Compare region_backend_service.backend[].group as a relative path

[tmshn]

In a region_backend_service resource, backend[].group argument is compared using full url. Therefore, if you pass a self_link of instance_group with beta feature enabled, you'll suffer from persisting perpetual diff.

E.g.:

resource "google_compute_instance_group_manager" "foo" {
  // ...configs (omitted)...
  // ↓ This is beta feature
  update_strategy       = "ROLLING_UPDATE" 
  rolling_update_policy = {
    type           = "PROACTIVE"
    minimal_action = "REPLACE"
  }
}

resource "google_compute_region_backend_service" "foo" {
  // ...configs (omitted)...
  backend = [
    {
        group = "${google_compute_instance_group_manager.foo.self_link}"
    },
  ]
}

With this configuration, following diff always occurs.

------------------------------------------------------------------------

An execution plan has been generated and is shown below.
Resource actions are indicated with the following symbols:
  ~ update in-place

Terraform will perform the following actions:

  ~ google_compute_region_backend_service.foo
      backend.1111111111.description: "" => ""
      backend.1111111111.group:       "https://www.googleapis.com/compute/v1/projects/my-project/zones/us-central1-a/instanceGroups/igm-foo" => ""
      backend.2222222222.description: "" => ""
      backend.2222222222.group:       "" => "https://www.googleapis.com/compute/beta/projects/my-project/zones/us-central1-a/instanceGroups/igm-foo"

Plan: 0 to add, 1 to change, 0 to destroy.

------------------------------------------------------------------------

To suppress this, I set compareSelfLinkRelativePaths as DiffSuppressFunc for backend[].group, like non-regional backend_service does.
https://github.com/terraform-providers/terraform-provider-google/blob/v1.12.0/google/resource_compute_backend_service.go#L80

[tmshn]

Should I add a test for this?

[emilymye]

I actually think this is the appropriate behavior - if a resource is changed to a beta resource, it should be recreated. If after the initial v1 --> beta change the diff persists, that's more an issue around how we compute the instance group for the initial diff.

[danawillow]

@emilymye that's not actually true- v1 and beta resources are the same underneath. the self_link only describes which API we used to access it. I'm going to go ahead and merge this, hope you don't mind :)

[danawillow]

@tmshn whoops, merged before realizing you'll also need to change the hash function. take a look at how we do it for regular backend services- it's already been fixed there.

[tmshn]

@danawillow Oh I missed that! Thanks to point out. I fixed it in #1491.

[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!