@aws-cdk/aws-ecs-patterns

The CDK Construct Library for AWS::ECS

Usage no npm install needed!

<script type="module">
  import awsCdkAwsEcsPatterns from 'https://cdn.skypack.dev/@aws-cdk/aws-ecs-patterns';
</script>

README

CDK Construct library for higher-level ECS Constructs

cdk-constructs: Stable

This library provides higher-level Amazon ECS constructs which follow common architectural patterns. It contains:

  • Application Load Balanced Services
  • Network Load Balanced Services
  • Queue Processing Services
  • Scheduled Tasks (cron jobs)
  • Additional Examples

Application Load Balanced Services

To define an Amazon ECS service that is behind an application load balancer, instantiate one of the following:

  • ApplicationLoadBalancedEc2Service
declare const cluster: ecs.Cluster;
const loadBalancedEcsService = new ecsPatterns.ApplicationLoadBalancedEc2Service(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('test'),
    environment: {
      TEST_ENVIRONMENT_VARIABLE1: "test environment variable 1 value",
      TEST_ENVIRONMENT_VARIABLE2: "test environment variable 2 value",
    },
  },
  desiredCount: 2,
});
  • ApplicationLoadBalancedFargateService
declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
});

loadBalancedFargateService.targetGroup.configureHealthCheck({
  path: "/custom-health-path",
});

Instead of providing a cluster you can specify a VPC and CDK will create a new ECS cluster. If you deploy multiple services CDK will only create one cluster per VPC.

You can omit cluster and vpc to let CDK create a new VPC with two AZs and create a cluster inside this VPC.

You can customize the health check for your target group; otherwise it defaults to HTTP over port 80 hitting path /.

Fargate services will use the LATEST platform version by default, but you can override by providing a value for the platformVersion property in the constructor.

Fargate services use the default VPC Security Group unless one or more are provided using the securityGroups property in the constructor.

By setting redirectHTTP to true, CDK will automatically create a listener on port 80 that redirects HTTP traffic to the HTTPS port.

If you specify the option recordType you can decide if you want the construct to use CNAME or Route53-Aliases as record sets.

If you need to encrypt the traffic between the load balancer and the ECS tasks, you can set the targetProtocol to HTTPS.

Additionally, if more than one application target group are needed, instantiate one of the following:

  • ApplicationMultipleTargetGroupsEc2Service
// One application load balancer with one listener and two target groups.
declare const cluster: ecs.Cluster;
const loadBalancedEc2Service = new ecsPatterns.ApplicationMultipleTargetGroupsEc2Service(this, 'Service', {
  cluster,
  memoryLimitMiB: 256,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  targetGroups: [
    {
      containerPort: 80,
    },
    {
      containerPort: 90,
      pathPattern: 'a/b/c',
      priority: 10,
    },
  ],
});
  • ApplicationMultipleTargetGroupsFargateService
// One application load balancer with one listener and two target groups.
declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationMultipleTargetGroupsFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  targetGroups: [
    {
      containerPort: 80,
    },
    {
      containerPort: 90,
      pathPattern: 'a/b/c',
      priority: 10,
    },
  ],
});

Network Load Balanced Services

To define an Amazon ECS service that is behind a network load balancer, instantiate one of the following:

  • NetworkLoadBalancedEc2Service
declare const cluster: ecs.Cluster;
const loadBalancedEcsService = new ecsPatterns.NetworkLoadBalancedEc2Service(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('test'),
    environment: {
      TEST_ENVIRONMENT_VARIABLE1: "test environment variable 1 value",
      TEST_ENVIRONMENT_VARIABLE2: "test environment variable 2 value",
    },
  },
  desiredCount: 2,
});
  • NetworkLoadBalancedFargateService
declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.NetworkLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
});

The CDK will create a new Amazon ECS cluster if you specify a VPC and omit cluster. If you deploy multiple services the CDK will only create one cluster per VPC.

If cluster and vpc are omitted, the CDK creates a new VPC with subnets in two Availability Zones and a cluster within this VPC.

If you specify the option recordType you can decide if you want the construct to use CNAME or Route53-Aliases as record sets.

Additionally, if more than one network target group is needed, instantiate one of the following:

  • NetworkMultipleTargetGroupsEc2Service
// Two network load balancers, each with their own listener and target group.
declare const cluster: ecs.Cluster;
const loadBalancedEc2Service = new ecsPatterns.NetworkMultipleTargetGroupsEc2Service(this, 'Service', {
  cluster,
  memoryLimitMiB: 256,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  loadBalancers: [
    {
      name: 'lb1',
      listeners: [
        {
          name: 'listener1',
        },
      ],
    },
    {
      name: 'lb2',
      listeners: [
        {
          name: 'listener2',
        },
      ],
    },
  ],
  targetGroups: [
    {
      containerPort: 80,
      listener: 'listener1',
    },
    {
      containerPort: 90,
      listener: 'listener2',
    },
  ],
});
  • NetworkMultipleTargetGroupsFargateService
// Two network load balancers, each with their own listener and target group.
declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.NetworkMultipleTargetGroupsFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  loadBalancers: [
    {
      name: 'lb1',
      listeners: [
        {
          name: 'listener1',
        },
      ],
    },
    {
      name: 'lb2',
      listeners: [
        {
          name: 'listener2',
        },
      ],
    },
  ],
  targetGroups: [
    {
      containerPort: 80,
      listener: 'listener1',
    },
    {
      containerPort: 90,
      listener: 'listener2',
    },
  ],
});

Queue Processing Services

To define a service that creates a queue and reads from that queue, instantiate one of the following:

  • QueueProcessingEc2Service
declare const cluster: ecs.Cluster;
const queueProcessingEc2Service = new ecsPatterns.QueueProcessingEc2Service(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  image: ecs.ContainerImage.fromRegistry('test'),
  command: ["-c", "4", "amazon.com"],
  enableLogging: false,
  desiredTaskCount: 2,
  environment: {
    TEST_ENVIRONMENT_VARIABLE1: "test environment variable 1 value",
    TEST_ENVIRONMENT_VARIABLE2: "test environment variable 2 value",
  },
  maxScalingCapacity: 5,
  containerName: 'test',
});
  • QueueProcessingFargateService
declare const cluster: ecs.Cluster;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  command: ["-c", "4", "amazon.com"],
  enableLogging: false,
  desiredTaskCount: 2,
  environment: {
    TEST_ENVIRONMENT_VARIABLE1: "test environment variable 1 value",
    TEST_ENVIRONMENT_VARIABLE2: "test environment variable 2 value",
  },
  maxScalingCapacity: 5,
  containerName: 'test',
});

when queue not provided by user, CDK will create a primary queue and a dead letter queue with default redrive policy and attach permission to the task to be able to access the primary queue.

Scheduled Tasks

To define a task that runs periodically, there are 2 options:

  • ScheduledEc2Task
// Instantiate an Amazon EC2 Task to run at a scheduled interval
declare const cluster: ecs.Cluster;
const ecsScheduledTask = new ecsPatterns.ScheduledEc2Task(this, 'ScheduledTask', {
  cluster,
  scheduledEc2TaskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('amazon/amazon-ecs-sample'),
    memoryLimitMiB: 256,
    environment: { name: 'TRIGGER', value: 'CloudWatch Events' },
  },
  schedule: appscaling.Schedule.expression('rate(1 minute)'),
  enabled: true,
  ruleName: 'sample-scheduled-task-rule',
});
  • ScheduledFargateTask
declare const cluster: ecs.Cluster;
const scheduledFargateTask = new ecsPatterns.ScheduledFargateTask(this, 'ScheduledFargateTask', {
  cluster,
  scheduledFargateTaskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('amazon/amazon-ecs-sample'),
    memoryLimitMiB: 512,
  },
  schedule: appscaling.Schedule.expression('rate(1 minute)'),
  platformVersion: ecs.FargatePlatformVersion.LATEST,
});

Additional Examples

In addition to using the constructs, users can also add logic to customize these constructs:

Configure HTTPS on an ApplicationLoadBalancedFargateService

import { HostedZone } from '@aws-cdk/aws-route53';
import { Certificate } from '@aws-cdk/aws-certificatemanager';
import { SslPolicy } from '@aws-cdk/aws-elasticloadbalancingv2';

const domainZone = HostedZone.fromLookup(this, 'Zone', { domainName: 'example.com' });
const certificate = Certificate.fromCertificateArn(this, 'Cert', 'arn:aws:acm:us-east-1:123456:certificate/abcdefg');

declare const vpc: ec2.Vpc;
declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  vpc,
  cluster,
  certificate,
  sslPolicy: SslPolicy.RECOMMENDED,
  domainName: 'api.example.com',
  domainZone,
  redirectHTTP: true,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
});

Add Schedule-Based Auto-Scaling to an ApplicationLoadBalancedFargateService

declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
});

const scalableTarget = loadBalancedFargateService.service.autoScaleTaskCount({
  minCapacity: 5,
  maxCapacity: 20,
});

scalableTarget.scaleOnSchedule('DaytimeScaleDown', {
  schedule: appscaling.Schedule.cron({ hour: '8', minute: '0'}),
  minCapacity: 1,
});

scalableTarget.scaleOnSchedule('EveningRushScaleUp', {
  schedule: appscaling.Schedule.cron({ hour: '20', minute: '0'}),
  minCapacity: 10,
});

Add Metric-Based Auto-Scaling to an ApplicationLoadBalancedFargateService

declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
});

const scalableTarget = loadBalancedFargateService.service.autoScaleTaskCount({
  minCapacity: 1,
  maxCapacity: 20,
});

scalableTarget.scaleOnCpuUtilization('CpuScaling', {
  targetUtilizationPercent: 50,
});

scalableTarget.scaleOnMemoryUtilization('MemoryScaling', {
  targetUtilizationPercent: 50,
});

Change the default Deployment Controller

declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  deploymentController: {
    type: ecs.DeploymentControllerType.CODE_DEPLOY,
  },
});

Deployment circuit breaker and rollback

Amazon ECS deployment circuit breaker automatically rolls back unhealthy service deployments without the need for manual intervention. Use circuitBreaker to enable deployment circuit breaker and optionally enable rollback for automatic rollback. See Using the deployment circuit breaker for more details.

declare const cluster: ecs.Cluster;
const service = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  circuitBreaker: { rollback: true },
});

Set deployment configuration on QueueProcessingService

declare const cluster: ecs.Cluster;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  command: ["-c", "4", "amazon.com"],
  enableLogging: false,
  desiredTaskCount: 2,
  environment: {},
  maxScalingCapacity: 5,
  maxHealthyPercent: 200,
  minHealthyPercent: 66,
});

Set taskSubnets and securityGroups for QueueProcessingFargateService

declare const vpc: ec2.Vpc;
declare const securityGroup: ec2.SecurityGroup;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  vpc,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  securityGroups: [securityGroup],
  taskSubnets: { subnetType: ec2.SubnetType.ISOLATED },
});

Define tasks with public IPs for QueueProcessingFargateService

declare const vpc: ec2.Vpc;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  vpc,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  assignPublicIp: true,
});

Define tasks with custom queue parameters for QueueProcessingFargateService

declare const vpc: ec2.Vpc;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  vpc,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  maxReceiveCount: 42,
  retentionPeriod: Duration.days(7),
  visibilityTimeout: Duration.minutes(5),
});

Set capacityProviderStrategies for QueueProcessingFargateService

declare const cluster: ecs.Cluster;
cluster.enableFargateCapacityProviders();

const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  capacityProviderStrategies: [
    {
      capacityProvider: 'FARGATE_SPOT',
      weight: 2,
    },
    {
      capacityProvider: 'FARGATE',
      weight: 1,
    },
  ],
});

Set a custom container-level Healthcheck for QueueProcessingFargateService

declare const vpc: ec2.Vpc;
declare const securityGroup: ec2.SecurityGroup;
const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  vpc,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  healthCheck: {
    command: [ "CMD-SHELL", "curl -f http://localhost/ || exit 1" ],
    // the properties below are optional
    interval: Duration.minutes(30),
    retries: 123,
    startPeriod: Duration.minutes(30),
    timeout: Duration.minutes(30),
  },
});

Set capacityProviderStrategies for QueueProcessingEc2Service

import * as autoscaling from '@aws-cdk/aws-autoscaling';

const vpc = new ec2.Vpc(this, 'Vpc', { maxAzs: 1 });
const cluster = new ecs.Cluster(this, 'EcsCluster', { vpc });
const autoScalingGroup = new autoscaling.AutoScalingGroup(this, 'asg', {
  vpc,
  instanceType: ec2.InstanceType.of(ec2.InstanceClass.BURSTABLE2, ec2.InstanceSize.MICRO),
  machineImage: ecs.EcsOptimizedImage.amazonLinux2(),
});
const capacityProvider = new ecs.AsgCapacityProvider(this, 'provider', {
  autoScalingGroup,
});
cluster.addAsgCapacityProvider(capacityProvider);

const queueProcessingFargateService = new ecsPatterns.QueueProcessingFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 512,
  image: ecs.ContainerImage.fromRegistry('test'),
  capacityProviderStrategies: [
    {
      capacityProvider: capacityProvider.capacityProviderName,
    },
  ],
});

Select specific vpc subnets for ApplicationLoadBalancedFargateService

declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  taskSubnets: {
    subnets: [ec2.Subnet.fromSubnetId(this, 'subnet', 'VpcISOLATEDSubnet1Subnet80F07FA0')],
  },
});

Set PlatformVersion for ScheduledFargateTask

declare const cluster: ecs.Cluster;
const scheduledFargateTask = new ecsPatterns.ScheduledFargateTask(this, 'ScheduledFargateTask', {
  cluster,
  scheduledFargateTaskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('amazon/amazon-ecs-sample'),
    memoryLimitMiB: 512,
  },
  schedule: appscaling.Schedule.expression('rate(1 minute)'),
  platformVersion: ecs.FargatePlatformVersion.VERSION1_4,
});

Set SecurityGroups for ScheduledFargateTask

const vpc = new ec2.Vpc(this, 'Vpc', { maxAzs: 1 });
const cluster = new ecs.Cluster(this, 'EcsCluster', { vpc });
const securityGroup = new ec2.SecurityGroup(this, 'SG', { vpc });

const scheduledFargateTask = new ecsPatterns.ScheduledFargateTask(this, 'ScheduledFargateTask', {
  cluster,
  scheduledFargateTaskImageOptions: {
    image: ecs.ContainerImage.fromRegistry('amazon/amazon-ecs-sample'),
    memoryLimitMiB: 512,
  },
  schedule: appscaling.Schedule.expression('rate(1 minute)'),
  securityGroups: [securityGroup],
});

Use the REMOVE_DEFAULT_DESIRED_COUNT feature flag

The REMOVE_DEFAULT_DESIRED_COUNT feature flag is used to override the default desiredCount that is autogenerated by the CDK. This will set the desiredCount of any service created by any of the following constructs to be undefined.

  • ApplicationLoadBalancedEc2Service
  • ApplicationLoadBalancedFargateService
  • NetworkLoadBalancedEc2Service
  • NetworkLoadBalancedFargateService
  • QueueProcessingEc2Service
  • QueueProcessingFargateService

If a desiredCount is not passed in as input to the above constructs, CloudFormation will either create a new service to start up with a desiredCount of 1, or update an existing service to start up with the same desiredCount as prior to the update.

To enable the feature flag, ensure that the REMOVE_DEFAULT_DESIRED_COUNT flag within an application stack context is set to true, like so:

declare const stack: Stack;
stack.node.setContext(cxapi.ECS_REMOVE_DEFAULT_DESIRED_COUNT, true);

The following is an example of an application with the REMOVE_DEFAULT_DESIRED_COUNT feature flag enabled:

import { App, Stack } from '@aws-cdk/core';
import * as ec2 from '@aws-cdk/aws-ec2';
import * as ecs from '@aws-cdk/aws-ecs';
import * as ecsPatterns from '@aws-cdk/aws-ecs-patterns';
import * as cxapi from '@aws-cdk/cx-api';
import * as path from 'path';

const app = new App();

const stack = new Stack(app, 'aws-ecs-patterns-queue');
stack.node.setContext(cxapi.ECS_REMOVE_DEFAULT_DESIRED_COUNT, true);

const vpc = new ec2.Vpc(stack, 'VPC', {
  maxAzs: 2,
});

new ecsPatterns.QueueProcessingFargateService(stack, 'QueueProcessingService', {
  vpc,
  memoryLimitMiB: 512,
  image: new ecs.AssetImage(path.join(__dirname, '..', 'sqs-reader')),
});

Deploy application and metrics sidecar

The following is an example of deploying an application along with a metrics sidecar container that utilizes dockerLabels for discovery:

declare const cluster: ecs.Cluster;
declare const vpc: ec2.Vpc;
const service = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  vpc,
  desiredCount: 1,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
    dockerLabels: {
      'application.label.one': 'first_label',
      'application.label.two': 'second_label',
    },
  },
});

service.taskDefinition.addContainer('Sidecar', {
  image: ecs.ContainerImage.fromRegistry('example/metrics-sidecar'),
});

Select specific load balancer name ApplicationLoadBalancedFargateService

declare const cluster: ecs.Cluster;
const loadBalancedFargateService = new ecsPatterns.ApplicationLoadBalancedFargateService(this, 'Service', {
  cluster,
  memoryLimitMiB: 1024,
  desiredCount: 1,
  cpu: 512,
  taskImageOptions: {
    image: ecs.ContainerImage.fromRegistry("amazon/amazon-ecs-sample"),
  },
  taskSubnets: {
    subnets: [ec2.Subnet.fromSubnetId(this, 'subnet', 'VpcISOLATEDSubnet1Subnet80F07FA0')],
  },
  loadBalancerName: 'application-lb-name',
});