cloudwatch_cpu_usage.tf

resource "aws_cloudwatch_metric_alarm" "cpu-utilization" {
  alarm_name                = "high-cpu-utilization-alarm"
  comparison_operator       = "GreaterThanOrEqualToThreshold"
  evaluation_periods        = "2"
  metric_name               = "CPUUtilization"
  namespace                 = "AWS/EC2"
  period                    = "120"
  statistic                 = "Average"
  threshold                 = "80"
  alarm_description         = "This metric monitors ec2 cpu utilization"
  alarm_actions             = [ "${aws_sns_topic.alarm.arn}" ]
dimensions {
    InstanceId = "${aws_instance.my_instance.*.id[count.index}"
  }
}