Quick question - what exactly does the “Use Snapshot” option do on the TrueCloud Backup Task? I read the documentation and it says the obvious, which is “Select to set the TrueCloud Backup Task to take a snapshot of the dataset before a push.”
My question is, how robust of a backup is this, and what happens if I don’t check that box? Would it fully clone/replicate the entire dataset every time? My guess is “use snapshot” means it stores the data once, and then updates with diffs on each push, i.e. snapshot. But the documentation isn’t clear on this, can someone please clarify? Thanks!
My understanding is that “Use Snapshot” does the following:
Takes a current state snapshot of the dataset before beginning the TrueCloud Backup.
Runs the TrueCloud backup
After the backup completes it deletes the snapshot created at step 1.
The purpose of this is to keep the data consistent/unmodified throughout the backup process. Remember, cloud backups will not be as fast as local ones due to bandwidth limitations in most of the case.
Now; as I believe the confusion may be like, but why we need to select it, isn’t obvious that we want the data unchanged throughout the backup in all cases?
Yes, in most of the normal scenarios that is correct. But if the dataset is an archived copy (which is not going to change mostly) this might not be necessary. Also the creation and auto deletion of the snapshots might create inconsistency if there is another replication task that is referring to the same dataset while TrueCloud backup was in progress (as the snapshot will disappear later at the end of backup).
In short, this checkbox gives us an additional option to let us manage these kind of scenarios with overlapping backups/replications.
Again, this is my understanding based on my own analysis. Anyone who may have additional thoughts please feel free to add more.
