Consistent change proposal on ecs service capacity provider

[ghost]

This issue was originally opened by @vladimirtiukhtin as hashicorp/terraform#23740. It was migrated here as a result of the provider split. The original body of the issue is below.

---

Terraform Version

Terraform v0.12.13
+ provider.aws v2.43.0
+ provider.random v2.2.1

Expected Behavior

When cluster's default capacity provider is used, no changes must be proposed to a service with no capacity provider defined.

Actual Behavior

Change is proposed

Steps to Reproduce

Create a cluster with default capacity provider

resource "aws_ecs_cluster" "cluster" {
  name = "test"
  capacity_providers = [
    "FARGATE",
    "FARGATE_SPOT"
  ]

  default_capacity_provider_strategy {
    capacity_provider = "FARGATE_SPOT"
    weight                   = 100
  }
}

Create a service not specifying capacity provider which forces it to use cluster's default

resource "aws_ecs_service" "service" {
  name                              = "test"
  cluster                           = "test"
  task_definition                   = aws_ecs_task_definition.task.arn
  desired_count                     = 1

  network_configuration {
    security_groups  = var.security_group_id
    subnets          = var.subnets
    assign_public_ip = false
  }
}

Apply. Now plan. A change to capacity provider will be proposed

[louis-ellevation]

I am currently having trouble due to this issue. Combined with this issue it seems I can't have an ECS service use the default capacity provider without said service incurring downtime every time the terraform config is applied.

[alex-bes]

I have managed to work around this issue as described in #11351 (comment)

[github-actions]

Marking this issue as stale due to inactivity. This helps our maintainers find and focus on the active issues. If this issue receives no comments in the next 30 days it will automatically be closed. Maintainers can also remove the stale label.

If this issue was automatically closed and you feel this issue should be reopened, we encourage creating a new issue linking back to this one for added context. Thank you!

[github-actions]

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.