azurerm_kubernetes_cluster: Add support for virtual machine scale set node pool

[dominik-lekse]

This pull request adds support for virtual machine scale set node pools in the resource and data source azurerm_kubernetes_cluster.

• The relevant test TestAccAzureRMKubernetesCluster_virtualMachineScaleSets requires the resource provider features MultiAgentpoolPreview and VMSSPreview. I assume these features are registered before the test is executed. Refer to https://docs.microsoft.com/en-us/azure/aks/use-multiple-node-pools.

Tests

=== RUN   TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
=== PAUSE TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
=== CONT  TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
--- PASS: TestAccAzureRMKubernetesCluster_virtualMachineScaleSets (896.94s)
PASS
ok      github.com/terraform-providers/terraform-provider-azurerm/azurerm       896.981s

New or Affected Resource(s)

• azurerm_kubernetes_cluster

References

• AKS with VMSS #2900

[katbyte]

Hi @dominik-lekse,

Thanks for the PR, i've left a few comments inline but overall this is looking good.

[dominik-lekse]

Hi @katbyte, thanks for the quick review. All suggestions confirmed.

[katbyte]

Thanks @dominik-lekse! this LGTM now 🚀

[katbyte]

I'm really sorry @dominik-lekse but i was looking at the wrong test results when I merged this 😞

It breaks all the other tests with:

DIFF:
        
        DESTROY/CREATE: azurerm_kubernetes_cluster.test
          addon_profile:                                        "" => "<computed>"
          addon_profile.#:                                      "1" => ""
          addon_profile.0.aci_connector_linux.#:                "0" => ""
          addon_profile.0.http_application_routing.#:           "0" => ""
          addon_profile.0.oms_agent.#:                          "0" => ""
          agent_pool_profile.#:                                 "1" => "1"
          agent_pool_profile.0.count:                           "2" => "2"
          agent_pool_profile.0.dns_prefix:                      "" => "<computed>"
          agent_pool_profile.0.fqdn:                            "acctestaks190512185214794318-7541a2a9.hcp.westeurope.azmk8s.io" => "<computed>"
          agent_pool_profile.0.max_pods:                        "30" => "<computed>"
          agent_pool_profile.0.name:                            "default" => "default"
          agent_pool_profile.0.os_disk_size_gb:                 "100" => "<computed>"
          agent_pool_profile.0.os_type:                         "Linux" => "Linux"
          agent_pool_profile.0.type:                            "AvailabilitySet" => "" (forces new resource)

because the api automatically sets it to AvailabilitySet. If that is the default unless type is specified then all we need to do is set the default to that value.

[dominik-lekse]

Hi @katbyte, apologies for being not strict enough on the test execution, I should have run at least one of the existing acceptance tests.

My assumption was that the Computed takes care of this case. With your suggestion to add the default value, the following tests pass including the TestAccAzureRMKubernetesCluster_basic. Due to resource limitations, I am not able to run all related acceptance tests.

With the default value, we have also covered upgrading of tf states produces by previous provider versions, right?

Since the commit is in the original branch, I will reopen the pull request.

=== RUN   TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
=== PAUSE TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
=== CONT  TestAccAzureRMKubernetesCluster_virtualMachineScaleSets
--- PASS: TestAccAzureRMKubernetesCluster_virtualMachineScaleSets (894.93s)
PASS
ok      github.com/terraform-providers/terraform-provider-azurerm/azurerm       894.976s

=== RUN   TestAccAzureRMKubernetesCluster_basic
=== PAUSE TestAccAzureRMKubernetesCluster_basic
=== CONT  TestAccAzureRMKubernetesCluster_basic
--- PASS: TestAccAzureRMKubernetesCluster_basic (894.15s)
PASS
ok      github.com/terraform-providers/terraform-provider-azurerm/azurerm       894.195s

[ghost]

This has been released in version 1.28.0 of the provider. Please see the Terraform documentation on provider versioning or reach out if you need any assistance upgrading. As an example:

provider "azurerm" {
	version = "~> 1.28.0"
}
# ... other configuration ...

[ghost]

I'm going to lock this issue because it has been closed for 30 days ⏳. This helps our maintainers find and focus on the active issues.

If you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. If you feel I made an error 🤖 🙉 , please reach out to my human friends 👉 [email protected]. Thanks!