Backup job failure for one VM, giving a NTDS Writer error

[ninjaburn81]

Note: In reference to open case #02295812

So going back and forth with technical support, and we are all at a bit of a loss on this one. I'm running B&R V9.5 with all recent patches/updates. I have a single VM that fails on the first backup attempt, and yet retries (less than 15 minutes) later almost always are successful. Oh, and the error is NOT daily.

Error is: Processing <VMNAME> Error: VSSControl: -1 Backup job failed. Cannot create a shadow copy of the volumes containing writer's data. Cannot prepare the [NTDS] data to a subsequent restore operation. Cannot process NTDS data. Cannot create a backup copy of the BCD. Cannot get [BcdStore] object. COM error: Code: 0xffffffff

VM details: Windows 2012 R2, running as a AD/DNS/RADIUS box. vSphere 5.5U3. Nothing remarkable about the VM or infrastructure.

I have tried all the normal things such as reboots, tests of the VSS, etc, but no joy on anything. Support also had me try Remove-WBCatalog to see if that would clear any cache issues, but nothing. No odd errors in the server logs from what I can see, and support has parsed the B&R logs but have come up with nothing.

Has anyone ever had an issue like this?? Thanks!

[ninjaburn81]

*crickets*

[foggy]

Hi Adam, have you reviewed this KB article?

[ninjaburn81]

Sorry, never got updates to this thread.

Yes, I did review the KB article before opening a ticket with support. Last thing support came back with is that he found "whispers" of this being a bug in Windows Server that happens randomly, but no known solution. So was more curious if anyone else is seeing this issue. I have two other DCs, but one is a physical box, and the other is sitting in our DR site in a warm state and isn't backed up by Veeam. So at a bit of a loss!

[ayates]

Yep we get this issue with one of our Domain Controllers too, I've done KB1697 and only temporary workaround we have so far is to reboot the server and it will be fine for a few days, maybe even a couple of weeks, then the error will return.
There are no errors in the Event Log of the server to suggest any other issues so not sure what triggers this issue.
The only thing special about this DC is it was migrated from an older vSphere environment before we started backing it up with Veeam, however another DC server that went through the same process never has the issue. Also two other DC that were newly built in another Data Centre also do not have the same issues.
Would be interested to know what triggers it so we can get a permanent fix, however we will need to rebuild at some point as its not optimised for our new storage, so will probably get solved then.