Memory hot-plug increase causing VM power off when using `nutanix_virtual_machine_v2` resource

[bbiegun-te]

On an attempt of increase memory on a VM which supports hot-plug addition the VM is being powered off when using the nutanix_virtual_machine_v2 resource.

The Nutanix documentation claims, that such operation is supported:

You can increase the memory allocation and (...) the number of vCPUs (sockets) while the VMs are powered on.

Ref: https://portal.nutanix.com/page/documents/details?targetId=AHV-Admin-Guide-v11_0:ahv-vm-memory-and-cpu-configuration-c.html

Recycling the VM is caused by incorrect condition under resource_nutanix_virtual_machine_v2.checkForHotPlugChanges() function.

There is no indication coming from the provider that a VM will be restarted on applying the changes.

Same change in UI doesn't cause the VM to be restarted.

Nutanix Cluster Information

• Nutanix Cluster (Prism Element / AOS): 7.3.1.1
• Nutanix Prism Central: pc.7.3.1.1

Terraform Version

Terraform v1.14.7 on linux_amd64
nutanix/nutanix v2.3.4
I'm unable to use the newest version of the provider as it is incompatible with PC. Mentioned code haven't changed between the versions.

Affected Resource(s)

• nutanix_virtual_machine_v2

Terraform Configuration Files

resource "nutanix_virtual_machine_v2" "compute" {
  memory_size_bytes    = var.computes_memory_size_mb * 1024 * 1024
  num_cores_per_socket = 1
  num_sockets          = var.computes_num_vcpus
  ...
}

Debug Output

Panic Output

Expected Behavior

VM memory is updated without power cycling the VM.

Actual Behavior

VM is being power cycled on increasing memory for hot-plug enabled VM.

Steps to Reproduce

1. create a VM with 1GB of memory using nutanix_virtual_machine_v2 resource
2. terraform apply
3. set VM memory to 2GB in terraform
4. terraform apply

Important Factors