Return
-
Title
Rapid Recovery Core Server Crashes during a HotAdd snapshot process -
Description
Rapid Recovery core servers running on an ESXi based system and using the HotAdd transport mechanism to backup protected machines crash during the backup window. Further investigation will show the core Operating System disk was removed by vmware, rather than the expected HotAdded volume.
The sudden removal of the Core Operating System OS drive will take the server offline.
-
Cause
This is related to a known vmware issue described in their knowledge base article Duplicate VMDK UUIDs -
Resolution
To work around this issue, deploy virtual machines to the same host and then migrate them later. This ensures that each disk has a unique identifier.To identify the virtual machines with the same UUID, run this command at the CLI via SSH to verify the UUID:vmkfstools -J getuuid .vmdkIf you have already identified the virtual machines with the same UUID, run this command to change the ddb UUID:vmkfstools -J setuuid .vmdk