Hyper-v + Netapp, snapshot removal error

[vchris]

Hello,
My support case number is the 00480194

And I have no logs as it's a functional error between Veeam and snapdrive...and that on veeam side I have no error at all.

EDIT: I attached to the case a filtered view of the snapdrive error corresponding to the jobs.

We have a 2 nodes HYPER-V (2012) cluster with a third node (out of cluster) for OFFHOST backup

The 3 nodes are accessing the same luns by iSCSI, the offhost backup work well. We have Host-utilities, DSM and snapdrive (for the hardware vss provider) on each nodes.
For each VMs backup Veeam triggers a snapshot on the Netapp, but it failed to remove it. (no veeam side error)
We have for each vms backuped the following error on the backup server:

EVENT ID: 245
DeleteSnapShot operation failed. Storage System Name = MYCONTROLLERNAME
lunPath = /vol/vol_hyperv_data_iscsi_1/hyperv_data/lun
Snapshot copy Name = {f63c69b6-4f17-41ce-8b7d-42db5af4ae94}
Error code = 0xc00410a6 Error description = Cannot delete backing snapshot as its associated backup snapshot exists.

When I try to remove the snapshot manually I have no issue (from Snapdrive, command line, shell…), it seems that Veeam ask to the VSS hardware provider (Snapdrive) to remove the snapshot before it is unmounted.

The problem is that the snapshot are never removed (before that the max 255 snapshot and the autodelete options start) or I have to script the cleaning after each jobs. Is there anybody that got the same issue ? Any idea how to solve that ?

[Gostev]

You should contact NetApp support, Veeam cannot really assist you with SnapDrive issues. Veeam uses standard Microsoft VSS API calls to manage snapshots, as per Microsoft requirements of performing Hyper-V backup. This sounds like a bug in SnapDrive interop with hardware VSS provider from NetApp.

[collinp]

I have the same issue, sometimes netapp snapshots aren't cleaned up automatically after the Veeam backups run. I run a daily batch script to cleanup the netapp snaps. Ocassionally I also see left over SnapDrive volume-level clones that I have to cleanup.

[vchris]

Hello guys,

Collinp: Can you provide me your script ? Actually I've set the snapreserve and the autodelete to avoid surprise.

Gostev: When I trigger the same command manually I don't have the issue :-/

[Gostev]

What same command? Do you trigger VSS API functions from command line??

[collinp]

The script isn't elegant and I have to use freeware "System Scheduler" to schedule the tasks. Doesn't seem to run from windows task scheduler. All volumes from both filers are listed in a single text file and both filers go through the list of volumes.

Code: Select all

rem c:\scripts\DeleteNetappSnapshots.bat
taskkill /F /IM putty.exe
start cmd /c "c:\putty.exe -ssh root@filer1 -pw Password1 -m c:\scripts\snapshot.txt"
start cmd /c "c:\putty.exe -ssh root@filer2 -pw Password1 -m c:\scripts\snapshot.txt"

c:\scripts\snapshot.txt contents below

Code: Select all

snap delete -a -f vol_filer1_SAS1
snap delete -a -f vol_filer2_SATA1