Verifying replica/backup integrity

[stevewalker]

Hi all,

I had a problem yesterday when I discovered some helper snapshots on one of my VMs that hadn't been removed correctly. In the end I ended up making a bit of a mess, the snapshot records got deleted eventually but the snapshot files were still present on the disk. The machine's virtual disks were attached to the snapshot files (deltas) rather than the original disks so I wasn't sure what to do next, as there were no snapshots showing when I right clicked on Manage snapshots. In the end, I changed the disk files back to the original VMDKs and booted the machine in independent non persistent mode and everything looked fine and CHKDSKs on the OS disk showed no errors. I shut down the machine and rebooted it without the independent mode set and everything has been fine since (including last night's Veeam backup and replication jobs).

However, I'm sure at some point yesterday some or all of the machinename-CTK.vmdk (changed block tracking?) files were either missing or recreated. Which brings the question, if I assume that all of the records about blocks being changed are stored in these files, how consistent will my backups or replicas be when they are recreated automatically?

I know that there is a check box in the properties of each backup/replication job to 'Verify integrity of ..' but the help file doesn't say specifically when/how often this integrity check occurs - and how it would deal with the circumstances I have detailed below.

Is there a more detailed explanation available somewhere that would help me understand how it works - or even better, how to tell Veeam to verify the replica on block level to make sure that it is consistent.

Cheers all,.
Steve

[Gostev]

Hello Steve, your backups and replicas are fine. If CTK files are gone, the VMware API will let us know that change information is not available, and our product will perform incremental pass using proprietary change tracking mechanism (full scan of source VM to determine blocks changed since last pass). At the same time, we will also ask VMware to enable change tracking back for this VM, so that the next incremental pass is again CBT-enabled.

And here is the existing topic with detailed explanation about integrity checks:
http://www.veeam.com/forums/viewtopic.p ... ity+Checks

Thanks!