[Bug]: Import missing details such as scheduler_au, and it causes issues in applying

[coreywebber]

Description

So I have an older deployment, which might be the issue. I have a simple resource definition written to mirror my current deployment. I imported the deployment and then ran a plan to see what missed. scheduler_au seems to be the issue, it's default is null.

The plan

  # astro_deployment.airflow_deployment will be updated in-place
  ~ resource "astro_deployment" "airflow_deployment" {
      ~ airflow_version             = "2.10.3" -> (known after apply)
      ~ astro_runtime_version       = "12.4.0" -> (known after apply)
      + cluster_id                  = (known after apply)
      ~ dag_tarball_version         = "2024-11-22T03:08:01.7254992Z" -> (known after apply)
      ~ desired_dag_tarball_version = "2024-11-22T03:08:01.7254992Z" -> (known after apply)
        id                          = "omitted"
      ~ image_repository            = "images.astronomer.cloud/omitted/omitted" -> (known after apply)
      ~ image_tag                   = "deploy-2024-11-22T03-03-08" -> (known after apply)
      ~ image_version               = "deploy-2024-11-22T03-03-08" -> (known after apply)
        name                        = "Production"
      ~ scaling_status              = {
          ~ hibernation_status = {
              ~ is_hibernating  = false -> (known after apply)
              + next_event_at   = (known after apply)
              + next_event_type = (known after apply)
              ~ reason          = "Hibernation not enabled for this deployment" -> (known after apply)
            } -> (known after apply)
        } -> (known after apply)
      - scheduler_au                = 10 -> null
      ~ scheduler_cpu               = "1" -> (known after apply)
      ~ scheduler_memory            = "2Gi" -> (known after apply)
      ~ scheduler_replicas          = 2 -> (known after apply)
      ~ status                      = "HEALTHY" -> (known after apply)
      + status_reason               = (known after apply)
      ~ updated_at                  = "2024-11-22 22:50:40.167 +0000 UTC" -> (known after apply)
      ~ updated_by                  = {
          ~ api_token_name = "Github Actions" -> (known after apply)
          + avatar_url     = (known after apply)
          + full_name      = (known after apply)
          ~ id             = "omitted" -> (known after apply)
          ~ subject_type   = "SERVICEKEY" -> (known after apply)
          + username       = (known after apply)
        } -> (known after apply)
        # (26 unchanged attributes hidden)
    }

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

The Error

│ Error: Provider produced inconsistent result after apply
│ 
│ When applying changes to astro_deployment.airflow_deployment, provider "provider[\"registry.opentofu.org/astronomer/astro\"]" produced an unexpected new value: .scheduler_au: was null, but now
│ cty.NumberIntVal(10).

Steps To Reproduce

Fresh project
Imported my deployment
Applied to see plan
I approved the changes
Got the error

Acceptance Criteria

I would like to import all existing properties, and have no changes except where there is drift. If it does need to by submitted to know, find, but don't error :)

Anything else?

No response