Why I am not able to perform failback from AWS to on-premise using failback client iso?

Error Message 1: Failed to get recovery instance volumes

Error:

Failed to get recovery instance volumes, please check the network configuration of your recovery instance

Root causes

The error typically occurs due to:

• Network connectivity failure between failback client and recovery instance on port 1500
• AWS Replication Agent service issues on the recovery instance
• Missing AWSElasticDisasterRecoveryRecoveryInstancePolicy on IAM role attached to the recovery instance

To resolve this issue

1. Verify Network Connectivity

2. Ensure the failback client can communicate with the recovery instance on TCP port 1500

3. Confirm security group configuration allows inbound traffic on port 1500

4. Test connectivity:

   For Windows:

   Test-NetConnection <recovery_instance_IP> -Port 1500
   

   For Linux:

   telnet <recovery_instance_IP> 1500
   

Note
Connection can be established through private routes (VPN/Direct Connect) or public routes (if public IP exists).

2. Check AWS Replication Agent Status

For Windows:

Get-Service | Where-Object Name -like "*AWSR*"

For Linux:

systemctl list-units | grep AWSR

Verify AWSReplicationService shows "Running" status

3. Validate IAM Role Configuration
   1. Navigate to AWS IAM console
   2. Locate the recovery instance IAM role
   3. Confirm AWSElasticDisasterRecoveryRecoveryInstancePolicy is attached

Error Message 2: Could not associate failback client

Error:

Could not associate failback client to recovery instances

Root causes

• Incorrect AWS credentials in failback client
• Missing required IAM permissions

To resolve this issue

1. Verify AWS Credentials

   • Confirm AWS Access Key ID and Secret Access Key
   • Test credentials:

     aws sts get-caller-identity
     

2. Validate IAM Permissions

   • Check IAM console
   • Verify AWSElasticDisasterRecoveryFailbackInstallationPolicy attachment

Error Message 3: Failback client not seen

Error:

Failback client not seen

Root causes

• Interrupted communication between Failback Client and DRS endpoint
• Terminated failback client process due to closed terminal sessions
• Network configuration issues

To resolve this issue

1. Verify DRS Endpoint Connectivity

   telnet drs.<region>.amazonaws.com 443
   

   Example:

   telnet drs.us-east-1.amazonaws.com 443
   

2. Maintain Failback Client Process

   • Ensure Failback Client runs in foreground with stdout file descriptor
   • Use interactive console session or screen tools to maintain active session

Error Message 4: Recovery instance connectivity verification failure

Error:

Could not verify recovery instance connectivity to DRS

To resolve this issue

1. Verify AWS Replication Agent Status

For Windows:

Get-Service | Where-Object Name -like "*AWSR*"

For Linux:

systemctl list-units | grep AWSR

Verify AWSReplicationService shows "Running" status

2. Verify Network Connectivity Path

   Option A: Public Internet Access Path

   • Check EC2 console for:
     • Public IP assignment
     • Auto-assigned public IP status
     • Elastic IP attachment

   Option B: Private VPC Interface Endpoint Path

   • Verify in VPC console > Endpoints:

     • DRS endpoint availability

     • Correct endpoint service name (com.amazonaws.<region>.drs)

     • "Enable Private DNS" option status

   Important
   Private DNS enablement is crucial for AWS service DNS name resolution

3. Verify Outbound Connectivity

4. Check security group rules for port 443

5. Test connectivity:

   For Windows:

   Test-NetConnection -ComputerName drs.<region>.amazonaws.com -Port 443
   

   For Linux:

   telnet drs.<region>.amazonaws.com 443
   

Error Message 5: Failback Drive Mapping Failure

Root causes

• Outdated Failback Client version
• Complex disk configurations
• Incompatible volume sizes
• Physical disk detection issues

To resolve this issue

1. Verify Failback Client Version

   • Check AWS DRS Failback Release Notes
   • Update to latest version if needed

2. Manual Volume Mapping

   Option A: Direct Block Device Mapping

   /dev/sdg
   

   Option B: Device Mapping Argument

   --device-mapping "ALL=/dev/nvme2n1=/dev/sda,/dev/nvme0n1=EXCLUDE"
   

   To exclude volumes:

   /dev/nvme0n1=EXCLUDE
   

   Note
   For detailed device mapping information, see AWS DRS Failback documentation

Error Message 6: Slow Failback Replication

Description

Failback replication speed is slower compared to initial replication, primarily due to EBS volume initialization latency when reading new blocks from Amazon S3.

To resolve this issue

1. Initialize EBS Volumes
   • Manually initialize volumes to prevent first-touch penalty
   • Consider using IO1/IO2 (Provisioned IOPS SSD) volumes

Note
Higher performance volume types will increase costs. Consider your cost-performance requirements when choosing volume types.