[SOLVED] Domain Controller backup job fails

[smartens]

Hi Everyone,

I'm evaluating Veeam Backup and Replication v.8, so admittedly I'm not well versed in this realm. I've encountered a strange error that I cannot seem to solve.

The set up is fairly simple. Two physical Windows Server 2012 Hosts, local storage on each, one VM in HyperV on each physical server. These are domain controllers (I know there is some debate around virtualized domain controllers, but let's leave that for another discussion). Both VMs are running 2008R2. These physical hosts are also a failover cluster with some VMs residing in CSVs. A backup job for a test VM in the CSVs succeeds repeatedly, no issues. The backup jobs for the DCs (on the local storage of each host) fails after one successful run.

After I configured the backup jobs (one for each VM), the first job succeeds for each DC. Great. Any subsequent job fails. The error is that the \\vm\admin$ share is inaccessible. It is not, I can browse to it from the Veeam server and the VM itself. However, the first backup job creates the folder VeeamVssSupport in the \\vm\admin$ share. This folder is inaccessible following the first successful job. Only a reboot will remove it from the share and allow future backup jobs to succeed. The guest processing is running with the default domain admin account. That account is a member of builtin\Administrators. Both the account and the group have Full Control for the \\vm\admin$ share. The default domain admin account is the owner of that folder and subfolders/files. UAC is disabled.

I've gone through Veeam support (Case # 00820248). I've read the related posts on the forum (at least all those I could find). I've Googled it. Veeam is recommending that I open a case with MS, but before I do that, I thought I'd post here and hope for some new knowledge.

Thanks to any and all who might reply.

Steve

[smartens]

UPDATE - just my luck, right after posting it I bumped into something that seems to be the solution I need.

I previously evaluated the Dell KACE unit which left an agent running on these servers that I forgot to uninstall. This agent seems to be responsible for a service called KSWMeterSvc which seems to have latched on to that VeeamVssSupport folder, unsure why. This service seems to be a Software Meter related to the KACE agent. Using Process Explorer, I was able to close those handles and the VeeamVssSupport folder disappeared after that.

After uninstalling the KACE agent (successfully on one, manually on the other) I ran the jobs again and they succeeded. Following the successful run, the VeeamVssSupport folder in \\vm\admin$ was not present and subsequent jobs run as expected. Problem solved.

Hopefully this can be helpful to someone else in the future.