Terraform always wants to recreate HDInsight Kafka Cluster

[Djiit]

Community Note

• Please vote on this issue by adding a 👍 reaction to the original issue to help the community and maintainers prioritize this request
• Please do not leave "+1" or "me too" comments, they generate extra noise for issue followers and do not help prioritize the request
• If you are interested in working on this issue or have submitted a pull request, please leave a comment

Terraform (and AzureRM Provider) Version

Affected Resource(s)

• azurerm_hdinsight_kafka_cluster

Terraform Configuration Files

resource "azurerm_hdinsight_kafka_cluster" "kafka-hdi_cluster" {
  name                = "mykafka"
  resource_group_name = azurerm_resource_group.kafka-resource_group[0].name
  location            = azurerm_resource_group.kafka-resource_group[0].location
  cluster_version     = "4.0"
  tier                = "Standard"

  component_version {
    kafka = "2.1"
  }

  gateway {
    enabled  = true
    username = module.vault.secrets["kafka-gateway-username"].value
    password = module.vault.secrets["kafka-gateway-password"].value
  }

  storage_account {
    storage_container_id = azurerm_storage_container.kafka-storage_container.id
    storage_account_key  = azurerm_storage_account.kafka-storage_account.primary_access_key
    is_default           = true
  }

  roles {
    head_node {
      vm_size  = "Standard_D3_V2"
      username = "obfuscated"
      password = "obfuscated"
    }

    worker_node {
      vm_size                  = "Standard_D3_V2"
      username                 = "obfuscated"
      password                 = "obfuscated"
      number_of_disks_per_node = 3
      target_instance_count    = 3
    }

    zookeeper_node {
      vm_size  = "Standard_D3_V2"
      username = "obfuscated"
      password = "obfuscated"
    }
  }
}

Expected Behavior

The resource should be left untouched., the Apply output should be green.

Actual Behavior

Terraform tells me it needs to recreate the resource because of some keys I didn't set 👍

  # azurerm_hdinsight_kafka_cluster.kafka-hdi_cluster must be replaced
-/+ resource "azurerm_hdinsight_kafka_cluster" "kafka-hdi_cluster" {
      ~ cluster_version     = "4.0.1000.1" -> "4.0"
      ~ https_endpoint      = "obfuscated.azurehdinsight.net" -> (known after apply)
      ~ id                  = "obfuscated" -> (known after apply)
        location            = "westeurope"
        name                = "mykafka"
        resource_group_name = "obfuscated"
      ~ ssh_endpoint        = "obfuscated.azurehdinsight.net" -> (known after apply)
      ~ tier                = "standard" -> "Standard"

        component_version {
            kafka = "2.1"
        }

      ~ gateway {
            enabled  = true
          ~ password = (sensitive value)
            username = "adminhdi"
        }

      ~ roles {
          ~ head_node {
                password = (sensitive value)
              - ssh_keys = [] -> null
                username = "adminhdi"
                vm_size  = "Standard_D3_V2"
            }

          ~ worker_node {
              - min_instance_count       = 0 -> null
                number_of_disks_per_node = 3
                password                 = (sensitive value)
              - ssh_keys                 = [] -> null
                target_instance_count    = 3
                username                 = "adminhdi"
                vm_size                  = "Standard_D3_V2"
            }

          ~ zookeeper_node {
                password = (sensitive value)
              - ssh_keys = [] -> null
                username = "adminhdi"
                vm_size  = "Standard_D3_V2"
            }
        }

        storage_account {
            is_default           = true
            storage_account_key  = (sensitive value)
            storage_container_id = "https://onediorkafkastaging.blob.core.windows.net/hdinsight"
        }
    }

What I tried :

• set the ssh_keys to [] but with no luck (you can't define username/password AND ssh_keys).

Steps to Reproduce

1. Define an HDI Kafka cluster.
2. terraform apply
3. terraform apply again
4. Terraform will tell you it needs to recreate the HDI cluster.

[tombuildsstuff]

hey @Djiit

Thanks for opening this issue.

Taking a look into this, this appears to have been fixed via #4489, which is due to a change in the way the API responds (it now returns "*****" for the "password" field, regardless of the value, so Terraform shows a diff/forces a new resource since this doesn't match the definition in code and it's not possible to change the password in the API).

Since this has been fixed via #4489 which will ship as a part of v1.35 I'm going to close this issue for the moment - but if you're still seeing this issue after upgrading to v1.35 please let us know and we'll take another look :)

Thanks!

[Djiit]

Good news ! Thanks 🙏

…

-- Julien Le jeu. 3 oct. 2019 à 16:30, Tom Harvey <notifications@github.com> a écrit :

hey @Djiit <https://github.com/Djiit> Thanks for opening this issue. Taking a look into this, this appears to have been fixed via #4489 <#4489>, which is due to a change in the way the API responds (it now returns "*****" for the "password" field, regardless of the value, so Terraform shows a diff/forces a new resource since this doesn't match the definition in code and it's not possible to change the password in the API). Since this has been fixed via #4489 <#4489> which will ship as a part of v1.35 I'm going to close this issue for the moment - but if you're still seeing this issue after upgrading to v1.35 please let us know and we'll take another look :) Thanks! — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#4485>, or mute the thread <https://github.com/notifications/unsubscribe-auth/AAI5TNBRT3JP6V2PLM6MXTTQMX6Y7ANCNFSM4I4YORAQ> .

[ghost]

This has been released in version 1.35.0 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 = "~> 1.35.0"
}
# ... 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 👉 hashibot-feedback@hashicorp.com. Thanks!