In Veeam, a media set is a logical group of tapes that are treated as a unit for writing backup data to physical media. This grouping lets you partition the archive into distinct groups, so you can manage retention, rotation, and restoration more effectively. By assigning backups to separate media sets, you can keep different data families (for example, production vs. long-term archive) separate and easier to catalog. It’s not meant to be a single fixed group for all backups, nor is it a rotation schedule or a VTL driver.

Excluding deleted blocks in this context relies on a change-tracking mechanism built into the guest OS file system. Windows NTFS provides the USN Journal, a change journal that records file and block-level modifications, including deletions. Veeam uses that journal to identify which blocks were deleted since the last backup so those blocks can be omitted from the next incremental pass, making backups smaller and faster. FAT32 and exFAT do not implement this kind of change journal, so they cannot support this optimization. ext4 is a Linux file system with its own journaling, but the specific ability to exclude deleted blocks in this Windows-oriented change-tracking workflow is tied to NTFS and its USN Journal. Thus NTFS is the required file system.

When CBT tracking data becomes corrupted, the way to fix it is to reset the tracking state for the affected VM. Reset-HvVmChangeTracking clears the current Hyper-V Change Tracking data for a virtual machine and prompts Hyper-V to rebuild the CBT state on the next backup, giving you a clean baseline and accurate detection of changed blocks going forward. The other options don’t perform this reset: they either don’t exist as a CBT-reset command, would only restart a service without clearing the CBT state, or would change configuration rather than reinitializing the tracked data.

Compression settings are part of a backup job’s configuration, so they’re used when the job starts its next run. When you update the setting, Veeam saves it to the job, and the next time that job begins, it uses the new compression level. If a backup is already running, it continues with the settings it started with, and the new setting won't affect that in‑progress run. No reboot or downtime is required for the change. In short, the new compression level takes effect at the next run of that job.