When you set up disaster recovery for Azure Virtual Machines (VMs), the VMs replicate from one region to another. If an outage occurs in the primary region, you fail Azure VMs over to the secondary region. When the primary region is up and running again, you fail VMs back over from secondary to primary, and reprotect them so that they start replicating again to secondary.
With this update, Site Recovery now cleans up the deallocated (failed back) VMs and NICs. In addition, if you disable replication after failback, Site Recovery cleans up the disks in the secondary region, in addition to the VMs and NICs.
Learn more about failover and failback for Azure Virtual Machines.