Hyper-V VNIC Failing

[BlueMan]

When Veeam is backing up our live Hyper-V cluster, the production VNIC randomly fails, so access to the node and VMs running on the node is not available. The node has to then be shutoff via the power button and the started again to get access to the VMs and node from the network.

I have been working with support but so far we have not been able to find a resolution. Has anyone encountered this before or can anyone supply any tips to try and fix the problem?

[foggy]

Looks like the cluster network becomes overloaded with the massive backup traffic. Could you please describe your cluster setup in a bit more details (including networking)? How many NICs do the Hyper-V hosts have? Also, what kind of VSS provider is used?

[BlueMan]

2 x HP Proliant DL360p
HP P2000 G3 SAS SAN
Cluster Shared Volume

The hosts have 4 NICs each. 1 for the production network (this is the one that fails). 1 for the heartbeat between the 2 hosts. 1 for a DMZ. 1 for live migration.

VSS Software Provider.

[BlueMan]

Should the hotfixes in the following article be applied to Windows Server 2012 R2 hosts or are they only relevant for 2012 hosts? http://www.veeam.com/kb1838

[foggy]

And how P2000 is connected to the hosts (FC/Ethernet)?

[BlueMan]

SAS

[foggy]

Sorry, missed that from your previous post.

Backup using software VSS provider is known to put significant load on the cluster network. In Veeam B&R v7, however, this effect is not that essential as it could be previously, since Veeam agent is always started on CSV owner node. What Veeam B&R version are you at?

[BlueMan]

7.0.0.833

[foggy]

What do our engineers say regarding that?