ecs-service-tf

A Terraform module to create an Amazon Web Services (AWS) ECS Service on top of an ECS cluster.

Usage

module "container_service" {
  source = "git::ssh://[email protected]/kgirthofer/ecs_service_terraform?ref=0.1.0"

  cluster_name = "test-cluster"
  
  alb_container_name = "nginx-frontend"
  alb_security_groups = [...]
  health_check_path = "/health_check"
  alb_container_port = 80
  public_subnet_ids = [...]
  alb_internal_bool = true

  custom_url = "whatever"

  service_desired_count = 3
  service_asg_min_cap = 3
  service_asg_max_cap = 12  

  project     = "docker-test-app"
  task_definition = "docker-test-app:1"
  environment = "stg"
}

Variables

• alb_container_name - Name of the container the ALB should route traffic too (Default: unknown)
• alb_container_port - Port the ALB will route traffic to on the container (Default: unknown)
• alb_internal_bool - Boolean flag to set the ALB to internal or public (Default: true)
• custom_url - Prefix for the URL - environment will append. I.e. test will be test-env.whatever.com (Default unknown)
• domain_zone_id - Route53 Hosted Zone ID (Default: unknown)
• service_desired_count - Number of services desired to run - Ignored after initial run (Default: 1)
• region - Where it all happens (Default: us-east-1)
• cpu_threshold_up - Percentage of CPU utilization to trigger scaling up action (Default: 60)
• cpu_up_evaluation_periods - How many periods of alarm until scaling up is triggered (Default: 3)
• cpu_up_time_period - How much time is one period (Default: 60)
• cpu_up_cooldown - How much time inbetween scaling actions in seconds (Default: 240)
• cpu_threshold_down - Percentage of CPU utilization to trigger scaling down action (Default: 20)
• cpu_down_evaluation_periods - How many periods of alarm until scalind down is triggered (Default 1)
• cpu_down_time_period - How much time is one period (Default: 300)
• cpu_down_cooldown - How much time inbetween scalind actions in seconds (Default: 300)
• service_asg_min_cap - Minimum number of services running (Default: 1)
• service_asg_max_cap - Maximum number of services running (Default: 3)
• project - Name of the project or application (Default: unknown)
• cluster_name - Name of the cluster to launch into (Default: unknown)
• task_definition - Full task definition - taskdef:refNum (Default: unknown)
• public_subnet_ids - List of subnet ID's to launch ALB into - format ["subnet-xxxxxxxx", "subnet-xxxxxxxx"] (Default: unknown)
• alb_security_groups - List of security groups to attach to the ALB - format ["sg-xxxxxxxx", "sg-xxxxxxxx"] (Default: unknown)
• environment - Service environment (Default: unknown)
• health_check_path - Path for the ALB to register target health (Default: unknown)
• port - ALB exposed port (Default: 80)
• vpc_id - VPC to launch into (Default: unknown)
• ecs_service_role_name - Service role needed to assuming permissions to adjust ECS cluster, ALB, and others (Default: unknown)

Outputs

• name - The container service Name
• lb_dns_name - The raw DNS name for the ELB (A Record)