[ISSUE] Configuration drift in databricks_pipeline when using gateway_definition

[alexott]

Creating a pipeline with LakeFlow Connect ingest gateway with this code:

resource "databricks_pipeline" "gateway" {
  name = "SQL Server Gateway"
  gateway_definition {
    connection_name         = "a9o-lakeflow-sql01"
    gateway_storage_catalog = "aott"
    gateway_storage_schema  = "tmp"
    gateway_storage_name    = "sql_server_staging"
  }
  channel = "PREVIEW"
}

leads to the following configuration drift:

  # databricks_pipeline.gateway must be replaced
-/+ resource "databricks_pipeline" "gateway" {
      - allow_duplicate_names  = false -> null
      - catalog                = "test" -> null # forces replacement
      + cause                  = (known after apply)
      ~ cluster_id             = "0121-082208-n00ldzcu" -> (known after apply)
      - continuous             = true -> null
      ~ creator_user_name      = "[email protected]" -> (known after apply)
      - development            = false -> null
      - expected_last_modified = 0 -> null
      + health                 = (known after apply)
      ~ id                     = "e69dc8e6-3826-4a82-9b46-3732f1b981e7" -> (known after apply)
      ~ last_modified          = 1737447728115 -> (known after apply)
        name                   = "SQL Server Gateway"
      - photon                 = false -> null
      ~ run_as_user_name       = "[email protected]" -> (known after apply)
      - serverless             = false -> null
      ~ state                  = "RUNNING" -> (known after apply)
      - target                 = "tmp" -> null
      ~ url                    = "https://aaaaa.cloud.databricks.com/#joblist/pipelines/e69dc8e6-3826-4a82-9b46-3732f1b981e7" -> (known after apply)
        # (2 unchanged attributes hidden)

      - cluster {
          - apply_policy_default_values  = false -> null
          - custom_tags                  = {} -> null
          - driver_node_type_id          = "r5.xlarge" -> null
          - enable_local_disk_encryption = false -> null
          - label                        = "default" -> null
          - node_type_id                 = "m4.large" -> null
          - num_workers                  = 1 -> null
          - spark_conf                   = {} -> null
          - spark_env_vars               = {} -> null
          - ssh_public_keys              = [] -> null
            # (3 unchanged attributes hidden)
        }

      ~ latest_updates (known after apply)
      - latest_updates {
          - creation_time = "2025-01-21T08:22:08.210Z" -> null
          - state         = "WAITING_FOR_RESOURCES" -> null
          - update_id     = "315db56d-b8f3-4448-b366-9c795a872bdd" -> null
        }

        # (1 unchanged block hidden)
    }

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