Attach multiple disk to multiple instances using google_compute_attached_disk

Terraform Providers Google
1

I have been trying to create a script that allows the attachment of multiple disks into multiple instances using google_compute_attached_disk. What should I do in the resource “google_compute_attached_disk”?

resource "google_compute_disk" "encrypted_disk" {
          count = "${var.resource_count}"
          name = "${var.disk_name}${count.index}"
          type  = "${var.disk_type}"
          zone  = "${var.instance_zone}"
          size = "${var.disk_size}"
          image = "${var.image}"
          labels = "${var.disk_labels}"
          disk_encryption_key {
            kms_key_self_link = "${var.encryption_key}"
          }
        }

resource "google_compute_attached_disk" "attached_encrypted_disk" {
  disk = "${google_compute_disk.encrypted_disk.self_link}"
  instance = "${google_compute_instance.compute_instance.self_link}"
}

resource "google_compute_instance" "compute_instance" {
  count = "${var.resource_count}"
  name         = "${var.instance_name}${count.index}"
  machine_type = "${var.instance_type}"
  zone         = "${var.instance_zone}"

  tags = "${var.target_tags}"

  boot_disk {
    kms_key_self_link = "${var.encryption_key}"
    auto_delete = "${var.disk_auto_delete}"
    initialize_params {
      size = "${var.image_size}"
      type = "${var.disk_type}"
      image = "${var.image}"
    }
  }

  network_interface {
    network = "${var.network}"
    network_ip = "${var.private_ip}"
  }

  lifecycle {
    ignore_changes = ["attached_disk"]
  }

  metadata = "${var.metadata}"

  labels = "${var.instance_labels}"

  scheduling {
    preemptible       = "${var.preemptible}"
    automatic_restart = "${var.automatic_restart}"
  }

  service_account {
    scopes = "${var.scopes}"
  }
}
2

Even I am also facing the same issue. reading few blogs I understand that using count as a loop is not recommendable. The reason is justified in this link (https://github.com/hashicorp/terraform/issues/17179), instead try to use for_each.