Recovering the virtual machine on the disaster recovery (DR) site or VMware vSphere Replication (VR) fails with the error: unable to create a failover image for group GID 66a70XXXXXXXX on VR server

Products

• VMware Live Recovery
• VMware vSphere ESXi

Issue/Introduction

Symptoms:

• Cannot recover the virtual machine on the disaster recovery (DR) site.
• Recovering the virtual machine on the DR site or VMware vSphere Replication (VR) proceeds up to 20% and then fails
• You see this error:
  
  unable to create a failover image for group GID 66a70XXXXXXXX on VR server
   

• In the hbrsrv.log files, you see entries similar to:
  
  <YYYY-MM-DD><time> [7FF0C0760700 info 'StorageManager' opID=f9d3f90b-####-####-####-########b2e6-HMS-12502896:hs-9d4f:hs-6f49] RemoteFile: Opened path (/vmfs/volumes/533fd3b2-####-####-####-########06f/SiteCoreQADB/hbrcfg.GID-66a70e8c-####-####-####-########04f.99130.nvram.302574)</time>

  <YYYY-MM-DD><time> [7FF0C0760700 error 'Main' opID=f9d3f90b-####-####-####-########2e6-HMS-12502896] [0] Config file hbrcfg.GID-66a70e8c-####-####-####-########04f.99130.nvram.302574 size is 74232, maximum supported size is 65536</time>

  Note: By default, the VR server log files (hbrsrv.log) are located at /var/log/vmware. in the VR appliance.

Cause

By default, the maximum supported size of the nvram file in the hbrconfig file is 6 GB. This issue occurs if the nvram file size exceeds 6 GB for the replicated virtual machine.

Resolution

maxConfigFileSize value 65535 with 81920.
 

Note: Ensure to take a backup of the hbrconfig file at /etc/vmware/ by running this command:

cp hbrsrv.xml hbrsrv.xml.bak
 

To replace the maxConfigFileSize value 65535 with 81920:

1. Log in to the HBR virtual machine.

2. Open the file /etc/vmware/hbrsrv.xml using the text editor. For more information, see Editing files on an ESX host using vi or nano (1020302).

3. Locate this entry and remove <!-- and --> before and after <maxConfigFileSize>

4. Replace the maxConfigFileSize value 65535 with 81920.

5. Restart the HBRSRV service using this command:
   
   /etc/init.d/hbrsrv restart

6. Re-run the failover operation.

Additional Information

Open a new SR with GSS support quoting this KB if the issue is not resolved. This is general information that has worked in the past and has been validated by Engineering.