beta provider fails when setting min_cpu_platform in auto_provisioning_defaults for google_container_cluster

[aaronstutzer]

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.
• If an issue is assigned to a user, that user is claiming responsibility for the issue.
• Customers working with a Google Technical Account Manager or Customer Engineer can ask them to reach out internally to expedite investigation and resolution of this issue.

Terraform Version & Provider Version(s)

Terraform v1.8.2
on

• provider registry.terraform.io/hashicorp/google v5.29.0
• provider registry.terraform.io/hashicorp/google-beta v5.29.0

Affected Resource(s)

google_container_cluster

Terraform Configuration

  cluster_autoscaling {
    enabled = true
    resource_limits {
      resource_type = "cpu"
      minimum       = var.cpu_min
      maximum       = var.cpu_max
    }
    resource_limits {
      resource_type = "memory"
      minimum       = var.memory_min
      maximum       = var.memory_max
    }
    auto_provisioning_defaults {
      service_account = var.sa_gke_nodes
      min_cpu_platform = "Intel Cascade Lake"
      management {
        auto_upgrade = true
      }
      shielded_instance_config {
        enable_integrity_monitoring = true
        enable_secure_boot          = false
      }
    }
  }

Debug Output

2024-05-14T12:21:58.579+0200 [ERROR] provider.terraform-provider-google_v5.29.0_x5: Response contains error diagnostic: tf_resource_type=google_container_cluster @caller=github.com/hashicorp/terraform-plugin-go@v0.22.1/tfprotov5/internal/diag/diagnostics.go:58 diagnostic_detail="" diagnostic_summary="project: required field is not set" tf_proto_version=5.4 tf_provider_addr=registry.terraform.io/hashicorp/google tf_req_id=7e28ccf1-787a-6db6-a0e9-bdbfa3d06968 @module=sdk.proto diagnostic_severity=ERROR tf_rpc=ApplyResourceChange timestamp="2024-05-14T12:21:58.579+0200"
2024-05-14T12:21:58.580+0200 [DEBUG] State storage *remote.State declined to persist a state snapshot
2024-05-14T12:21:58.580+0200 [ERROR] vertex "google_container_cluster.prod" error: project: required field is not set
2024-05-14T12:21:58.580+0200 [DEBUG] states/remote: state read serial is: 202; serial is: 202
2024-05-14T12:21:58.580+0200 [DEBUG] states/remote: state read lineage is: 2aabb3ac-527d-a139-ec90-5a6a1102efc8; lineage is: 2aabb3ac-527d-a139-ec90-5a6a1102efc8
╷
│ Error: project: required field is not set
│
│ with google_container_cluster.prod,
│ on gke.tf line 72, in resource "google_container_cluster" "prod":
│ 72: resource "google_container_cluster" "prod" {
│
╵
2024-05-14T12:21:58.958+0200 [DEBUG] provider.terraform-provider-google_v5.29.0_x5: 2024/05/14 12:21:58 [DEBUG] [transport] [server-transport 0x14000162600] Closing: Server.Stop called
2024-05-14T12:21:58.958+0200 [DEBUG] provider.terraform-provider-google_v5.29.0_x5: 2024/05/14 12:21:58 [DEBUG] [transport] [server-transport 0x14000162600] loopyWriter exiting with error: transport closed by client
2024-05-14T12:21:58.959+0200 [DEBUG] provider.stdio: received EOF, stopping recv loop: err="rpc error: code = Unavailable desc = error reading from server: EOF"
2024-05-14T12:21:58.964+0200 [DEBUG] provider: plugin process exited: path=.terraform/providers/registry.terraform.io/hashicorp/google/5.29.0/darwin_arm64/terraform-provider-google_v5.29.0_x5 pid=82611
2024-05-14T12:21:58.964+0200 [DEBUG] provider: plugin exited

Expected Behavior

new cluster should be deployed with min_cpu_platform "Intel Cascade Lake"

Actual Behavior

when i deploy with google-beta provider i get the above error and terraform terminates.
when i use the google provider the deployment is successful but the min_cpu_platform is ignored.

Steps to reproduce

1. set min_cpu_platform
2. terraform apply

Important Factoids

No response

References

google documentation
google_container_cluster documentation

[ggtisc]

Hi @aaronstutzer!

Please provide your complete terraform code to reproduce this scenario because you only have a little fragment of your code(you can exclude sensitive data, but we need to be sure that you are setting valid values)