• Verify all hosts and CVMs are in the same time zone.

  • If putting the CVMs and hosts in a different native VLAN, it is easier to do before creating VMs, but can be done afterwards if down time is not an issue.

  • Get cluster parameters - ncli cluster get-params
  • Set Cluster time zone = ncli cluster set-timezone timezone=America/New_York

  • Cluster status = cluster status

  • Check /etc/resolv.conf on the hosts. Mine was missing after the initial install, and started throwing errors after a few days. I resolved it by creating the file with similar DNS servers as the CVMs.

  • Nutanix Cluster Check (NCC)
  • Run Multiple checks: ncc health_checks system_checks plugin_list=cluster_version_check,cvm_reboot_check
  • Run all: ncc health_checks run_all

  • Rerun failing checks: ncc --rerun_failing_plugins=True

  • Checks can be run in parallel with --parallel=4 flag (4 is the max)

  • Display status in terminal with –use-npyscreen

  • Power Off all VMs that are powered on: for vm_name in `acli vm.list power_state=on | grep -v ^'VM name' | awk '{print $1}'`; do acli vm.force_off $vm_name; done

  • Change CVM vlan: change_cvm_vlan VLAN_ID

  • Set the OpenVSwitch Vlan on the host: ovs-vsctl set port br0 tag=10

  • Power on all powered off VMs: for vm_name in `acli vm.list power_state=off | grep -v ^'VM name' | awk '{print $1}'`; do acli vm.on $vm_name; done