Enhance Monitoring disabled in RDS auto scaling read instance

[gmattar]

I created an Aurora Postgres DB and enabled enhance monitoring (EM). It worked well for a single instance dev DB but not for the autoscaling production version. In this case, the write instance gets the EM enabled but the autoscaled write instance don't.
I checked the aws_appautoscaling terraform docs for any missing parameter on my. side but couldn't find (only role_arn but that seems to be another thing).

this is a simplified version of the module

terraform {
  required_version = "= 0.12.28"
}

provider "aws" {
  version = "= 2.70.0"
}

locals {
  rds_enhanced_monitoring_arn  = join("", aws_iam_role.rds_enhanced_monitoring.*.arn)
  rds_enhanced_monitoring_name = join("", aws_iam_role.rds_enhanced_monitoring.*.name)
}

//
// Create the database
//

data "aws_iam_policy_document" "monitoring_rds_assume_role" {
  statement {
    actions = ["sts:AssumeRole"]

    principals {
      type        = "Service"
      identifiers = ["monitoring.rds.amazonaws.com"]
    }
  }
}

resource "aws_iam_role" "rds_enhanced_monitoring" {
  name               = "rds-enhanced-monitoring"
  assume_role_policy = data.aws_iam_policy_document.monitoring_rds_assume_role.json
  tags               = local.tags
}

resource "aws_iam_role_policy_attachment" "rds_enhanced_monitoring" {
  role       = local.rds_enhanced_monitoring_name
  policy_arn = "arn:aws:iam::aws:policy/service-role/AmazonRDSEnhancedMonitoringRole"
}

resource "aws_rds_cluster" "default" {
  cluster_identifier                  = local.name
  deletion_protection                 = var.deletion_protection
  engine                              = "aurora-postgresql"
  engine_version                      = "11.6"
  storage_encrypted                   = true
  apply_immediately                   = true
  enabled_cloudwatch_logs_exports     = ["postgresql"]

  lifecycle {
    create_before_destroy = true
  }
}

resource "aws_rds_cluster_instance" "default" {
  monitoring_interval          = var.monitoring_interval
  performance_insights_enabled = true
  monitoring_role_arn          = local.rds_enhanced_monitoring_arn
}

resource "aws_appautoscaling_target" "read_replica_count" {
  min_capacity       = var.replica_scale_min
  max_capacity       = var.replica_scale_max
  resource_id        = "cluster:${aws_rds_cluster.default.cluster_identifier}"
  scalable_dimension = "rds:cluster:ReadReplicaCount"
  service_namespace  = "rds"
  role_arn           = local.rds_enhanced_monitoring_arn
}

resource "aws_appautoscaling_policy" "autoscaling_read_replica_count" {
  name               = "target-metric"
  policy_type        = "TargetTrackingScaling"
  resource_id        = "cluster:${aws_rds_cluster.default.cluster_identifier}"
  scalable_dimension = "rds:cluster:ReadReplicaCount"
  service_namespace  = "rds"

  target_tracking_scaling_policy_configuration {
    predefined_metric_specification {
      predefined_metric_type = var.predefined_metric_type
    }

    scale_in_cooldown  = var.replica_scale_in_cooldown
    scale_out_cooldown = var.replica_scale_out_cooldown
    target_value       = var.predefined_metric_type == "RDSReaderAverageCPUUtilization" ? var.replica_scale_cpu : var.replica_scale_connections
  }

  depends_on = [aws_appautoscaling_target.read_replica_count]
}

[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!