Resolving DR Test Failures in Virtual Machines for RDM

Products

• VMware Live Recovery

Issue/Introduction

Symptoms 

• During a disaster recovery (DR) test using Array-Based/vSphere Replication with a newly created Protection Group and Recovery Plan, failures were encountered.
• Specifically, the DR test failed for Raw Device Mapping (RDM) virtual machines, resulting in the following error:

"Failed to recover RDM. Virtual machine xxxx.vmdk cannot be found on recovered datastore. Failed to recover RDM".

• Consequently, some virtual machines within the Protection Group xxx could not be successfully recovered.

Environment

Site recovery manager version: 8.x
Site recovery manager version: 9.x
vSphere vCenter Server: 7.x
vSphere vCenter Server: 8.x
vSphere ESXi : 7.x
vSphere ESXI : 8.x

Cause

The failure to recover RDM virtual machines during the DR test is typically due to one or more of the following reasons:

1. Misconfiguration of Datastores:
   - The target datastore where the RDM virtual machine is expected to be recovered may not be properly configured or accessible. This can prevent the recovery process from locating the necessary virtual machine files.

2. Incorrect RDM Mapping:
   - The RDM mappings may not be correctly replicated to the recovery site. This can occur if the RDM pointers are not properly aligned with the corresponding virtual machines, resulting in the recovery process being unable to find the RDM disks.

3. Replication Issues:
   - Incomplete or failed replication of the RDM disks can cause the recovery process to fail. This can happen if there are issues with the replication configuration or if the replication process was interrupted.

4. Environmental Factors:
   - Network issues, storage accessibility, or other environmental factors at the recovery site can also impact the ability to recover RDM virtual machines.

Resolution

This kb outlines the underlying issues and provides steps for resolution to ensure a successful disaster recovery test for RDM virtual machines:

1. Verify Datastore Configuration:
   - Ensure that the target datastore for recovery is correctly configured and accessible.
   - Confirm that all necessary LUNs are mapped correctly and visible to the recovery site.

2. Check RDM Mappings:
   - Validate that the RDM mappings are intact and correctly associated with the corresponding virtual machines.
   - Ensure that the RDM pointers are correctly replicated to the recovery site.

3. Test Individual VMs:
   - Perform individual recovery tests for RDM virtual machines to isolate the issue.
   - Check for any specific errors or logs that provide more details on the failure.

5. Consult Documentation:
   - Refer to official documentation for any specific configuration requirements for RDMs in a DR scenario.
   - Follow best practices outlined for DR testing with Array-Based Replication and vSphere Replication.

6. Engage Support:
   - If the issue persists, contact support with detailed logs and error messages.
   - Provide the configuration details of the Protection Group and Recovery Plan for further analysis.

By following these steps, you can address the issues encountered during the DR test and ensure a successful recovery of RDM virtual machines.

Additional Information

For more detailed guidance, please refer to the below documentation.

Site Recovery Manager and Virtual Machines Attached to RDM Disk Devices