Terraform AWS ECS service stack creation

This module is for whole ECS service stack creation: service, task definition, container definition, alb listener rule, target group, route53 record, security group etc.

Important note:

Use connect_to_lb and service_domain to connect service container to load balancer and create route53 A record
Use vpc_cidr_block, route_53_zone_name, lb_dns_name only when you dont have previously created resources

Example

Single container

module "ecs_service" {
  source = "zahornyak/ecs-service/aws"

  environment        = "production"
  vpc_id             = "vpc-080fd3099892"
  vpc_cidr_block     = "10.0.0.0/16" # use when you dont have previously created vpc
  service_subnets    = ["subnet-0c264c7154cb", "subnet-09e0d8b22e2"]
  # assign_public_ip = true # if you are using public subnets
  cluster_name       = "production-cluster"
  route_53_zone_id   = "Z01006347593463S0ZFL7A2" # use when you dont have previously created Route53 zone
  route_53_zone_name = "example.com" 
  lb_arn             = "arn:aws:elasticloadbalancing:eu-central-1:1234567890:loadbalancer/app/plugin-development-alb/46555556595fd4b2"
  lb_listener_arn    = "arn:aws:elasticloadbalancing:eu-central-1:1234567890:listener/app/plugin-development-alb/46555556595fd4b2/83d6940f8c9f02db"
  lb_dns_name        = "my-loadbalancer-1234567890.us-west-2.elb.amazonaws.com" # use when you dont have previously created load balancer
  create_ssl         = true # requests ssl for service and attach it to listener rule

  service_name  = "backend"
  desired_count = 1

  container_definitions = {
    proxy = {
      service_domain   = "api-test"
      connect_to_lb    = true
      container_image  = "nginx:latest"
      container_name   = "backend"
      container_cpu    = 256
      container_memory = 256
      containerPort    = 80
      environment      = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
  }

  service_memory = 1024
  service_cpu    = 512
}

Multiple containers

module "ecs_service" {
  source  = "zahornyak/ecs-service/aws"

  environment     = "production"
  vpc_id          = "vpc-080fd3099892"
  service_subnets = ["subnet-0c264c7154cb", "subnet-09e0d8b22e2"]
  # assign_public_ip = true # if you are using public subnets
  cluster_name     = "production-cluster"
  route_53_zone_id = "Z01006347593463S0ZFL7A2"
  lb_arn           = "arn:aws:elasticloadbalancing:eu-central-1:1234567890:loadbalancer/app/plugin-development-alb/46555556595fd4b2"
  lb_listener_arn  = "arn:aws:elasticloadbalancing:eu-central-1:1234567890:listener/app/plugin-development-alb/46555556595fd4b2/83d6940f8c9f02db"
  create_ssl       = true # requests ssl for service and attach it to listener rule

  service_name  = "backend"
  desired_count = 1

  container_definitions = {
    proxy = {
      service_domain   = "api-test"
      connect_to_lb    = true
      container_image  = "nginx:latest"
      container_name   = "proxy"
      container_cpu    = 256
      container_memory = 256
      containerPort    = 80
      environment = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
    
    backend = {
      container_image  = "nginx:latest"
      container_name   = "backend"
      container_cpu    = 256
      container_memory = 256
      container_depends_on = [
        {
          containerName = "proxy"
          condition     = "START"
        }
      ]
      containerPort = 3000
      healthcheck = {
        retries     = 5
        command     = ["CMD-SHELL", "curl -f http://localhost:3000"]
        timeout     = 15
        interval    = 30
        startPeriod = 10
      }
      environment = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
    
    admin = {
      service_domain   = "api-worker"
      connect_to_lb    = true
      container_image  = "nginx:latest"
      container_name   = "worker"
      container_cpu    = 256
      container_memory = 256
      container_depends_on = [
        {
          containerName = "backend"
          condition     = "START"
        }
      ]
      containerPort = 3050
      healthcheck = {
        retries     = 5
        command     = ["CMD-SHELL", "curl -f http://localhost:3050"]
        timeout     = 15
        interval    = 30
        startPeriod = 10
      }
      environment = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
    }
  service_memory  = 1024
  service_cpu     = 512
}

No load balancer

module "ecs_service" {
  source = "zahornyak/ecs-service/aws"

  environment     = var.environment
  vpc_id          = var.vpc_id
  service_subnets = var.subnets
  vpc_cidr_block  = var.vpc_cidr_block

  # assign_public_ip = true # if you are using public subnets
  cluster_name = aws_ecs_cluster.ecs_cluster.name

  service_name  = "backend"
  desired_count = 1

  container_definitions = {
    proxy = {
      container_image  = "nginx:latest"
      container_name   = "proxy"
      container_cpu    = 256
      container_memory = 256
      containerPort    = 80
      environment      = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
    
    backend = {
      container_image      = "nginx:latest"
      container_name       = "backend"
      container_cpu        = 256
      container_memory     = 256
      container_depends_on = [
        {
          containerName = "proxy"
          condition     = "START"
        }
      ]
      containerPort = 3000
      healthcheck   = {
        retries     = 5
        command     = ["CMD-SHELL", "curl -f http://localhost:3000"]
        timeout     = 15
        interval    = 30
        startPeriod = 10
      }
      environment = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
    }
  }

  service_memory = 1024
  service_cpu    = 512
}

Example of using environment valiables for containers(using ssm_secrets which creates ssm parameters and puts them into container definition)

module "ecs-service" {
  source  = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here
  container_definitions = {
    proxy = {
      container_image  = "nginx:latest"
      container_name   = "proxy"
      container_cpu    = 256
      container_memory = 256
      containerPort    = 80
      environment      = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
      ssm_secrets = {
        DEBUG = {
          value = "true"
        }
      }
    }
  }
}

Example of using environment valiables for containers(using ssm_env_file which parses and creates ssm parameters and puts them into container definition)

module "ecs-service" {
  source  = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here
  container_definitions = {
    proxy = {
      container_image  = "nginx:latest"
      container_name   = "proxy"
      container_cpu    = 256
      container_memory = 256
      containerPort    = 80
      environment      = [
        {
          "name"  = "foo"
          "value" = "bar"
        }
      ]
      ssm_env_file = "./env"
    }
  }
}

.env example

LOG_LEVEL=verbose
LOG_TARGET=console
LOG_FORMAT=json

CRONJOB_ENABLED=true
DEPLOYMENT=develop

Autoscaling with scaling values

module "ecs-service" {
  source = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here

  min_service_tasks = 1
  max_service_tasks = 6

  cpu_scaling_target_value = 40
  cpu_scale_in_cooldown    = 350
  cpu_scale_out_cooldown   = 200

  memory_scaling_target_value = 90
  memory_scale_in_cooldown    = 350
  memory_scale_out_cooldown   = 300
}

Autoscaling with scaling values (no memory or cpu scaling)

module "ecs-service" {
  source = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here

  min_service_tasks = 1
  max_service_tasks = 6

  cpu_scaling_target_value = 40
  cpu_scale_in_cooldown    = 350
  cpu_scale_out_cooldown   = 200
  
}

Capacity provider strategy, ordered placement, placement_constraints strategy example configuration

module "ecs-service" {
  source = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here

  capacity_provider_strategy = {
    main = {
      capacity_provider = "FARGATE_SPOT"
      base              = 1
      weight            = 1
    }
  }


  ordered_placement_strategy = {
    test = {
      type  = "binpack"
      field = "cpu"
    }
  }


  placement_constraints = {
    example = {
      type       = "memberOf"
      expression = "attribute:ecs.availability-zone in [us-west-2a, us-west-2b]"
    }
  }
}

Service discovery example

module "ecs-service" {
  source = "zahornyak/ecs-service/aws"
  # insert the 7 required variables here

  create_service_discovery = true
  discovery_registry_id    = "service_discovery_registry_id"
  
}

Requirements

Name	Version
terraform	>= 1.4
aws	>= 4.37

Providers

Name	Version
aws	>= 4.37

Modules

Name	Source	Version
acm	terraform-aws-modules/acm/aws	~> 3.3
ecs_task_exec_policy	terraform-aws-modules/iam/aws//modules/iam-policy	~> 4.4
ecs_task_execution_role	terraform-aws-modules/iam/aws//modules/iam-assumable-role	~> 4.4
ecs_task_policy	terraform-aws-modules/iam/aws//modules/iam-policy	~> 4.4
ecs_task_role	terraform-aws-modules/iam/aws//modules/iam-assumable-role	~> 4.4
env_variables	zahornyak/multiple-ssm-parameters/aws	0.0.11
service_container_definition	registry.terraform.io/cloudposse/ecs-container-definition/aws	~> 0.58
service_container_sg	registry.terraform.io/terraform-aws-modules/security-group/aws	~> 4.3

Resources

Name	Type
aws_appautoscaling_policy.target_tracking_scaling_cpu_service	resource
aws_appautoscaling_policy.target_tracking_scaling_memory_service	resource
aws_appautoscaling_target.service_scaling	resource
aws_cloudwatch_log_group.service_logs	resource
aws_ecs_service.service	resource
aws_ecs_task_definition.service	resource
aws_lb_listener_certificate.this	resource
aws_lb_listener_rule.service	resource
aws_lb_target_group.service	resource
aws_route53_record.lb_records	resource
aws_service_discovery_service.service	resource
aws_caller_identity.current	data source
aws_iam_policy_document.ecs_task_exec_policy	data source
aws_iam_policy_document.ecs_task_policy	data source
aws_lb.this	data source
aws_region.current	data source
aws_route53_zone.this	data source
aws_vpc.this	data source

Inputs

Name	Description	Type	Default	Required
assign_public_ip	Assign_public_ip set true if you are using public subnets.	`bool`	`false`	no
capacity_provider_strategy	capacity_provider_strategy	`any`	`{}`	no
cluster_name	Name of the ECS Cluster.	`string`	n/a	yes
container_definitions	Custom container definitions.	`any`	`{}`	no
cpu_scale_in_cooldown	cpu scale_in_cooldown	`number`	`null`	no
cpu_scale_out_cooldown	cpu scale_out_cooldown	`number`	`null`	no
cpu_scaling_target_value	cpu_scaling target_value	`number`	`null`	no
create_service_discovery	creates service discovery service and connects in to ecs service	`bool`	`false`	no
create_ssl	defines if create ssl for services domains	`bool`	`true`	no
deployment_circuit_breaker	deployment_circuit_breaker configuration	`any`	`{}`	no
deployment_maximum_percent	deployment_maximum_percent. For example 200 will create twice more container and if everything is ok, deployment is succesfull.	`number`	`200`	no
deployment_minimum_healthy_percent	deployment_minimum_healthy_percent.	`number`	`100`	no
deregistration_delay	Deregistration delay for target group.	`number`	`5`	no
desired_count	Desired count for service.	`number`	`null`	no
discovery_registry_id	service discovery registry_id	`string`	`null`	no
docker_volume	docker volume	`any`	`null`	no
efs_volume	efs volume	`any`	`null`	no
environment	Environment name. For example 'production'	`string`	n/a	yes
external_dns	when you dont have route53 zone, you can use external dns	`any`	`null`	no
health_check	Custom healthcheck for target group.	`any`	`null`	no
health_check_grace_period_seconds	health_check_grace_period_seconds	`number`	`null`	no
launch_type	Launch type for service: 'FARGATE', 'EC2' etc.	`string`	`"FARGATE"`	no
lb_arn	Load balancer arn.	`string`	`null`	no
lb_dns_name	Load balancer dns name. Use only if you dont have previously created Load Balancer	`string`	`null`	no
lb_listener_arn	Listener arn for load balancer connection	`string`	`null`	no
lb_zone_id	load balancer zone id	`string`	`null`	no
max_service_tasks	Maximum service tasks.	`number`	`null`	no
memory_scale_in_cooldown	memory scale_in_cooldown	`number`	`null`	no
memory_scale_out_cooldown	memory scale_out_cooldown	`number`	`null`	no
memory_scaling_target_value	memory scaling_target_value	`number`	`null`	no
min_service_tasks	Minimum service tasks.	`number`	`null`	no
network_mode	Network mode for task. For example 'awsvpc' or 'bridge' etc.	`string`	`"awsvpc"`	no
ordered_placement_strategy	ordered_placement_strategy	`any`	`{}`	no
parameter_prefix	prefix for parameter store parameter. For example '/develop/service/'. So parameter 'DEBUG' will have '/develop/service/DEBUG' name on the parameter store	`string`	`null`	no
placement_constraints	placement_constraints	`any`	`{}`	no
protocol_version	target group protocol version	`string`	`null`	no
requires_compatibilities	Compatibilities for ECS task. Available: 'FARGATE', 'FARGATE_SPOT', 'EC2' etc.	`list(string)`	[ "FARGATE" ]	no
retention_in_days	retention_in_days	`number`	`60`	no
route_53_zone_id	Route 53 zone id.	`string`	`null`	no
route_53_zone_name	route 53 zone name. Use only when you dont have previously created Route53 zone	`string`	`null`	no
runtime_platform	runtime platform	`any`	`null`	no
security_groups	additional security_groups for service	`list(string)`	`[]`	no
service_cpu	CPU amount for the service.	`number`	n/a	yes
service_memory	Memory amount for the service.	`number`	n/a	yes
service_name	Name of the service.	`string`	n/a	yes
service_subnets	Subnets for service	`list(string)`	n/a	yes
task_exec_role_policy_arns	Additional policies to attach to task execution role of ECS container.	`list(string)`	`[]`	no
task_role_policy_arns	Additional policies to attach to task role of ECS container.	`list(string)`	`[]`	no
tg_protocol	target group protocol(for example 'HTTP' or 'TCP')	`string`	`"HTTP"`	no
tg_target_type	target group target type(ip or instance etc)	`string`	`"ip"`	no
vpc_cidr_block	cidr block for vpc. Use that variable when you dont have previously created VPC	`string`	`null`	no
vpc_id	VPC id.	`string`	n/a	yes

Outputs

Name	Description
acm_arn	acm arn
cloudwatch_log_group_arns	aws cloudwatch log group arns
container_definitions	container definitions of your task definition
ecs_service_arn	ecs_service_arn
ecs_service_name	ecs service name
ecs_service_security_group_ids	ecs service security group ids
ecs_task_definition_arn	task definition arn
ecs_task_execution_role_arn	ecs task execution role arn
ecs_task_policy_arn	ecs task policy arn
ecs_task_role_arn	ecs task role arn
lb_listener_certificate	lb listener certificate
lb_listener_rule_arns	load balancer listener rules arns
records_lb_names	load balancers records names
service_container_sg_ids	service container sg ids
target_group_arns	target group arns

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

README.md

README.md

Terraform AWS ECS service stack creation

Important note:

Example

Single container

Multiple containers

No load balancer

Example of using environment valiables for containers(using ssm_secrets which creates ssm parameters and puts them into container definition)

Example of using environment valiables for containers(using ssm_env_file which parses and creates ssm parameters and puts them into container definition)

Autoscaling with scaling values

Autoscaling with scaling values (no memory or cpu scaling)

Capacity provider strategy, ordered placement, placement_constraints strategy example configuration

Service discovery example

Requirements

Providers

Modules

Resources

Inputs

Outputs

Files

README.md

Latest commit

History

README.md

File metadata and controls

Terraform AWS ECS service stack creation

Important note:

Example

Single container

Multiple containers

No load balancer

Example of using environment valiables for containers(using ssm_secrets which creates ssm parameters and puts them into container definition)

Example of using environment valiables for containers(using ssm_env_file which parses and creates ssm parameters and puts them into container definition)

Autoscaling with scaling values

Autoscaling with scaling values (no memory or cpu scaling)

Capacity provider strategy, ordered placement, placement_constraints strategy example configuration

Service discovery example

Requirements

Providers

Modules

Resources

Inputs

Outputs