Customizing the Network Settings When Using AWS Backup to Restore VMware-Based Instances to Amazon EC2

AWS Backup for VMware Overview

AWS Backup support for VMware provides robust capabilities for protecting VMware workloads running in on-premises environments or VMware Cloud on AWS environments. One of the most powerful features is the ability to restore these VMware-based backups directly to Amazon EC2 instances, enabling cloud-based disaster recovery scenarios and workload migration paths to AWS. AWS Backup seamlessly integrates with VMware environments through the AWS Backup Gateway, allowing organizations to:

• Back up VMware virtual machines using native VMware snapshots (including the use of Changed Block Tracking)
• Store backups efficiently in AWS using incremental forever backup chains
• Maintain application consistency for supported operating systems
• Enable hybrid cloud protection strategies
• Facilitate cross-platform recovery to Amazon EC2

When disaster strikes, AWS Backup allows you to restore your VMware-based backups directly to Amazon EC2 instances. This capability eliminates the need for intermediate conversion steps, maintains system integrity during the restoration process, and provides flexibility in target instance configuration.

Customization of Networking During Restore

While restoring VMware backups to EC2, you might need to customize the network configuration to match your AWS networking requirements or preserve the existing IP addressing configuration of the workload as it was deployed on-premises or in VMware Cloud on AWS. AWS Backup supports this through restore metadata, specifically the NetworkInterfaces configuration, in a manner similar manner to how custom metadata would be passed during an Amazon EC2 restore.

Retrieve Existing Recovery Point Restore Metadata

Prior to creating an AWS Backup restore job, it is recommended to first retrieve the metadata already attached to the recovery point, and we can do this using the AWS Backup GetRecoveryPointRestoreMetadata API. Some of these values, notably the disks, hypervisorId, and vmId will be used in the restore job invocations.

Command Input

aws backup get-recovery-point-restore-metadata --backup-vault-name "BACKUP_VAULT_NAME" --recovery-point-arn "arn:aws:backup:REGION:ACCOUNT_ID:recovery-point:RECOVERY_POINT_ID"

Command Output

{
    "BackupVaultArn": "arn:aws:backup:REGION:ACCOUNT_ID:backup-vault:BACKUP_VAULT_NAME",
    "RecoveryPointArn": "arn:aws:backup:REGION:ACCOUNT_ID:recovery-point:RECOVERY_POINT_ID",
    "RestoreMetadata": {
        "aws:backup:request-id": "REQUEST_ID",
        "disks": "[{\"diskId\":\"2000\",\"label\":\"Hard disk 1\"}]",
        "hypervisorId": "hype-HYPERVISOR_ID",
        "vmId": "vm-VM_ID"
    },
    "ResourceType": "VirtualMachine"
}

Specify Primary Private IP Address

AWS Backup provides us the ability to customize the VPC, Subnet, and Security Group configuration of Amazon EC2 instances directly through the AWS Backup console, but what if you needed to specify an IP address within the subnet? The sample provided below uses the AWS Backup StartRestoreJob API to create a restore job on VMware-based recovery point and allows us to pass in the primary private IP address we wish to assign to the launched Amazon EC2 instance.

Create Restore Job Sample

aws backup start-restore-job --cli-input-json '{
    "RecoveryPointArn": "arn:aws:backup:REGION:ACCOUNT_ID:recovery-point:RECOVERY_POINT_ID",
    "Metadata": {
        "disks": "[{\"diskId\":\"2000\",\"label\":\"Hard disk 1\"}]", 
        "hypervisorId": "hype-HYPERVISOR_ID", 
        "vmId": "vm-VM_ID", 
        "RestoreTo": "EC2Instance", 
        "InstanceType": "t2.medium", 
        "InstanceInitiatedShutdownBehavior": "stop", 
        "Tenancy": "default", 
        "EbsOptimized": "false", 
        "NetworkInterfaces": "[{\"associatePublicIpAddress\":false,\"DeviceIndex\":0,\"groups\":[\"sg-SECURITY_GROUP_ID\"],\"privateIpAddresses\":[{\"primary\":true,\"privateIpAddress\":\"PRIMARY_PRIVATE_IPV4\"}],\"subnetId\":\"subnet-SUBNET_ID\"}]"
    },
    "IamRoleArn": "IAM_ROLE_ARN",
    "ResourceType": "VirtualMachine",
    "CopySourceTagsToRestoredResource": false
}'

Specify Existing Elastic Network Interface

Another option we could invoke would be attaching an existing Elastic Network Interface (ENI) to the Amazon EC2 instance during the time of restore. This is a particularly helpful option if you have pre-created and pre-configured ENI's in AWS as part of your backup workflow. Similar to the example above, we would again leverage the NetworkInterfaces configuration to specify an existing NetworkInterfaceId we wish to attach to the Amazon EC2 instance.

Create Restore Job Sample

aws backup start-restore-job --cli-input-json '{
    "RecoveryPointArn": "arn:aws:backup:REGION:ACCOUNT_ID:recovery-point:RECOVERY_POINT_ID",
    "Metadata": {
        "disks": "[{\"diskId\":\"2000\",\"label\":\"Hard disk 1\"}]",
        "hypervisorId": "hype-HYPERVISOR_ID",
        "vmId": "vm-VM_ID",
        "RestoreTo": "EC2Instance",
        "InstanceType": "t2.medium",
        "InstanceInitiatedShutdownBehavior": "stop",
        "Tenancy": "default",
        "EbsOptimized": "false",
        "NetworkInterfaces": "[{\"DeviceIndex\":0,\"NetworkInterfaceId\":\"NETWORK_INTERFACE_ID\"}]"
    },
    "IamRoleArn": "IAM_ROLE_ARN",
    "ResourceType": "VirtualMachine",
    "CopySourceTagsToRestoredResource": false
}'

Summary

In this article, we explored how AWS Backup enables customization of network settings when restoring VMware-based backups to Amazon EC2 instances. We covered several key aspects, including:

• The fundamental capabilities of AWS Backup for VMware, including its support for cross-platform recovery to Amazon EC2
• How to retrieve existing recovery point restore metadata using the GetRecoveryPointRestoreMetadata API
• Two practical approaches for network customization during restore:
  • Specifying a primary private IP address within a subnet
  • Attaching an existing Elastic Network Interface (ENI)

These networking capabilities provide organizations with the flexibility needed to maintain specific IP addressing requirements or leverage pre-configured network interfaces when restoring VMware workloads to AWS. This flexibility is particularly valuable in disaster recovery scenarios where maintaining specific network configurations is critical for application functionality and business continuity.

For more information about AWS Backup's VMware capabilities and restore options, visit the AWS Backup documentation.