What is the issue?

In some situations a guest could fail, potentially causing corruption.

Hyper-V machines running BA 10.1 with CSG enabled create a temporary checkpoint to allow scanning of the guests. Sometimes the checkpoint process can stall, and when this happens CryptoSafeGuard thinks the checkpoint has failed. The checkpoint could still eventually be created, but it won’t be removed even after the scanning phase has completed.

Guests with Hyper-V checkpoints run on separate differencing disks, and over time can come to use a significant amount of disk space. If the above problem occurs in every backup, the amount of “incremental” changes in each differencing disk could result in the host volumes becoming filled up. If this happens, guest integrity is at risk, as files being written by guests that are hosted on those volumes could fail, potentially causing corruption.

Who is affected?

The problem can occur under the following conditions:

1. Running BackupAssist v10.1.0, v10.1.0r1, or v10.1.0r2
2. CryptoSafeGuard is enabled
3. One or more jobs are scheduled with Hyper-V guest selections
4. Selected guests take more than 30 seconds to checkpoint

What is the resolution?

The problem can occur under the following conditions:

1. Log onto any affected Hyper-V machines
2. Review all guest checkpoints using Hyper-V Manager
3. Delete any unwanted checkpoints: To delete the check point, select the VM in Hyper-V Manager and then right click the checkpoint as shown below. The option Delete Checkpoint will appear as an option
   
4. Upgrade the BackupAssist install(s) to BackupAssist v10.1.1

Further questions?

If you have any other questions, you can contact our support team.