[AS] - tf apply always needs in-place update for opentelekomcloud_as_policy_v2

[uli-schneider]

Terraform provider version

Terraform v1.5.7
on darwin_arm64

• provider registry.terraform.io/hashicorp/random v3.7.2
• provider registry.terraform.io/opentelekomcloud/opentelekomcloud v1.36.54

Affected Resource(s)

opentelekomcloud_as_policy_v2

Terraform Configuration Files

## Create Auto Scaling Policy - Scale Out
resource "opentelekomcloud_as_policy_v2" "as_policy_web_shared_scale_out" {
  scaling_policy_name   = "as-policy-web-shared-elb-scale-out-${local.dp_name}-${var.region}"
  scaling_policy_type   = "ALARM"
  scaling_resource_id   = opentelekomcloud_as_group_v1.as_group_web_shared_elb.id
  scaling_resource_type = "SCALING_GROUP"
  #
  alarm_id = opentelekomcloud_ces_alarmrule.alarm_rule_as_web_shared_elb_scale_out.id
  #
  cool_down_time = 300
  #
  scaling_policy_action {
    operation = "ADD"
    size      = 1
  }
}

Debug Output/Panic Output

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:

  # opentelekomcloud_as_policy_v2.as_policy_web_shared_scale_in will be updated in-place
  ~ resource "opentelekomcloud_as_policy_v2" "as_policy_web_shared_scale_in" {
        id                    = "45b043f9-0ade-4a83-8b64-f2eae6be3d4e"
        # (7 unchanged attributes hidden)

      - scheduled_policy {}

        # (1 unchanged block hidden)
    }

Steps to Reproduce

1. terraform apply

Expected Behavior

As "scheduled_policy" is an optional value, it should not force an "update in-place" during terraform apply.

Actual Behavior

With each run, some unneeded changes are processed.

Important Factoids

References

[muneeb-jan]

Hi @uli-schneider ,

From the output, it seems that there is an existing policy as_policy_web_shared_scale_in that is being updated in place. Could it be that you created or updated this at some point to SCHEDULED and then changed to alarm? That would explain the update in place. The other explanation is that when you create the scale out resource, the state is being refreshed and for an already existing resource scale_in, the scheduled_policy in state could be set to an empty block implicitly. Since the TF files have no scheduled_policy defined, the TF plans to remove it (provider drift cleanup).