Deploy_to_AWS_ECS.md

Hands-on: Deploy to AWS Elastic Container Service (ECS)

In this hands-on lab you will deploy a simple container to Amazon AWS Elastic Container Services (ECS) using GitHub Actions.

1. Create a Amazon ECR container registry:
   Open the Cloud Shell and log in with your AWS account. Run the following commands and adjust AWS_REGION with your desired region (you can find all regions in the top right corner in AWS). Adjust ECR_REPOSITORYto be unique and replace xyz with your user name.

   $ AWS_REGION=us-east-2

   $ ECR_REPOSITORY=xyz-demo-shop

   $ aws ecr create-repository \
       --repository-name $ECR_REPOSITORY \
       --region $AWS_REGION

2. Fork the repository: Fork this repository (if you haven't done so) by clicking the Fork button at the top right corner.

3. Create access key: In AWS go to My Security Credentials | Access keys| Create New Access Key:

   

   Note the Access Key ID and Secret Access Key:

   

4. Create repository secrets: In your forked repository go to Settings | Secrets and create to new secrets AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY using the values of the previous step.

5. Modify workflow file: Edit .github/workflows/Deploy-AWS.yml and adjust the values under env:. Use AWS_REGION and ECR_REPOSITORY from step 1. If you are a new user on AWS, just replace xyz with your username. If you already have a cluster running, you can use the name here. The service will be created later.

   env:
     AWS_REGION: us-east-2
     ECR_REPOSITORY: xyz-demo-shop
     ECS_SERVICE: xyz-demo-shop-service
     ECS_CLUSTER: default
     ECS_TASK_DEFINITION: ch9_release/aws-task-definition.json
     CONTAINER_NAME: xyz-demo-shop

   Commit the changes.

6. Run the workflow: Manually trigger the workflow.

   

   It is expected that the second job Deploy will fail as you have not jet configured ECS. But Build will work and publish our image to the registry.

   Expand Build, tag, and push image to Amazon ECR and copy the image name:

   

7. Create cluster and ECS: Open the Getting Started with Amazon Elastic Container Service (Amazon ECS) using Fargate wizard (be aware that this URL contains the region!).

   Click Configure on the custom tile.

   

   Enter the 'container name' you set in yaml and the image name you copied in the previous step. Set the port to 80:

   

   Click Next.

   Select the Application Load Balancer and click Next:

   

   Set the Cluster name to the value from the yaml in step 5.

   

   Click `Next' and create the cluster.

   

   Wait until all resource have been created and click View service:

8. Copy and save the task run definition: Click on the task definition link:

   

   Copy all the content from the JSON tab to your clipboard.

   

   Open the file ch9_release/aws-task-definition.json, edit it, and past the content. Commit the file.

9. Check that your container is running: To see that your container is running we have to find the DNS name of the load balancer. Go back to your cluster and open the target group:

   

   Click the load balancer in the group:

   

   In the load balancer you see the DNS name under Description. Copy it:

   

   Paste it into a new browser window or tab. You should see the Tailwind Traders website from our container:

   

10. Run the workflow again: Manually trigger the workflow again.

    

    This time your workflow runs without error and deploys your latest version of the container.