Protecting your data with snapshots

A snapshot is a read-only image of an Amazon FSx for NetApp ONTAP volume at a point in time. Snapshots offer protection against accidental deletion or modification of files in your volumes. With snapshots, your users can easily view and restore individual files or folders from an earlier snapshot to undo changes, recover deleted content, and compare file versions.

A snapshot contains the data that has changed since the last snapshot which consumes the file system's SSD storage capacity. Snapshots are not included in any volume backups. Snapshots are enabled by default on your volumes using the default snapshot policy. Snapshots are stored in the .snapshot directory at the root of a volume. You can store a maximum of 1,023 snapshots per volume at any point in time. Once you reach this limit, you must delete an existing snapshot before a new snapshot of your volume can be created.

Snapshot policies

The snapshot policy defines how the system creates snapshots for a volume. The policy specifies when to create snapshots, how many copies to retain, and how to name them. There are three built-in snapshot policies for FSx for ONTAP:

By default, every volume is associated with the file system's default snapshot policy. We recommend using this policy for most workloads.

The default policy automatically creates snapshots on the following schedule, with the oldest snapshot copies deleted to make room for newer copies:

The default-1weekly policy works in the same way as the default policy, except that it only retains one snapshot from the weekly schedule.

The none policy doesn't take any snapshots. You can assign this policy to volumes to prevent automatic snapshots from being taken.

You can also create a custom snapshot policy using the ONTAP CLI or REST API. For more information, see Create a Snapshot Policy in the NetApp ONTAP Product Documentation. You can choose a snapshot policy while creating or updating a volume in the Amazon FSx console, the AWS CLI, or the Amazon FSx API. For more information, see Creating volumes and Updating volumes.

Deleting snapshots

Snapshots consume storage capacity only for the data on a volume that has changed since the last snapshot. For this reason, if your workload writes data rapidly, snapshots from old data can take up a significant amount of a volume's storage capacity.

For example, the volume show-space ONTAP CLI command output shows 140 KB of User Data. However, the volume had 9.8 GB of User Data before the user data was deleted. Even if you've deleted the files from your volume, a snapshot might still reference old user data. Because of this, Snapshot Reserve and Snapshot Spill in the prior example take up a total of 9.8 GB of space, even though there is virtually no user data on the volume.

To free up space on volumes, you can delete older snapshots that you no longer need. Because snapshots are incremental, you do not reclaim the amount of storage equal to the size of the snapshot when you delete it. You can see the amount of storage you can reclaim when deleting a snapshot by using the volume snapshot compute-reclaimable -vserver ONTAP ClI command, using your data to replace svm_name, vol_name, and snapshot_name.

fsid8970abc52::> volume snapshot compute-reclaimable -vserver svm_name -volume vol_name -snapshot snapshot_name 
A total of 667648 bytes can be reclaimed.

You can delete snapshots either by creating a snapshot auto-delete policy or by manually deleting snapshots. Deleting a snapshot deletes the changed data stored on the snapshot.

Snapshot reserve

Snapshot copy reserve sets a specific percent of a volume's storage capacity for storing Snapshot copies, with a default value of 5 percent. The Snapshot copy reserve must have sufficient space allocated for the Snapshot copies, including volume backups. If the Snapshot copies exceeds the Snapshot reserve space, you must delete existing Snapshot copies from the active file system to recover the storage capacity for the use of the file system. You can also modify the percent of disk space that is allotted to Snapshot copies.

Whenever Snapshots consume more than 100% of the Snapshot reserve, they begin to occupy primary SSD storage space. This process is called Snapshot spill. When the Snapshots continue to occupy the active file system space, the file system is at risk of becoming full. If the file system becomes full due to Snapshot spill, you can create files only after you delete enough Snapshots.

When enough disk space is available for snapshots in the snapshot reserve, deleting files from the primary SSD tier frees disk space for new files, while the Snapshot copies that reference those files consume only the space in the Snapshot copy reserve.

Because there is no way to prevent Snapshots from consuming disk space greater than the amount reserved for them (the Snapshot reserve), it is important to reserve enough disk space for Snapshots so that the primary SSD tier always has space available to create new files or modify existing ones.

If a snapshot is created when the disks are full, deleting files from the primary SSD tier does not create any free space because all that data is also referenced by the newly created Snapshot. You must delete the Snapshot in order to free up storage in order to create or update any files.

You can modify the amount of Snapshot reserve on a volume using the NetApp ONTAP CLI. For more information, see Updating your volume's snapshot reserve.