Force Instant Recovery network

[MarcelB]

We are using two physical networks between a physical Veeam server and the ESX hosts. One network VLAN30 is dedicated for management of the servers. We do not want data to and from the Veeam server to run over this network.
For backup and restore data we created a separate network VLAN50.
I am able to force restores (which will use the network) over VLAN50 by using a host file on the Veeam server pointing to the IP-adres of the VLAN50 adapter of the ESX host.

When performing an instant recovery, the ESX server initiates a NFS-datastore mount over the management network. Suppose a server with lots of i/o is being made operational after a instant recovery there is a risk the networkcapacity of the managment network is all used by the virtual machine.

Is there a method besides creating a local /etc/hosts to force an ESXi host to use the VMkernel port which is configuered for our backup network instead of the management network?

thanks
Marcel

[Gostev]

Hi Marcel, yes - basically you need to make sure Veeam Backup server name resolves to IP address on backup (not management) network on ESX host side. Other method would be to add vPower NFS datastore to your ESX host manually, as described here (Veeam will not attempt to modify existing datastore). Thanks!

[MarcelB]

Thanks. I mounted the NFS datastore manualy using the Ip of the backup network. Worked perfectly!

[Andreas Neufert]

There is a nother option. You can use the Windows Firewall at the vPOWER NFS Service Server to allow NFS Ports only on the Interfaces that you want to use. (TCP111)