cloud_sql_instance volumes in google_cloud_run_v2_service inconsistent ordering

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 version
OpenTofu v1.8.5
on darwin_amd64
+ provider registry.opentofu.org/hashicorp/google v6.11.2

Affected Resource(s)

google_cloud_run_v2_service

Terraform Configuration

resource "google_cloud_run_v2_service" "default" {
  name = "cloudrun-service"

  location            = "us-central1"
  project             = "innablr-dev"

  template {
    execution_environment = "EXECUTION_ENVIRONMENT_GEN2"

    containers {
      image = "us-docker.pkg.dev/cloudrun/container/hello"
      volume_mounts {
        name       = "bucket"
        mount_path = "/var/www"
      }
      volume_mounts {
        name       = "cloudsql"
        mount_path = "/cloudsql"
      }
    }

    volumes {
      name = "cloudsql"
      cloud_sql_instance {
        instances = ["innablr-dev:australia-southeast2:innablr-dev"]
      }
    }

    volumes {
      name = "bucket"
      gcs {
        bucket    = google_storage_bucket.default.name
        read_only = false
      }
    }
  }
}

resource "google_storage_bucket" "default" {
  name     = "innablr-test-junk-cloudrun-service"
  location = "US"
  project  = "innablr-dev"
}

Debug Output

terraform2.log

Expected Behavior

terraform plan should show no changes

Actual Behavior

terraform plan wants to swap the order of the volumes. even if I apply this, the next plan/apply is the same.

❯ TF_LOG=debug TF_LOG_PATH=terraform2.log terraform apply
google_storage_bucket.default: Refreshing state... [id=innablr-test-junk-cloudrun-service]
google_cloud_run_v2_service.default: Refreshing state... [id=projects/innablr-dev/locations/us-central1/services/cloudrun-service]

OpenTofu used the selected providers to generate the following execution plan. Resource actions are indicated with the following symbols:
  ~ update in-place

OpenTofu will perform the following actions:

  # google_cloud_run_v2_service.default will be updated in-place
  ~ resource "google_cloud_run_v2_service" "default" {
      ~ deletion_protection     = false -> true
        id                      = "projects/innablr-dev/locations/us-central1/services/cloudrun-service"
        name                    = "cloudrun-service"
        # (26 unchanged attributes hidden)

      ~ template {
            # (7 unchanged attributes hidden)

          ~ volumes {
              ~ name = "bucket" -> "cloudsql"

              + cloud_sql_instance {
                  + instances = [
                      + "innablr-dev:australia-southeast2:innablr-dev",
                    ]
                }

              - gcs {
                  - bucket    = "innablr-test-junk-cloudrun-service" -> null
                  - read_only = false -> null
                }
            }
          ~ volumes {
              ~ name = "cloudsql" -> "bucket"

              - cloud_sql_instance {
                  - instances = [
                      - "innablr-dev:australia-southeast2:innablr-dev",
                    ] -> null
                }

              + gcs {
                  + bucket    = "innablr-test-junk-cloudrun-service"
                  + read_only = false
                }
            }

            # (2 unchanged blocks hidden)
        }

        # (1 unchanged block hidden)
    }

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

Steps to reproduce

terraform plan

Important Factoids

It looks like google-side API is returning the volumes in a particular order (cloudsql last?) which doesn't match what we're setting. If we SWAP the order of declaration of the volumes, then it plan/apply clean.

2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:     "volumes": [
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:       {
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         "name": "bucket",
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         "gcs": {
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:           "bucket": "innablr-test-junk-cloudrun-service"
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         }
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:       },
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:       {
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         "name": "cloudsql",
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         "cloudSqlInstance": {
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:           "instances": [
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google: "innablr-dev:australia-southeast2:innablr-dev"
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:           ]
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:         }
2024-11-16T15:58:17.128+1100 [DEBUG] provider.terraform-provider-google:       }

The same thing happens with the volume_mounts attribute.

References

https://github.com/hashicorp/terraform-provider-google/issues/13368 https://github.com/GoogleCloudPlatform/magic-modules/pull/8929

hashicorp / terraform-provider-google

cloud_sql_instance volumes in google_cloud_run_v2_service inconsistent ordering #20360