invalid planned change after resouce creation failure

[rvandegrift]

While troubleshooting hashicorp/terraform-provider-google#8261 I came upon a bug in terraform core. In that issue, a provider generates an invalid name for a resource. The google api rejects the creation.

If I apply again, terraform shows the failed resource as tainted, and exits with an invalid plan error.

Terraform Version

Terraform v0.14.4
+ provider registry.terraform.io/hashicorp/google v3.53.0
+ provider registry.terraform.io/hashicorp/google-beta v3.53.0

Terraform Configuration Files

variable "project" {
  type = string
}

provider "google" {
  project = var.project
}

provider "google-beta" {
  project = var.project
}

data "google_compute_network" "network" {
  name = "default"
}

data "google_compute_subnetwork" "subnet" {
  name   = "default"
  region = "us-central1"
}

resource "google_container_cluster" "cluster" {
  provider   = google-beta
  name       = "cluster-1"
  location   = "us-central1"
  network    = data.google_compute_network.network.self_link
  subnetwork = data.google_compute_subnetwork.subnet.self_link

  remove_default_node_pool = true
  node_pool {
    name       = "default-pool"
    node_count = 0
  }

  lifecycle {
    prevent_destroy = true

    ignore_changes = [
      node_pool,
      min_master_version,
    ]
  }
}


resource "google_container_node_pool" "nodes" {
  provider           = google-beta
  name_prefix        = "ca-nonprod-core-"
  location           = "us-central1"
  cluster            = google_container_cluster.cluster.name
  initial_node_count = 1

  node_config {
    machine_type = "n1-standard-4"
    disk_size_gb = 100
    image_type   = "COS_CONTAINERD"

    metadata = {
      disable-legacy-endpoints = "true"
    }

    oauth_scopes = [
      "https://www.googleapis.com/auth/logging.write",
      "https://www.googleapis.com/auth/monitoring",
      "https://www.googleapis.com/auth/devstorage.read_only",
    ]
  }

  lifecycle {
    create_before_destroy = true
  }
}

Debug Output

https://gist.github.com/rvandegrift/b9bc65e5d35c97b80890125e0ca27096

Expected Behavior

Terraform could continue. It wants to delete a resource that doesn't exist.

Actual Behavior

Terraform fails with:

Error: Invalid planned change

The plan contains a replace change for google_container_node_pool.nodes, which
doesn't exist yet. This is a bug in Terraform.

Steps to Reproduce

Additional Context

References

• google_container_node_pool with name_prefix generates names that are too long terraform-provider-google#8261

[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 have found a problem that seems similar to this, please open a new issue and complete the issue template so we can capture all the details necessary to investigate further.