Implementation

• Set up Environment Variables

  1. Open the new AWS Cloud9 terminal window as per the instructions listed in Module 2.

  2. Configure the AWS CLI with your current region as default (you can retrieve the region you chose from instance metadata).

  export AWS_REGION=$(curl -s http://169.254.169.254/latest/meta-data/placement/region)
  aws configure set default.region ${AWS_REGION}
  aws configure get default.region

  3. Set up environment variable for AWS account id, for later reuse.

  export AWS_ACCOUNT_ID=$(aws sts get-caller-identity --query Account --output text)
  echo $AWS_ACCOUNT_ID

  In subsequent steps you will be referring these values, so storing them in the environment variables takes away the need of copying and pasting the values, or writing long command line instructions.
  

• Create ECR Repository

  1. Create an Amazon ECR repository to store movie-app docker image, set up environment variable for the Repo URI, and note the Repo URI in the output. You can choose any name for your repository, this guide suggests using name as movie-app-repo.

  export IMAGE_REPO_URI=$( \
  aws ecr create-repository \
  --repository-name movie-app-repo \
  --query repository.repositoryUri \
  --output text \
  )
  echo Repo URI: $IMAGE_REPO_URI

  The result should look something like this:
  

  

  Keep this information on a separate notepad as you’ll need this image repository URI when you specify container definition in your Amazon ECS Task Definition.
  

• Create Docker Image

  In this step you’ll create a Dockerfile and then build a docker image for your sample application.

  1. Move to the source code directory.
  

  cd ~/environment/amazon-ecs-dotnet-app-graviton/app/

  2. Create Dockerfile in the root of the project.
  

  touch Dockerfile

  3. Open Dockerfile, either in Cloud9 IDE or via command-line text editor such as nano or vi, then copy the commands listed below and paste them into the file, and then save the file. Each command has an inline documentation to explain the purpose of the command.

  FROM mcr.microsoft.com/dotnet/sdk:5.0 AS build
  WORKDIR /src
  
  # These lines copy MvcMovie.csproj project files in the current directory 
  # and then restore for the specified runtime, using -r.
  COPY ["MvcMovie.csproj", "./"]
  RUN dotnet restore "MvcMovie.csproj" -r linux-musl-arm64
  
  # Build dotnet project and publishes the application and its dependencies to a folder for deployment to a hosting system.
  COPY . .
  WORKDIR "/src/."
  RUN dotnet publish "MvcMovie.csproj" -c Release -o /app/publish -r linux-musl-arm64 --self-contained false --no-restore
  
  # Docker build command
  # You need to use an architecture-specific tag if you want to target ARM based instances. 
  # Note that .NET is only supported on Alpine on ARM64 and x64, and not ARM32.
  FROM mcr.microsoft.com/dotnet/aspnet:5.0-alpine-arm64v8
  WORKDIR /app
  EXPOSE 80
  COPY --from=build /app/publish .
  ENTRYPOINT ["dotnet", "MvcMovie.dll"]

  4. Build and publish docker image with tag movie-app. Tag your image with the Amazon ECR registry, repository, and optional image tag name combination to use. For detail, you can refer here later. The commands below can take a few minutes.

  docker build -t movie-app .
  docker tag movie-app $IMAGE_REPO_URI

  5. Confirm the image is created successfully.

  docker images movie-app

  The command line outputs should look something like below:
  

  

• Publish Docker Image

  In this step you’ll login to your Amazon ECR Repository, and then publish the new image you built, in the last step.

  1. To be able to push the docker image to the Amazon ECR, you’ll login into the default private Amazon Elastic Container Registry. The URL for your default private registry (replacing the values from environment variables you created earlier) is: https://$AWS_ACCOUNT_ID.dkr.ecr.$AWS_REGION.amazonaws.com.
  

  aws ecr get-login-password | docker login --username AWS --password-stdin $AWS_ACCOUNT_ID.dkr.ecr.$AWS_REGION.amazonaws.com

  2. Finally, you’ll push the docker image you created, to the repository, via following command.

  docker push $IMAGE_REPO_URI
  aws ecr describe-images --repository-name movie-app-repo

  Output should look like this: