Backup Exec Vmware incremental backup jobs periodically take up much more space than expected when backing up thin provisioned VMs
Problem
Backup Exec incremental backup jobs using Vmware AVVI periodically take up much more space than expected when backing up Windows 2012/2012 R2 thin provisioned VMs. the issue has been observed after upgrading VSphere to 6.7, however the issue could occur on other versions as well
Error Message
Cause
Windows 2012 / 2012 R2 OS schedules defrag to be run once a week.. The command line –c –h –k (for 2012) and –c –h –k –g (for 2012R2) for the defrag task will perform storage optimization and slab consolidation on thin provisioned media as well. The /k option (slab consolidation), which is not exactly the same functionality as option /d "disk defragmentation", appear to be the one that affects the "Change Block Tracking" information returned to the Backup Exec Media Server by the VCenter or ESX servers, since the slab consolidation moves blocks around generating write operations on disk which are registered by the CBT VSphere functionality.
According to reports in the VMware official blogs website (https://blogs.vmware.com/vsphere/2011/09/should-i-defrag-my-guest-os.html) .the following statement can be read:
"Thin Provisioned VMs. If you defragment a Thin Provisioned VM, as file blocks are moved around, the TP VMDK bloats up, consuming much more disk space"
Solution
There are several feasible solutions to this issue:
1) Disable defrag on the affected virtual machines
2) Modify the defrag options in the "Schedules" (Schedule task") Windows Os application and remove the /K option
3) Modify the defragmentation task so that it would be run after the last incremental backup job, right before the Full backup job runs.