A planned failover is a process that allows for the seamless switching from the original VMs at Site A to their replicas at Site B with no data loss. This is the correct course of action when a power outage is expected, as it is a controlled process that ensures that all data is synchronized before the switch is made. This action minimizes data loss and ensures that the VMs on Site B are up-to-date.

The quickest option for migrating a physical Linux file server with multiple volumes to VMware ESXi is to perform Instant VM Recovery. This feature allows you to run the server directly from the backup file in a VMware environment, significantly reducing the time and complexity involved in migration.References:
Veeam Backup & Replication User Guide, Veeam Instant VM Recovery for Physical Servers

To ensure that a backup job includes both the virtual machine's disk files and the Microsoft SQL database transaction logs, Application-aware processing should be used. This feature enables Veeam Backup & Replication to create transactionally consistent backups of VMs running VSS-aware applications like Microsoft SQL Server, ensuring that both the VM disks and the SQL transaction logs are properly backed up.References:Veeam Backup & Replication User Guide, Veeam Application-Aware Processing Guide

Veeam Backup & Replication allows you to set up failover plans to control the order in which VMs are started during a failover. By creating a failover plan, you can specify that the database servers boot before the application servers. This is achieved by setting up VM dependencies within the failover plan, ensuring that the database servers (which are critical for the application servers to function) are operational before the application servers start. Creating a disaster recovery template (B), creating a replica mapping (C), or initiating a planned failover (D) does not directly address the boot order of the servers during failover.