Why is AWS Backup unable to back up my FSx for ONTAP volume?

Last updated: 2022-11-21

When backing up my Amazon FSx for NetApp ONTAP volume using AWS Backup, I'm receiving an error or the back-up progress is slow.

Short description

Backups commonly fail if there is insufficient space remaining on the file system. Even if you have space available on the volume, the volume is thinly provisioned. This means that the volume only consumes storage capacity from your file system for the data stored in the volume. So, even if you have space available on the volume, you might not have available space on the file system SSD (aggregate : aggr1).

Before creating a backup of an FSx for ONTAP volume, Amazon FSx first verifies that the volume isn't full. Full volumes can't be backed up because FSx for ONTAP requires a small amount of free space to create a snapshot. FSx for ONTAP considers a volume full if its utilization is above the volume's fullness threshold. By default, the fullness threshold is 98%.

In addition, when reporting free space for a volume, FSx for ONTAP considers the smaller of the following:

  • The available capacity in the volume.
  • The available capacity in the file system's SSD tier.

For example, if you create a 100-TB SSD file system, FSx for ONTAP reports that volume as having 1 TB of free capacity. So, FSx for ONTAP reports the volume as 99% full. 99% exceeds the volume's default fullness threshold of 98%. The backup of your volume might fail because your volume is reporting that it's more than 98% full.

Resolution

Run the describe-backups AWS Command Line Interface (AWS CLI) command to view any failure messages for the backup. In the following example command, replace xyzabc with the ID of your backup.

$ aws fsx describe-backups --backup-id backup-xyzabc --region region_id

Note: If you receive errors when running AWS CLI commands, make sure that you're using the most recent version of the AWS CLI.

After verifying that the failure is due to lack of storage on the volume, do the following:

Note: Keep in mind that the backup created by AWS Backup is stored in the .snapshot directory on the volume itself. This means that the backup consumes space on the volume.

Increase the space-full-threshold-percent to 100%

FSx for ONTAP volumes are thinly provisioned. So, you can provision a volume with more size than the file system. This might cause space issues when creating snapshots. For example, if you create a 100-TB volume in a 1-TB file system, you can create snapshots using the NetApp ONTAP CLI. But, you can't create snapshots with AWS Backup. For a 1-TB file system, you can create backups of a volume that's about 42 TB. Any additional TB in size to the volume causes AWS Backup to fail to create a backup.

To correct this, run the following command using the NetApp ONTAP CLI:

::> vol modify -vserver <svm_name> -volume <vol_name> -space-full-threshold-percent 100

The preceding command increases the full threshold of your volume. This prevents FSx for ONTAP from reporting your volume as full, allowing backups to proceed.

Increase the file system's SSD storage capacity

Increase your file system's SSD storage capacity so that free SSD capacity is always at least 2% larger than the size of your largest volume. For example, if your largest volume is 100 TB in size, size your SSD capacity to always have at least 2 TB of free capacity. By doing this, even if you keep your volumes' space-full-threshold-percent at the default value of 98%, FSx for ONTAP doesn't report your volumes as full.

Check the snapshot autodelete initiation setting on the volume

If your volume has snapshot autodelete turned on with the initiation set as snap_reserve, your backup might fail. The snap_reserve parameter deletes snapshots when the space reserved for the snapshots reaches the value assigned in Space Reserved for Snapshot Copies. The default value is 5%. When a backup is created, Amazon FSx takes a snapshot of the volume, then compares this snapshot to the previous backup snapshot, if one exists. Amazon FSx then copies only the changed data into the backup. When a previous snapshot has a larger size than the value assigned in Space Reserved for Snapshot Copies, the new backup automatically deletes at creation. This means that Amazon FSx can't create backups of the volume.

To avoid this, increase the size of the volume so that the Space Reserved for Snapshot Copies parameter exceeds the size of the previous backup. To set the snapshot reserve to a value larger than the size of the previous backup, run the following NetApp ONTAP CLI command:

::> volume modify -vserver com -volume vol_name -percent-snapshot -space 20

The preceding command sets the Space Reserved for Snapshot Copies option to 20%. For more information, see Manage the Snapshot copy reserve NetApp documentation.