Create a pipeline with an Amazon ECR source and ECS-to-CodeDeploy deployment

Step 1: Create image and push to an Amazon ECR repository

docker pull nginx
docker images
aws ecr create-repository --repository-name nginx --region ap-southeast-1 --profile global

docker tag nginx:latest aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/nginx:latest
aws ecr get-login-password --region ap-southeast-1 --profile global | docker login --username AWS --password-stdin aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/nginx
docker push aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/nginx:latest

Step 2: Create task definition and AppSpec source files and push to a CodeCommit repository

1. Replace the placeholder to `nginx` in taskdef.json

aws ecs register-task-definition --cli-input-json file://taskdef.json --region ap-southeast-1 --profile global

1. After the task definition is registered, edit your file to replace the nginx back to <IMAGE1_NAME> placeholder text in the image field.

2. To create an AppSpec file appspec.yaml

Step 3: Create ECS resources

1. Create your Application Load Balancer and target groups
2. Create your Amazon ECS cluster and service

   aws ecs create-service --service-name my-service --cli-input-json file://create-service.json --region ap-southeast-1 --profile global
   aws ecs describe-services --cluster ecs-demo --services my-service --region ap-southeast-1 --profile global
   

Step 4: Create your CodeDeploy application and deployment group

1. Create Application: MyECSDemoApplication with ECS compute platform
2. Create a CodeDeploy deployment group
   • Name: ecs-deployment
   • Service Role: MyCodeDeployRole
   • ECS cluster: ecs-demo
   • ECS service name: my-service
   • LoadBalancer: ecs-demo-alb, production listener port: 80, test listener port: 8080, TargetGroup 1: ecs-demo-alb-blue-tg (for 80), TargetGroup 2: ecs-demo-alb-gree-tg (for 8080)
   • Choose Reroute traffic immediately

Step 5: Create your pipeline

• pipeline name: ECS-pipeline
• Source Stage: Provider: CodeCommit, Repository: MyECSDemo, Branch: master
• Skip build stage
• Deploy stage
  • Deployment provider, choose Amazon ECS (Blue/Green)
  • Cluster name: ECSDemo
  • Service name: nginx-service
• Deploy stage action
  • action name: ECSBlueGreenDeploy
  • action provider: Amazon ECS (Blue/Green)
  • Input artifacts: SourceArtifact
  • AWS CodeDeploy application name: MyECSDemoApplication
  • AWS CodeDeploy deployment group: ecs-deployment
  • Amazon ECS task definition: SourceArtifact - taskdef.json
  • AWS CodeDeploy AppSpec file: SourceArtifact - appspec.yaml
• Add an Amazon ECR source action to your pipeline Source Stage
  • Name: MyNginxImage
  • Provider: ECR, Repository: nginx
  • Output artifacts: MyNginxImage
• Wire your source artifacts to the deploy action
  • Edit ECSBlueGreenDeploy action
  • Add new Input artifacts: ECRImage
  • Dynamically update task definition image
    • Input artifact with image details: MyNginxImage
    • Placeholder text in the task definition: IMAGE1_NAME

Step 6: Verify your pipeline ran successfully

1. Verify your pipeline ran successfully
2. Access the Public DNS of your ALB in your browser to view sample ECS application (for both 80 and 8080)

Step 7: Add a Build Specification File to Your Source Repository

1. Create CodeCommit Repository with name MyECSDemo
2. Checkin the buildspec.yml

   git clone ssh://git-codecommit.ap-southeast-1.amazonaws.com/v1/repos/MyECSDemo
   git add .
   git commit -m "Adding container build specification."
   git push
   

Step 8: Creating ECS resources

1. Create the Task Definition for hello-world with hello-world ECR image and map to 9080 port

   aws ecr create-repository --repository-name hello-world --region ap-southeast-1 --profile global
   cd webpage-demo
   docker build . -t hello-world:latest
   docker tag hello-world:latest aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/hello-world:latest
   aws ecr get-login-password --region ap-southeast-1 --profile global | docker login --username AWS --password-stdin aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/hello-world
   docker push aws_account_id.dkr.ecr.ap-southeast-1.amazonaws.com/hello-world:latest
   

2. Create the Service hello-world under ecs-demo Cluster

   aws ecs create-service --service-name hello-world --cli-input-json file://create-service-hello-world.json --region ap-southeast-1 --profile global
   aws ecs describe-services --cluster ecs-demo --services my-service --region ap-southeast-1 --profile global
   

Step 8: Creating the Continuous Deployment Pipeline

• pipeline name: ECS-pipeline
• Source: Provider: CodeCommit, Repository: MyECSDemo, Branch: master
• Skip build stage
• Build stage page, for Build provider choose AWS CodeBuild, and then choose Create project.
  • Project name: ECS-Nginx
  • Environment image, choose Managed image.
  • Operating system, choose Amazon Linux 2.
  • Runtime(s), choose Standard.
  • Image, choose aws/codebuild/amazonlinux2-x86_64-standard:3.0.
  • Image version and Environment type, use the default values.
  • Privileged: Select Enable this flag if you want to build Docker images or want your builds to get elevated privileges.
  • Service Role: MyCodeBuildRole (with AmazonEC2ContainerRegistryPowerUser permission)
  • Builspec: Using the buildspec.yml in the webpage-demo directory
  • CloudWatch logs: /aws/codebuild/codebuilddemo
  • Build type: Single build
• Deploy stage
  • Deployment provider, choose Amazon ECS.
  • Cluster name: ecs-demo
  • Service name: hello-world

Step 9: Verify your pipeline ran successfully

1. Verify your pipeline ran successfully
2. Access the Public DNS of your ALB in your browser to view sample ECS application (9080)

TroubleShooting:

Cannot connect to the Docker daemon