Backup from storage snapshots is the correct choice because it has the lowest performance impact on the production VM. It leverages storage hardware snapshots, allowing the backup to be created almost instantly from the snapshot, minimizing I/O load on the production datastore compared to methods that read directly from the VM’s virtual disks during production hours.

Enabling Changed Block Tracking (CBT) is the correct answer. CBT allows the hypervisor to track which blocks on a VM’s virtual disk have changed since a specific point in time. When CBT is enabled, Veeam Backup & Replication queries this information and only reads the changed blocks, dramatically reducing the amount of data processed during incremental backups and shortening the backup window.

Configuration Backup is the correct answer. This feature is designed to back up the Veeam Backup & Replication configuration database, which includes the configuration of all managed servers (like backup proxies and repositories), jobs, settings, and credentials. It is essential for disaster recovery of the backup infrastructure itself.

Configuring immutability on the cloud object storage bucket is correct. The 3-2-1-1-0 rule calls for three copies of data, on two different media, with one copy off-site, one copy immutable, and zero errors. While a hardened Linux repository provides immutability on-premises, using immutable object storage (like Amazon S3 Object Lock or Azure Blob Immutable Storage) for the cloud copy satisfies the requirement for an immutable, off-site copy.

Deploying a Veeam Backup & Replication instance in a standalone configuration is the most efficient. This creates a fully independent backup server at the branch office. It manages local backups and can be configured to replicate VMs to a backup server in the main data center, sending only the compressed and deduplicated replication traffic over the WAN link, which conserves bandwidth.

The ‘Perform authoritative restore of Active Directory’ option is correct. This specialized option, available when restoring a domain controller VM, increments the Update Sequence Number (USN) of the restored Active Directory objects. This marks the restored data as the most recent, forcing other domain controllers in the forest to replicate from it, thereby avoiding conflicts and ensuring the restored DC functions correctly.

The key advantage is that it shifts the I/O-intensive merge process to a separate synthetic full activity. Forward incremental backups simply create a new incremental file (.vib). The synthetic full operation, which creates a new full backup file by synthesizing it from the previous full and subsequent increments, can be scheduled for off-peak hours. This decouples the heavy write load from the daily backup window, improving performance during production hours.

Using the guest OS file index to perform a File-Level Restore is the most efficient method. When application-aware processing is enabled for Linux VMs (using Veeam Linux FLR helpers), Veeam indexes the file system within the backup. This allows you to browse the backup contents for a specific point in time (from three days ago) and restore individual files or directories directly to the original VM or an alternate location, without any unnecessary recovery steps.

Configuration Backup Encryption is the correct answer. This feature allows you to encrypt the Veeam configuration backup file with a separate password. The configuration backup contains, among other things, the encryption passwords for all your backup jobs. By storing this encrypted configuration backup file in a secure location (like a safe), you ensure that even if all job passwords are forgotten, you can restore the configuration backup (using its password, which is stored separately per company policy) to retrieve the original job encryption passwords.

Backup from storage snapshots is the correct choice because it has the lowest performance impact on the production VM. It leverages storage hardware snapshots, allowing the backup to be created almost instantly from the snapshot, minimizing I/O load on the production datastore compared to methods that read directly from the VM’s virtual disks during production hours.0