What can cause a single VM to take longer to backup?

[dwrice0]

I have a backup job with 30+ VMs. In this job, we have one or two VMs that are not the largest in the job, but they take much longer to backup than any other VM. These VMs are same OS, use the same storage and are backed up by the same proxies, network path and are backed up to the same repository. At the end of the backups, not much data is actually transferred. It just seems to take way longer to process through the VM's data.

I've noticed this across multiple customers but have never been able to find out the common thread.

[DaveWatkins]

Corrupt CBT data that can't be reset automatically is the classic cause. Is there any mention of CBT in the log for those VM's?

[dwrice0]

Bam. That is most likely it. Found this in the job details: "CBT data is invalid, failing over to legacy incremental backup. No action is required, next job run should start using CBT again. If CBT data remains invalid, follow KB1113 to perform CBT reset. Usual cause is power loss." Thanks!

[DaveWatkins]

Also make sure your VM's are a high enough hardware version to support CBT, I've never tried backing up ones that aren't but that might also be the cause