]

all the home dataset snapshots have now been replaced by the clone dataset.

 

[guermantes]

From the Guide:

"Promote Dataset: only appears on clones. When a clone is promoted, the origin filesystem becomes a clone of the clone making it possible to destroy the filesystem that the clone was created from. Otherwise, a clone cannot be deleted while the origin filesystem exists. "

It seems to me that the last sentence is in opposition to point 3 stated elsewhere in the guide:

"Rollback is a potentially dangerous operation and causes any configured replication tasks to fail as the replication system uses the existing snapshot when doing an incremental backup. To restore the data within a snapshot, the recommended steps are:

1. Clone the desired snapshot.
2. Share the clone with the share type or service running on the FreeNAS® system.
3. After users have recovered the needed data, delete the clone in the Active Pools tab.

This approach does not destroy any on-disk data and has no impact on replication."

 

[guermantes]

Pardon my desperation earlier.
I found an old thread and followed advice there. I simply promoted the original home dataset and the clone became a clone once again and I will delete it once I am certain it is safe to delete without losing my original dataset.

I still don't understand what the status of the original dataset became when I promoted the clone. Neither do I understand the last sentence in the "promote dataset" explantion fron the user guide, quoted above. My understanding was that promoting a clone would detach it from its snapshot so that the clone would remain even if the snapshot was removed. But that thing about the original filesystem becoming a clone is something I cannot really grasp, nor see the point of. Since the new dataset has a different name, all replication and sharing falls apart and has to be redone. And since my original share would have all the files still from a later stage in time, how then is that a clone of something that was at an earlier stage. I am really puzzled by all this.