Is the cluster only connected to that specific switch? Or is there another switch (MLAG) where the backup or LACP interfaces are connected to?
If this is the only switch, yes, shutdown the cluster.
I agree with Jeroen… As you expect downtime for the network, please poweroff running virtual machines if any.
When that’s finished, stop the Nutanix cluster from commandline… No need to poweroff the hosts after that. When network is up again, you can start the cluster.
This will avoid any corrupt data.
understand these way ,
All CVMs communicate with each other through TOR, where all your nodes are connected. If the switch goes down, cluster services will inevitably be affected. In my experience, I have witnessed network outages that didn't result in any data loss on the cluster. However, as Bart and Jeroen mentioned, you should follow the precautions below:"
https://portal.nutanix.com/page/documents/details?targetId=AHV-Admin-Guide-v6_7:ahv-cluster-shut-down-t.html