I am working with a customer who has implemented a new 4 node Virtualised SAN solution. They now only have enough Veeam licenses for half of their sockets and no budget this year to upgrade Veeam.
The bandaid solution is to get all VM's in the small environment migrated to 2 of the hosts, run the backups then, rebalance the VM's. The options we considered are:
1) place two hosts in maintenance mode and and shut down the Virtual SAN appliance on each node
2) Turn off DRS, discover all VM's on two hosts and vMotion them off, excluding the Virtalised SAN Appliances
I have been able to script option 1 but it really is a bad solution considering it takes half of the SAN offline. The reason we are shutting down the SAN nodes is that the enter maintenance mode task will not complete with them running.
Option 2 is my preferred option however I do not know how to use the Get-VM set of commands to get all VM's excluding the names of the Virtualised SAN nodes.
Does anyone have any previous experience with Pre / Post backup scripts to achieve what we are trying to do?
Cheers in advance.
Connect-VIServer 10.20.1.212 -User firstname.lastname@example.org -Password ##### (not encrypted yet)
Get-VMHost -Name 10.x.x.x | Set-VMHost -State Maintenance -Evacuate
Did you already consider using DRS rules?
With 2 sets of VM to Host rules (1 for backup and 1 for normal operations), you should be able to move your VMs to/from the 2 ESXi nodes.
You have to direct the Veeam solution to those 2 ESXi nodes that hold the VMs (and exclude the VSA VMs).
Working with DRS rules and VM to Host groups can be scripted with our DRSRule module.