snowflake_failover_group object types showing changes on every run

[gdelia-pm]

Terraform CLI and Provider Versions

 .\terraform.exe version
Terraform v1.7.3
on windows_amd64
+ provider registry.terraform.io/snowflake-labs/snowflake v0.86.0

Terraform Configuration

resource "snowflake_failover_group" "SOURCE_FAILOVER_GROUP" {
  depends_on = [snowflake_database.DB, snowflake_storage_integration.AZURE_STORAGE_INTEGRATION_DEV, snowflake_role.FUNCTIONAL_ROLES, snowflake_warehouse.WH,]

  name                      = "FAILOVER_GROUP1"
  object_types              = ["ACCOUNT PARAMETERS", "DATABASES", "INTEGRATIONS", "NETWORK POLICIES", "RESOURCE MONITORS", "ROLES", "USERS", "WAREHOUSES"]
  allowed_accounts          = ["ACCOUNT.CORE1", "ACCOUNT.CORE_DR1"]
  allowed_integration_types = ["SECURITY INTEGRATIONS", "STORAGE INTEGRATIONS"]
  allowed_databases         = toset([for db in snowflake_database.DB : db.name])

  replication_schedule {
    interval = 15 # minutes
  }
}

Expected Behavior

After the initial apply no further plan should be generated.

Actual Behavior

On every plan, object type PARAMETERS is detected as removed, while ACCOUNT PARAMETERS is set to be added.
This seems to be a result of how snowflake stores the account parameter option, as it's noted as just parameter when running show failover groups. I think the tf provider is reading the parameters on each load and seeing that as different than the accepted object type of account parameters.

I can work around this by creating the group with the "ACCOUNT PARAMETERS" object type, applying the config, then changing my object type to "PARAMETERS" to suppress the change, but if I have to make another change that breaks things.

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

Terraform will perform the following actions:

  # snowflake_failover_group.SOURCE_FAILOVER_GROUP will be updated in-place
  ~ resource "snowflake_failover_group" "SOURCE_FAILOVER_GROUP" {
        id                        = "FAILOVER_GROUP1"
        name                      = "FAILOVER_GROUP1"
      ~ object_types              = [
          - "PARAMETERS",
          + "ACCOUNT PARAMETERS",
            # (7 unchanged elements hidden)
        ]
        # (5 unchanged attributes hidden)

        # (1 unchanged block hidden)
    }

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

Steps to Reproduce

1. add the failover resource with object type "ACCOUNT PARAMETERS"
2. terraform apply
3. terraform apply will show an unexpected change
4. changing the object type on the failover resource to "PARAMETERS" will stop detecting a new change
5. terraform destroy
6. terraform apply will fail with the PARAMETERS object type set for a new resource

How much impact is this issue causing?

Medium

Logs

No response

Additional Information

No response

[sfc-gh-asawicki]

Hey @gdelia-pm. Thanks for reaching out to us.

I confirmed that this is undocumented Snowflake behavior (returning PARAMETERS instead of ACCOUNT PARAMETERS). I have adjusted the SDK logic in #2522. It should be merged before 0.87.0 release.

[sfc-gh-asawicki]

Hey @gdelia-pm . We have released the fix as part of v0.87.0 release. Please follow the migration guide during the update. Please confirm that the issue is resolved in the newest version. Thanks!

[gdelia-pm]

Confirming this is fixed in 0.87. Thanks!