Terraform destroy does not handle 404 on Google Compute Instance

[clstokes]

If I create a Google Compute Instance with Terraform, then remove the instance via the Google Cloud console, a terraform destroy fails with a no longer exists error (likely a 404) and won't continue. Other Terraform providers seem to handle this case by handling the 404 as a successful destroy and continuing on.

Tested with Terraform v0.6.6.

Terraform config:

resource "google_compute_instance" "main" {
  name         = "main"
  machine_type = "n1-standard-1"
  zone         = "us-central1-f"

  disk {
    image = "ubuntu-1404-trusty-v20151113"
  }

  network_interface {
    network = "default"
    access_config {
      // Ephemeral IP
    }
  }
}

Abridged Terraform output...

$ terraform apply

...

google_compute_instance.main: Creating...
  can_ip_forward:                             "" => "0"
  disk.#:                                     "" => "1"
  disk.0.auto_delete:                         "" => "1"
  disk.0.image:                               "" => "ubuntu-1404-trusty-v20151113"
  machine_type:                               "" => "n1-standard-1"
  metadata_fingerprint:                       "" => "<computed>"
  name:                                       "" => "main"
  network_interface.#:                        "" => "1"
  network_interface.0.access_config.#:        "" => "1"
  network_interface.0.access_config.0.nat_ip: "" => "<computed>"
  network_interface.0.address:                "" => "<computed>"
  network_interface.0.name:                   "" => "<computed>"
  network_interface.0.network:                "" => "default"
  self_link:                                  "" => "<computed>"
  tags_fingerprint:                           "" => "<computed>"
  zone:                                       "" => "us-central1-f"
google_compute_instance.main: Creation complete

Apply complete! Resources: 1 added, 0 changed, 0 destroyed.

...

$ 
$ # Instance removed via Google Cloud console
$ 
$ terraform destroy -force

...

google_compute_instance.main: Refreshing state... (ID: main)
Error refreshing state: 1 error(s) occurred:

* google_compute_instance.main: Resource hashicorp-clstokes no longer exists
$

[lwander]

This has already been brought up in #3742, and is being addressed in #3913. However, @apparentlymart made the case that the typical behavior is to allow terraform show to resolve this by calling Read(..) on the deleted resources, marking them as no longer existing. What do you think, @clstokes?

[clstokes]

I'm definitely good with @apparentlymart's suggestion. I'll close this in favor of #3742. Thanks @lwander!

[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.