Backup Chain and Retention Policy
For backups of Nutanix AHV VMs, AHV Backup Proxy implements the forever forward incremental and forward incremental backup method. For details on backup methods, see the Backup Methods section of the Veeam Backup & Replication User Guide. Note that for forward incremental backup chains, you can create only active full backups. Synthetic full backups are not supported for AHV VM backups.
During the first backup session, AHV Backup Proxy copies the whole content of the VM and creates a full backup file (VBK) in the repository. The full backup file acts as a starting point of the backup chain. When you add several VMs to a backup job, AHV Backup Proxy creates individual backup chains on the backup repository, one for each VM processed by the job.
During the second and further backup sessions, AHV Backup Proxy copies only those data blocks that have changed since the previous backup session. Changed data blocks are stored as an incremental backup file in the repository. Incremental backup files depend on the full backup file and preceding incremental backup files of the backup chain.
To back up Nutanix AHV VMs, you must configure a backup job in the AHV Backup Proxy web console. For Nutanix AHV VMs, AHV Backup Proxy creates per-VM backup chains: one backup chain contains data for one VM only. Even if you add several VMs to the backup job, AHV Backup Proxy creates several backup chains on the backup repository, one per every VM processed by the job.
One backup job can be used to process one or several VMs. You can instruct AHV Backup Proxy to run jobs automatically by schedule or start them manually.
Backup jobs and backup files created by AHV Backup Proxy are visible in the Veeam Backup & Replication console. However, some of the Veeam Backup & Replication functionality is not available for backups of Nutanix AHV VMs. For details, see Backup Job Limitations.
If you want to create application-consistent backups of AHV VMs, follow the guidelines of Nutanix. For details, see the Application-Consistent Snapshot Guidelines section of the Prism Web Console Guide.
Changed Block Tracking (CBT)
When AHV Backup Proxy performs incremental backup, it needs to know what data blocks have changed since the previous job session. To get the list of changed data blocks, AHV Backup Proxy uses the Nutanix AHV changed block tracking mechanism (CBT). CBT increases the speed and efficiency of incremental backups.
Nutanix AHV CBT uses REST API to determine the changed metadata regions in Nutanix AHV virtual disks. CBT compares two snapshots: a snapshot of the current virtual disks environment with the snapshot that was obtained during the previous backup. Only the changed metadata regions that the API detects are backed up and written to backup repository. After the backup completes, the current snapshot will become the previous snapshot for comparison in the next backup operation. Only one snapshot per VM will be retained for comparison.
Every successful backup job session creates a new restore point that lets you roll back VM data to an earlier point in time. To control the number of restore points in the backup chain, you must specify retention policy settings. The retention policy defines how many restore points you want to retain on disk and thus how ‘far’ you are able to roll back. After the allowed number of restore points is exceeded, AHV Backup Proxy automatically removes the earliest restore point from the backup chain.
To define the retention policy for a backup job, you must specify the necessary number of restore points in the Restore points to keep field in the backup job settings. By default, AHV Backup Proxy keeps 7 restore points. In the Schedule step of the New Job wizard you can select the following units of retention policy.
For details on the retention policy for AHV VM backups, see the Forward Incremental Backup Retention Policy section of the Veeam Backup & Replication User Guide.