Beta - check Databricks cluster Azure attribute best practices

This product is not supported for your selected Datadog site. ().

Metadata

Id: 38028698-e663-4ef7-aa92-773fef0ca86f

Cloud Provider: Databricks

Platform: Terraform

Severity: Low

Category: Best Practices

Learn More

Description

One or more Azure attribute best practices are not followed for this Databricks cluster. The rule flags clusters when:

  • azure_attributes.availability is set to SPOT or SPOT_AZURE,
  • azure_attributes.first_on_demand is 0, or
  • azure_attributes.first_on_demand is missing.

These settings must ensure the use of at least one on-demand instance and avoid exclusive reliance on spot VMs.

Compliant Code Examples

resource "databricks_cluster" "negative" {
  cluster_name            = "Shared Autoscaling"
  spark_version           = data.databricks_spark_version.latest.id
  node_type_id            = data.databricks_node_type.smallest.id
  autotermination_minutes = 20
  autoscale {
    min_workers = 1
    max_workers = 50
  }
  azure_attributes {
    availability           = "SPOT_WITH_FALLBACK_AZURE"
    first_on_demand        = 1
    spot_bid_price_percent = 100
  }
}

Non-Compliant Code Examples

resource "databricks_cluster" "positive2" {
  cluster_name            = "data"
  spark_version           = data.databricks_spark_version.latest.id
  node_type_id            = data.databricks_node_type.smallest.id
  autotermination_minutes = 20
  autoscale {
    min_workers = 1
    max_workers = 50
  }
  azure_attributes {
    availability           = "SPOT_WITH_FALLBACK_AZURE"
    first_on_demand        = 0
    spot_bid_price_percent = 100
  }
}

resource "databricks_cluster" "positive3" {
  cluster_name            = "data"
  spark_version           = data.databricks_spark_version.latest.id
  node_type_id            = data.databricks_node_type.smallest.id
  autotermination_minutes = 20
  autoscale {
    min_workers = 1
    max_workers = 50
  }
  azure_attributes {
    availability           = "SPOT_WITH_FALLBACK_AZURE"
    zone_id                = "auto"
    spot_bid_price_percent = 100
  }
}

resource "databricks_cluster" "positive1" {
  cluster_name            = "data"
  spark_version           = data.databricks_spark_version.latest.id
  node_type_id            = data.databricks_node_type.smallest.id
  autotermination_minutes = 20
  autoscale {
    min_workers = 1
    max_workers = 50
  }
  azure_attributes {
    availability           = "SPOT_AZURE"
    first_on_demand        = 1
    spot_bid_price_percent = 100
  }
}