VSA Backup Operations - CommCell® Console

Collect File Details vs. Live File Recovery

In version 11, there are two available methods to recover files within a virtual machine that was protected using VSA backups: the traditional file recovery or the Live File Recovery, also referred to as the block-level browse or Live Browse.

There are differences between the two methods:

• File Recovery – indexes every virtual machine during the backup, which increases backup time but provides quicker browse and restore.
• Live File Recovery – doesn't require indexing during backup, which improves backup performance, but slows browse operations since the virtual machine must be mounted from the disk library.

For file servers where browse and recovery operations are frequent, consider using traditional recovery mode, and for all other virtual machines use the default Live File Recovery.

The 'Collect File Details' option dictates which method is used. If it is checked, it uses traditional file recovery. When unchecked, it uses Live File Recovery. The benefit of collecting file details is to conduct faster browse operations since the indexes already exist. However, generating indexes on hundreds of VMs at backup time will have a negative impact on performance.

For Linux virtual machines, the traditional file recovery only supports EXT2 and EXT3 file systems. If the VM is using EXT4, JFS, XFS or any other advanced file systems, in such cases the block-level browse is the recommended method.

Note! This applies to subclient using V1 indexing only.

Backup Failed VMs Only

From the incremental backup job or scheduled job | Click Advanced | Data tab

The option to backup failed VMs only can be used to create special schedules or run special backup jobs to protect VMs that were skipped during regular VSA backups. Let's take a subclient that has 100 VMs defined, for which the last backup job failed two VMs. Running another backup job for the subclient would back up the entirety of the 100 VMs. With the 'Backup Failed VMs Only' option enabled, the system looks for failed VMs in the previous subclient backup job and only protects these machines.