segfault while handling nutanix_user_group resources

[phillipmacey]

Nutanix Cluster Information

PC 2024.3.1.4 (also had the same problem on older versions. Probably 2024.2 ish)

Terraform Version

Tested months ago using 1.x terraform versions having similar issues but I don't have the details captured. Restarted testing today initially using 1.11.3 and subsequently upgraded to 1.12.2 with no change. (MacOS build installed via brew)
terraform -v
Terraform v1.11.3
on darwin_arm64

• provider registry.terraform.io/nutanix/nutanix v2.2.1

Re-tested on updated version (same behaviour):
terraform -v
Terraform v1.12.2
on darwin_arm64

• provider registry.terraform.io/nutanix/nutanix v2.2.1

Affected Resource(s)

• resource.nutanix_user_group

Terraform Configuration Files

I have attempted to strip out all the bits that are not relevant to the crash based on my understanding that it is only the nutanix_user_group resource that has issues. (If I target some other non-group resource, terraform seems to run just fine. There are a number of other resources in my full code base)

variable "idp" {
  type          = map(string)
  description   = "Maps an Identity Provider by name to its UUID for ease of use. idp[\"name_of_idp\"] gives you the uuid"
}

variable "tenants" {
  type        = map(object({
    name                      = string,
    description               = string,
    accessGroupName           = optional(string, null)
    adminGroupName            = optional(string, null)
    # use this to map into the idp["name_of_idp"] variable
    idp                       = optional(string, null)
  }))
  default     = {}
  description = "List of tenants and their required attributes"
}

idp     = {
  "name-of-our-idp": "uuid-of-our-idp"
}


tenants = {
  "org-test"                = {
    name                      = "org Test"
    description               = "A test tenant for org to use"
  }
}


resource "nutanix_user_groups" "tenantBasicAccessGroups" {
  for_each      = var.tenants
  saml_user_group {
    # If accessGroupName is set for the tenant, use it. Otherwise, default to 'xxxx idp group'
    name        = coalesce(each.value.accessGroupName, join("", [each.key, " idp group"]))
    idp_uuid    = var.idp["name-of-our-idp"]
  }
}

Debug Output

https://gist.github.com/phillipmacey/8001da637def6e6c5afaa60a3b05bba8

Panic Output

Terraform passed on a segfault from the nutanix provider:

https://gist.github.com/phillipmacey/1a6ba93c042d0321036adce463157809

Expected Behavior

No segfaults, any errors handled cleanly

Actual Behavior

segfault bombed out the whole terraform run

Steps to Reproduce

1. Run any terraform plan or apply command: `terraform apply -target 'name-of-nutanix_user_group-resource'

Without targeting, the same fault still happens when it arrives at the nutanix_user_group resources. Other non-group resources can be targeted to get a successful run. (Didn't include any of the HCL for those non-group resources above but they do exist)

Important Factors

References