If I restore a volume using snapshot will it restores the data tiered to the fabricpool bucket

[edgy nova]

I have a aggregate where the data in the volumes is getting tiered to fabricpool, by mistake the data in the fabricpool got wiped.

Can I restore the all data using volume snapshots from ONTAP?

Or the data tiered will be lost completely?

[timber harness]

I don't think that is possible. Even theoretically, it would only be possible if you still had an aggregate snapshot from before you attached the FabricPool to the aggregate. Since aggregate snapshots are by default disable, and even if they aren't they get rotated rather quickly, I guess you wonj't have any luck with that

[hasty gyro]

Depends on where the snapshots are. They probably got tiered off too.

You can mount the .snapshot directory and look, but I doubt you'd find anything.

[finite harness]

You have to understand that FabricPool is basically only an extension to your current volume/aggr. Some of the blocks are still saved locally and some (older) ones are saved in the bucket of the object storage. It's not a copy. Every block only exists only once, either locally or in the bucket. If a block is in the bucket and that bucket is gone, the block is also gone. There is no other copy of that block locally.

So it doesn't matter if you have FabricPool enabled or not, you still always need backups. It depends on the FabricPool policy but most likely the majority of the blocks of your snapshots are already moved to the bucket (since they are most likely cold already). So check how you backed up these volumes and restore from backup.

FabricPool is only a cost-reduction solution. You move the inactive blocks to lower-cost storage so you can utilize your higher-cost flash storage more efficiently. That's it.
Never disable backups only because you enabled FabricPool. They are still needed.

[edgy nova]

Thanks @timber harness @hasty gyro for your inputs.
The snapshots are not tiered off.
So what I understood is that if a volume is fabricpool enabled and the data tiered off to capacity tier then snapshot will not be an restore option.. does the snapshot only track the data on the performance tier?

[finite harness]

What's the tiering-policy of your volumes?
volume object-store tiering show

"auto" policy also does tier off cold blocks from snapshots:

"The auto tiering policy moves cold user data blocks in both the Snapshot copies and the active file system to the cloud tier.

The default tiering minimum cooling period is 31 days and applies to the entire volume, for both the active file system and the Snapshot copies."

https://docs.netapp.com/us-en/ontap/fabricpool/tiering-policies-concept.html#types-of-fabricpool-tiering-policies

[hasty gyro]

And snaps are not a backup.

[rigid plank]

well, snapshots are a backup, but only local unless mirrored to secondary systems.

[hasty gyro]

backup
/băk′ŭp″/
noun

1. A reserve or substitute.
   __2. (Computers) A copy of a program or file that is stored separately from the original. __
2. Support or backing.

from The American Heritage® Dictionary of the English Language, 5th Edition

[rigid plank]

again, we can dig into the semantics of "separately", but it is logically separate and immutable, so, no.

[finite harness]

What you mean with logically separate?
A backup is simply a physical copy of your data. Snapshots are no physical copies. They are referencing the same physical blocks. If this single block gets corrupted then both your production data and your snapshots are damaged.

[rigid plank]

that can be said of any "backup" copy as well, so still no