Tutorial: Deploy an application to Amazon EKS - Amazon CodeCatalyst

Tutorial: Deploy an application to Amazon EKS

In this tutorial, you learn how to deploy a containerized application into Amazon Elastic Kubernetes Service using an Amazon CodeCatalyst workflow, Amazon EKS, and a few other AWS services. The deployed application is a simple 'Hello, World!' website built on an Apache web server Docker image. The tutorial walks you through the required preparation work such as setting up a development machine and an Amazon EKS cluster, and then describes how to create a workflow to build the application and deploy it into the cluster.

After the initial deployment is complete, the tutorial instructs you to make a change to your application source. This change causes a new Docker image to be built and pushed to your Docker image repository with new revision information. The new revision of the Docker image is then deployed into Amazon EKS.

Tip

Instead of working your way through this tutorial, you can use a blueprint that does a complete Amazon EKS setup for you. You'll need to use the EKS App Deployment blueprint. For more information, see Creating a project with a blueprint.

Prerequisites

Before you begin this tutorial:

  • You need an Amazon CodeCatalyst space with a connected AWS account. For more information, see Creating a space.

  • In your space, you need an empty project called:

    codecatalyst-eks-project

    Use the Start from scratch option to create this project.

    For more information, see Creating an empty project in Amazon CodeCatalyst.

  • In your project, you need an empty CodeCatalyst source repository called:

    codecatalyst-eks-source-repository

    For more information, see Store and collaborate on code with source repositories in CodeCatalyst.

  • In your project, you need a CodeCatalyst CI/CD environment (not a Dev Environment) called:

    codecatalyst-eks-environment

    Configure this environment as follows:

    • Choose any type, such as Non-production.

    • Connect your AWS account to it.

    • For the Default IAM role, choose any role. You'll specify a different role later.

    For more information, see Deploying into AWS accounts and VPCs.

Step 1: Set up your development machine

The first step in this tutorial is to configure a development machine with a few tools that you'll use throughout this tutorial. These tools are:

  • the eksctl utility – for cluster creation

  • the kubectl utility – a prerequisite for eksctl

  • the AWS CLI – also a prerequisite for eksctl

You can install these tools on your existing development machine if you have one, or you can use a CodeCatalyst Dev Environment, which is Cloud-based. The benefit of a CodeCatalyst Dev Environment is that it's easy to spin up and take down, and is integrated with other CodeCatalyst services, allowing you to work through this tutorial in fewer steps.

This tutorial assumes you'll be using a CodeCatalyst Dev Environment.

The following instructions describe a quick way to launch a CodeCatalyst Dev Environment and configure it with the required tools, but if you want detailed instructions, see:

To launch a Dev Environment
  1. Open the CodeCatalyst console at https://codecatalyst.aws/.

  2. Navigate to your project, codecatalyst-eks-project.

  3. In the navigation pane, choose Code, and then choose Source repositories.

  4. Choose the name of your source repository, codecatalyst-eks-source-repository.

  5. Near the top choose Create Dev Environment, and then choose AWS Cloud9 (in browser).

  6. Make sure that Work in existing branch and main are selected, and then choose Create.

    Your Dev Environment launches in a new browser tab, and your repository (codecatalyst-eks-source-repository) is cloned into it.

To install and configure kubectl
  1. In the Dev Environment terminal, enter:

    curl -o kubectl https://amazon-eks.s3.us-west-2.amazonaws.com/1.18.9/2020-11-02/bin/linux/amd64/kubectl
  2. Enter:

    chmod +x ./kubectl
  3. Enter:

    mkdir -p $HOME/bin && cp ./kubectl $HOME/bin/kubectl && export PATH=$PATH:$HOME/bin
  4. Enter:

    echo 'export PATH=$PATH:$HOME/bin' >> ~/.bashrc
  5. Enter:

    kubectl version --short --client
  6. Check that a version appears.

    You have now installed kubectl.

To install and configure eksctl
Note

eksctl is not strictly required because you can use kubectl instead. However, eksctl has the benefit of automating much of the cluster configuration, and is therefore the tool recommended for this tutorial.

  1. In the Dev Environment terminal, enter:

    curl --silent --location "https://github.com/weaveworks/eksctl/releases/latest/download/eksctl_$(uname -s)_amd64.tar.gz" | tar xz -C /tmp
  2. Enter:

    sudo cp /tmp/eksctl /usr/bin
  3. Enter:

    eksctl version
  4. Check that a version appears.

    You have now installed eksctl.

To verify that the AWS CLI is installed
  1. In the Dev Environment terminal, enter:

    aws --version
  2. Check that a version appears to verify that the AWS CLI is installed.

    Complete the remaining procedures to configure the AWS CLI with the necessary permissions to access AWS.

To configure the AWS CLI

You must configure the AWS CLI with access keys and a session token to give it access to AWS services. The following instructions provide a quick way to configure the keys and token, but if you want detailed instructions, see Configuring the AWS CLI in the AWS Command Line Interface User Guide.

  1. Create an IAM Identity Center user, as follows:

    1. Sign in to the AWS Management Console and open the AWS IAM Identity Center console at https://console.aws.amazon.com/singlesignon/.

      (You might need to choose Enable if you've never signed in to IAM Identity Center before.)

      Note

      Make sure you sign in using the AWS account that is connected to your CodeCatalyst space. You can verify which account is connected by navigating to your space and choosing the AWS accounts tab. For more information, see Creating a space.

    2. In the navigation pane, choose Users, and then choose Add user.

    3. In Username, enter:

      codecatalyst-eks-user
    4. Under Password, choose Generate a one-time password that you can share with this user.

    5. In Email address and Confirm email address, enter an email address that doesn't already exist in IAM Identity Center.

    6. In First name, enter:

      codecatalyst-eks-user
    7. In Last name, enter:

      codecatalyst-eks-user
    8. In Display name, keep:

      codecatalyst-eks-user codecatalyst-eks-user
    9. Choose Next.

    10. On the Add user to groups page, choose Next.

    11. On the Review and add user page, review the information and choose Add user.

      A One-time password dialog box appears.

    12. Choose Copy and then paste the sign-in information to a text file. The sign-in information consists of the AWS access portal URL, a user name, and a one-time password.

    13. Choose Close.

  2. Create a permission set, as follows:

    1. In the navigation pane, choose Permission sets, and then choose Create permission set.

    2. Choose Predefined permission set and then select AdministratorAccess. This policy provides full permissions to all AWS services.

    3. Choose Next.

    4. In Permission set name, remove AdministratorAccess and enter:

      codecatalyst-eks-permission-set
    5. Choose Next.

    6. On the Review and create page, review the information and choose Create.

  3. Assign the permission set to codecatalyst-eks-user, as follows:

    1. In the navigation pane, choose AWS accounts, and then select the check box next to the AWS account that you're currently signed in to.

    2. Choose Assign users or groups.

    3. Choose the Users tab.

    4. Select the check box next to codecatalyst-eks-user.

    5. Choose Next.

    6. Select the check box next to codecatalyst-eks-permission-set.

    7. Choose Next.

    8. Review the information and choose Submit.

      You have now assigned codecatalyst-eks-user and codecatalyst-eks-permission-set to your AWS account, binding them together.

  4. Obtain codecatalyst-eks-user's access keys and session token, as follows:

    1. Make sure you have the AWS access portal URL and the username and one-time password for codecatalyst-eks-user. You should have copied this information to a text editor earlier.

      Note

      If you do not have this information, go to the codecatalyst-eks-user details page in IAM Identity Center, choose Reset password, Generate a one-time password [...], and Reset password again to display the information on the screen.

    2. Sign out of AWS.

    3. Paste the AWS access portal URL into your browser's address bar.

    4. Sign in with:

      • Username:

        codecatalyst-eks-user
      • Password:

        one-time-password

    5. In Set new password, enter a new password and choose Set new password.

      An AWS account box appears on the screen.

    6. Choose AWS account, and then choose the name of the AWS account to which you assigned the codecatalyst-eks-user user and permission set.

    7. Next to codecatalyst-eks-permission-set, choose Command line or programmatic access.

    8. Copy the commands in the middle of the page. They look similar to the following:

      export AWS_ACCESS_KEY_ID="AKIAIOSFODNN7EXAMPLE" export AWS_SECRET_ACCESS_KEY="wJalrXUtnFEMI/K7MDENG/bPxRfiCYEXAMPLEKEY" export AWS_SESSION_TOKEN="session-token"

      ...where session-token is a long random string.

  5. Add the access keys and session token to the AWS CLI, as follows:

    1. Return to your CodeCatalyst Dev Environment.

    2. At the terminal prompt, paste the commands you copied. Press Enter.

      You have now configured the AWS CLI with access keys and a session token. You can now use AWS CLI to complete the tasks required by this tutorial.

      Important

      If at any time during this tutorial you see messages similar to:

      Unable to locate credentials. You can configure credentials by running "aws configure".

      Or:

      ExpiredToken: The security token included in the request is expired

      ...it's because your AWS CLI session has expired. In this case, do not run the aws configure command. Instead, use the instructions in step 4 of this procedure that starts with Obtain codecatalyst-eks-user's access key and session token to refresh your session.

Step 2: Create an Amazon EKS cluster

In this section, you create a cluster in Amazon EKS. The instructions below describe a quick way to create the cluster using eksctl, but if you want detailed instructions, see:

Note

Private clusters are not supported by the CodeCatalyst integration with Amazon EKS.

Before you begin

Make sure you have completed the following tasks on your development machine:

  • Installed the eksctl utility.

  • Installed the kubectl utility.

  • Installed the AWS CLI and configured it with access keys and a session token.

For information on how to complete these tasks, see Step 1: Set up your development machine.

To create a cluster
Important

Do not use the Amazon EKS service's user interface to create the cluster because the cluster won't be configured correctly. Use the eksctl utility, as described in the following steps.

  1. Go to your Dev Environment.

  2. Create a cluster and nodes:

    eksctl create cluster --name codecatalyst-eks-cluster --region us-west-2

    Where:

    • codecatalyst-eks-cluster is replaced with the name you want to give your cluster.

    • us-west-2 is replaced with your Region.

    After 10-20 minutes, a message similar to the following appears:

    EKS cluster "codecatalyst-eks-cluster" in "us-west-2" region is ready

    Note

    You will see multiple waiting for CloudFormation stack messages while AWS creates your cluster. This is expected.

  3. Verify that your cluster was created successfully:

    kubectl cluster-info

    You will see a message similar to the following, indicating a sucessful cluster creation:

    Kubernetes master is running at https://long-string.gr7.us-west-2.eks.amazonaws.com CoreDNS is running at https://long-string.gr7.us-west-2.eks.amazonaws.com/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

Step 3: Create an Amazon ECR image repository

In this section, you create a private image repository in Amazon Elastic Container Registry (Amazon ECR). This repository stores the Docker image for the tutorial.

For more information about Amazon ECR, see the Amazon Elastic Container Registry User Guide.

To create an image repository in Amazon ECR
  1. Go to your Dev Environment.

  2. Create an empty repository in Amazon ECR:

    aws ecr create-repository --repository-name codecatalyst-eks-image-repo

    Replace codecatalyst-eks-image-repo with the name you want to give the Amazon ECR repository.

    This tutorial assumes you named your repository codecatalyst-eks-image-repo.

  3. Display the Amazon ECR repository’s details:

    aws ecr describe-repositories \ --repository-names codecatalyst-eks-image-repo
  4. Note the “repositoryUri”: value, for example, 111122223333.dkr.ecr.us-west-2.amazonaws.com/codecatalyst-eks-image-repo.

    You need it later when adding the repository to your workflow.

Step 4: Add source files

In this section, you add application source files to your source repository (codecatalyst-eks-source-repository). They consist of:

  • An index.html file – Displays a 'Hello, World!' message in the browser.

  • A Dockerfile – Describes the base image to use for your Docker image and the Docker commands to apply to it.

  • A deployment.yaml file – The Kubernetes manifest that defines the Kubernetes service and deployment.

The folder structure is as follows:

|— codecatalyst-eks-source-repository |— Kubernetes |— deployment.yaml |— public-html | |— index.html |— Dockerfile

index.html

The index.html file displays a 'Hello, World!' message in the browser.

To add the index.html file
  1. Go to your Dev Environment.

  2. In codecatalyst-eks-source-repository, create a folder called public-html.

  3. In /public-html, create a file called index.html with the following contents:

    <html> <head> <title>Hello World</title> <style> body { background-color: black; text-align: center; color: white; font-family: Arial, Helvetica, sans-serif; } </style> </head> <body> <h1>Hello, World!</h1> </body> </html>
  4. At the terminal prompt, enter:

    cd /projects/codecatalyst-eks-source-repository
  5. Add, commit, and push:

    git add . git commit -m "add public-html/index.html" git push

    The index.html is added to your repository in a public-html folder.

Dockerfile

The Dockerfile describes the base Docker image to use and the Docker commands to apply to it. For more information about the Dockerfile, see the Dockerfile Reference.

The Dockerfile specified here indicates to use the Apache 2.4 base image (httpd). It also includes instructions for copying a source file called index.html to a folder on the Apache server that serves webpages. The EXPOSE instruction in the Dockerfile tells Docker that the container is listening on port 80.

To add the Dockerfile
  1. In codecatalyst-eks-source-repository, create a file called Dockerfile with the following contents:

    FROM httpd:2.4 COPY ./public-html/index.html /usr/local/apache2/htdocs/index.html EXPOSE 80

    Do not include a file extension.

    Important

    The Dockerfile must reside in your repository’s root folder. The workflow’s Docker build command expects it to be there.

  2. Add, commit, and push:

    git add . git commit -m "add Dockerfile" git push

    The Dockerfile is added to your repository.

deployment.yaml

In this section, you add a deployment.yaml file to your repository. The deployment.yaml file is a Kubernetes manifest that defines two Kubernetes resources types or kinds to run: a 'service' and a 'deployment'.

  • The 'service' deploys a load balancer into Amazon EC2. The load balancer provides you with an Internet-facing public URL and standard port (port 80) that you can use to browse to the 'Hello, World!' application.

  • The 'deployment' deploys three pods, and each pod will contain a Docker container with the 'Hello, World!' application. The three pods are deployed onto the nodes that were created when you created the cluster.

The manifest in this tutorial is short; however, a manifest can include any number of Kubernetes resource types, such as pods, jobs, ingresses, and network policies. Further, you can use multiple manifest files if your deployment is complex.

To add a deployment.yaml file
  1. In codecatalyst-eks-source-repository, create a folder called Kubernetes.

  2. In /Kubernetes, create a file called deployment.yaml with the following contents:

    apiVersion: v1 kind: Service metadata: name: my-service labels: app: my-app spec: type: LoadBalancer selector: app: my-app ports: - protocol: TCP port: 80 targetPort: 80 --- apiVersion: apps/v1 kind: Deployment metadata: name: my-deployment labels: app: my-app spec: replicas: 3 selector: matchLabels: app: my-app template: metadata: labels: app: my-app spec: containers: - name: codecatalyst-eks-container # The $REPOSITORY_URI and $IMAGE_TAG placeholders will be replaced by actual values supplied by the build action in your workflow image: $REPOSITORY_URI:$IMAGE_TAG ports: - containerPort: 80
  3. Add, commit, and push:

    git add . git commit -m "add Kubernetes/deployment.yaml" git push

    The deployment.yaml file is added to your repository in a folder called Kubernetes.

You have now added all your source files.

Take a moment to double-check your work and make sure you placed all the files in the correct folders. The folder structure is as follows:

|— codecatalyst-eks-source-repository |— Kubernetes |— deployment.yaml |— public-html | |— index.html |— Dockerfile

Step 5: Create AWS roles

In this section, you create AWS IAM roles that your CodeCatalyst workflow will need in order to function. These roles are:

  • Build role – Grants the CodeCatalyst build action (in the workflow) permission to access your AWS account and write to Amazon ECR and Amazon EC2.

  • Deploy role – Grants the CodeCatalyst Deploy to Kubernetes cluster action (in the workflow) permission to access your AWS account and Amazon EKS.

For more information about IAM roles, see IAM roles in the AWS Identity and Access Management User Guide.

Note

To save time, you can create a single role, called the CodeCatalystWorkflowDevelopmentRole-spaceName role, instead of the two roles listed previously. For more information, see Creating the CodeCatalystWorkflowDevelopmentRole-spaceName role for your account and space. Understand that the CodeCatalystWorkflowDevelopmentRole-spaceName role has very broad permissions which may pose a security risk. We recommend that you only use this role in tutorials and scenarios where security is less of a concern. This tutorial assumes you are creating the two roles listed previously.

To create the build and deploy roles, complete the following series of procedures.

1. To create a trust policy for both roles
  1. Go to your Dev Environment.

  2. In the Cloud9-long-string directory, create a file called codecatalyst-eks-trust-policy.json with the following contents:

    { "Version": "2012-10-17", "Statement": [ { "Sid": "", "Effect": "Allow", "Principal": { "Service": [ "codecatalyst-runner.amazonaws.com", "codecatalyst.amazonaws.com" ] }, "Action": "sts:AssumeRole" } ] }
2. To create the build policy for the build role
  • In the Cloud9-long-string directory, create a file called codecatalyst-eks-build-policy.json with the following contents:

    { "Version": "2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "ecr:*", "ec2:*" ], "Resource": "*" } ] }
    Note

    The first time the role is used to run workflow actions, use the wildcard in the resource policy statement and then scope down the policy with the resource name after it is available.

    "Resource": "*"
3. To create the deploy policy for the deploy role
  • In the Cloud9-long-string directory, create a file called codecatalyst-eks-deploy-policy.json with the following contents:

    { "Version": "2012-10-17", "Statement": [ { "Effect": "Allow", "Action": [ "eks:DescribeCluster", "eks:ListClusters" ], "Resource": "*" } ] }
    Note

    The first time the role is used to run workflow actions, use the wildcard in the resource policy statement and then scope down the policy with the resource name after it is available.

    "Resource": "*"

You have now added three policy documents to your Dev Environment. Your directory structure now looks like this:

|— Cloud9-long-string |— .c9 |— codecatalyst-eks-source-repository |— Kubernetes |— public-html |— Dockerfile codecatalyst-eks-build-policy.json codecatalyst-eks-deploy-policy.json codecatalyst-eks-trust-policy.json
4. To add the build policy to AWS
  1. In the Dev Environment terminal, enter:

    cd /projects
  2. Enter:

    aws iam create-policy \ --policy-name codecatalyst-eks-build-policy \ --policy-document file://codecatalyst-eks-build-policy.json
  3. Press Enter.

  4. In the command output, note the "arn": value, for example, arn:aws:iam::111122223333:policy/codecatalyst-eks-build-policy. You need this ARN later.

5. To add the deploy policy to AWS
  1. Enter:

    aws iam create-policy \ --policy-name codecatalyst-eks-deploy-policy \ --policy-document file://codecatalyst-eks-deploy-policy.json
  2. Press Enter.

  3. In the command output, note the deploy policy's "arn": value, for example, arn:aws:iam::111122223333:policy/codecatalyst-eks-deploy-policy. You need this ARN later.

6. To create the build role
  1. Enter:

    aws iam create-role \ --role-name codecatalyst-eks-build-role \ --assume-role-policy-document file://codecatalyst-eks-trust-policy.json
  2. Press Enter.

  3. Enter:

    aws iam attach-role-policy \ --role-name codecatalyst-eks-build-role \ --policy-arn arn:aws:iam::111122223333:policy/codecatalyst-eks-build-policy

    Where arn:aws:iam::111122223333:policy/codecatalyst-eks-build-policy is replaced with the ARN of the build policy you noted earlier.

  4. Press Enter.

  5. At the terminal prompt, enter:

    aws iam get-role \ --role-name codecatalyst-eks-build-role
  6. Press Enter.

  7. Note the role's "Arn": value, for example, arn:aws:iam::111122223333:role/codecatalyst-eks-build-role. You need this ARN later.

7. To create the deploy role
  1. Enter:

    aws iam create-role \ --role-name codecatalyst-eks-deploy-role \ --assume-role-policy-document file://codecatalyst-eks-trust-policy.json
  2. Press Enter.

  3. Enter:

    aws iam attach-role-policy \ --role-name codecatalyst-eks-deploy-role \ --policy-arn arn:aws:iam::111122223333:policy/codecatalyst-eks-deploy-policy

    Where arn:aws:iam::111122223333:policy/codecatalyst-eks-deploy-policy is replaced with the ARN of the deploy policy you noted earlier.

  4. Press Enter.

  5. Enter:

    aws iam get-role \ --role-name codecatalyst-eks-deploy-role
  6. Press Enter.

  7. Note the role's "Arn": value, for example, arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role. You need this ARN later.

You have now created build and deploy roles and noted their ARNs.

Step 6: Add AWS roles to CodeCatalyst

In this step, you add the build role (codecatalyst-eks-build-role) and deploy role (codecatalyst-eks-deploy-role) to the AWS account that you connected to your space. This makes the roles available for use in your workflow.

To add build and deploy roles to your AWS account
  1. In the CodeCatalyst console, navigate to your space.

  2. At the top, choose Settings.

  3. In the navigation pane, choose AWS accounts. A list of accounts appears.

  4. In the Amazon CodeCatalyst display name column, copy the display name of the AWS account where you created your build and deploy roles. (It might be a number.) You'll need this value later, when creating your workflow.

  5. Choose the display name.

  6. Choose Manage roles from AWS management console.

    The Add IAM role to Amazon CodeCatalyst space page appears. You might need to sign in to access the page.

  7. Select Add an existing role you have created in IAM.

    A drop-down list appears. The list displays the build and deploy roles, and any other IAM roles with a trust policy that includes the codecatalyst-runner.amazonaws.com and codecatalyst.amazonaws.com service principals.

  8. From the drop-down list, add:

    • codecatalyst-eks-build-role

    • codecatalyst-eks-deploy-role

    Note

    If you see The security token included in the request is invalid, it might be because you do not have the right permissions. To fix this issue, sign out of AWS as sign back in with the AWS account that you used when you created your CodeCatalyst space.

  9. Return to the CodeCatalyst console and refresh the page.

    The build and deploy roles should now appear under IAM roles.

    These roles are now available for use in CodeCatalyst workflows.

Step 7: Update the ConfigMap

You must add the deploy role that you created in Step 5: Create AWS roles to the Kubernetes ConfigMap file to give the Deploy to Kubernetes cluster action (in your workflow) the ability to access and interact with your cluster. You can use eksctl or kubectl to perform this task.

To configure the Kubernetes ConfigMap file using eksctl
  • In the Dev Environment terminal, enter:

    eksctl create iamidentitymapping --cluster codecatalyst-eks-cluster --arn arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role --group system:masters --username codecatalyst-eks-deploy-role --region us-west-2

    Where:

    • codecatalyst-eks-cluster is replaced with the cluster name of the Amazon EKS cluster.

    • arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role is replaced with the ARN of the deploy role that you created in Step 5: Create AWS roles.

    • codecatalyst-eks-deploy-role (next to --username) is replaced with the name of the deploy role that you created in Step 5: Create AWS roles.

      Note

      If you decided not to create a deploy role, replace codecatalyst-eks-deploy-role with the name of the CodeCatalystWorkflowDevelopmentRole-spaceName role. For more information about this role, see Step 5: Create AWS roles.

    • us-west-2 is replaced with your Region.

    For details on this command, see Manage IAM users and roles.

    A message similar to the following appears:

    2023-06-09 00:58:29 [ℹ]  checking arn arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role against entries in the auth ConfigMap
    2023-06-09 00:58:29 [ℹ]  adding identity "arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role" to auth ConfigMap
To configure the Kubernetes ConfigMap file using kubectl
  1. In the Dev Environment terminal, enter:

    kubectl edit configmap -n kube-system aws-auth

    The ConfigMap file appears on the screen.

  2. Add the text in red italics:

    # Please edit the object below. Lines beginning with a '#' will be ignored,
    # and an empty file will abort the edit. If an error occurs while saving this file will be
    # reopened with the relevant failures.
    #
    apiVersion: v1
    data:
      mapRoles: |
        - groups:
          - system:bootstrappers
          - system:nodes
          rolearn: arn:aws:iam::111122223333:role/eksctl-codecatalyst-eks-cluster-n-NodeInstanceRole-16BC456ME6YR5
          username: system:node:{{EC2PrivateDNSName}}
        - groups:
          - system:masters
          rolearn: arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role
          username: codecatalyst-eks-deploy-role
      mapUsers: |
        []
    kind: ConfigMap
    metadata:
      creationTimestamp: "2023-06-08T19:04:39Z"
      managedFields:
      ...

    Where:

    • arn:aws:iam::111122223333:role/codecatalyst-eks-deploy-role is replaced with the ARN of the deploy role that you created in Step 5: Create AWS roles.

    • codecatalyst-eks-deploy-role (next to username:)is replaced with the name of the deploy role that you created in Step 5: Create AWS roles.

      Note

      If you decided not to create a deploy role, replace codecatalyst-eks-deploy-role with the name of the CodeCatalystWorkflowDevelopmentRole-spaceName role. For more information about this role, see Step 5: Create AWS roles.

    For details, see Enabling IAM principal access to your cluster in the Amazon EKS User Guide.

You have now given the deploy role, and by extension the Deploy to Amazon EKS action, system:masters permissions to your Kubernetes cluster.

Step 8: Create and run a workflow

In this step, you create a workflow that takes your source files, builds them into a Docker image, and then deploys the image into tree pods in your Amazon EKS cluster.

The workflow consists of the following building blocks that run sequentially:

  • A trigger – This trigger starts the workflow run automatically when you push a change to your source repository. For more information about triggers, see Starting a workflow run automatically using triggers.

  • A build action (BuildBackend) – On trigger, the action builds the Docker image using the Dockerfile and pushes the image to Amazon ECR. The build action also updates the $REPOSITORY_URI and $IMAGE_TAG variables in the deployment.yaml file with the correct values, and then creates an output artifact of this file and any others in the Kubernetes folder. In this tutorial, the only file in the Kubernetes folder is deployment.yaml but you could include more files. The artifact is used as the input for the deploy action, which is next.

    For more information about the build action, see Building with workflows.

  • A deploy action (DeployToEKS) – On completion of the build action, the deploy action looks for the output artifact generated by the build action (Manifests), and finds the deployment.yaml file inside of it. The action then follows the instructions in the deployment.yaml file to run three pods—each containing a single 'Hello, World!' Docker container—inside your Amazon EKS cluster.

To create a workflow
  1. Go to the CodeCatalyst console.

  2. Navigate to your project (codecatalyst-eks-project).

  3. In the navigation pane, choose CI/CD, and then choose Workflows.

  4. Choose Create workflow.

  5. For Source repository, choose codecatalyst-eks-source-repository.

  6. For Branch, choose main.

  7. Choose Create.

  8. Delete the YAML sample code.

  9. Add the following YAML code to create a new workflow definition file:

    Note

    For more information about the workflow definition file, see Workflow YAML definition.

    Note

    In the YAML code that follows, you can omit the Connections: sections if you want. If you omit these sections, you must ensure that the role specified in the Default IAM role field in your environment includes the permissions and trust policies of both roles described in Step 6: Add AWS roles to CodeCatalyst. For more information about setting up an environment with a default IAM role, see Creating an environment.

    Name: codecatalyst-eks-workflow SchemaVersion: 1.0 Triggers: - Type: PUSH Branches: - main Actions: BuildBackend: Identifier: aws/build@v1 Environment: Name: codecatalyst-eks-environment Connections: - Name: codecatalyst-account-connection Role: codecatalyst-eks-build-role Inputs: Sources: - WorkflowSource Variables: - Name: REPOSITORY_URI Value: 111122223333.dkr.ecr.us-west-2.amazonaws.com/codecatalyst-eks-image-repo - Name: IMAGE_TAG Value: ${WorkflowSource.CommitId} Configuration: Steps: #pre_build: - Run: echo Logging in to Amazon ECR... - Run: aws --version - Run: aws ecr get-login-password --region us-west-2 | docker login --username AWS --password-stdin 111122223333.dkr.ecr.us-west-2.amazonaws.com #build: - Run: echo Build started on `date` - Run: echo Building the Docker image... - Run: docker build -t $REPOSITORY_URI:latest . - Run: docker tag $REPOSITORY_URI:latest $REPOSITORY_URI:$IMAGE_TAG #post_build: - Run: echo Build completed on `date` - Run: echo Pushing the Docker images... - Run: docker push $REPOSITORY_URI:latest - Run: docker push $REPOSITORY_URI:$IMAGE_TAG # Replace the variables in deployment.yaml - Run: find Kubernetes/ -type f | xargs sed -i "s|\$REPOSITORY_URI|$REPOSITORY_URI|g" - Run: find Kubernetes/ -type f | xargs sed -i "s|\$IMAGE_TAG|$IMAGE_TAG|g" - Run: cat Kubernetes/* # The output artifact will be a zip file that contains Kubernetes manifest files. Outputs: Artifacts: - Name: Manifests Files: - "Kubernetes/*" DeployToEKS: DependsOn: - BuildBackend Identifier: aws/kubernetes-deploy@v1 Environment: Name: codecatalyst-eks-environment Connections: - Name: codecatalyst-account-connection Role: codecatalyst-eks-deploy-role Inputs: Artifacts: - Manifests Configuration: Namespace: default Region: us-west-2 Cluster: codecatalyst-eks-cluster Manifests: Kubernetes/

    In the preceding code, replace:

    • Both instances of codecatalyst-eks-environment with the name of the environment you created in Prerequisites.

    • Both instances of codecatalyst-account-connection with the display name of your account connection. The display name might be a number. For more information, see Step 6: Add AWS roles to CodeCatalyst.

    • codecatalyst-eks-build-role with the name of the build role you created in Step 5: Create AWS roles.

    • 111122223333.dkr.ecr.us-west-2.amazonaws.com/codecatalyst-eks-image-repo (in the Value: property) with the URI of the Amazon ECR repository you created in Step 3: Create an Amazon ECR image repository.

    • 111122223333.dkr.ecr.us-west-2.amazonaws.com (in the Run: aws ecr command) with the URI of the Amazon ECR repository without the image suffix (/codecatalyst-eks-image-repo).

    • codecatalyst-eks-deploy-role with the name of the deploy role you created in Step 5: Create AWS roles.

    • Both instances of us-west-2 with your AWS Region code. For a list of Region codes, see Regional endpoints in the AWS General Reference.

    Note

    If you decided not to create build and deploy roles, replace codecatalyst-eks-build-role and codecatalyst-eks-deploy-role with the name of the CodeCatalystWorkflowDevelopmentRole-spaceName role. For more information about this role, see Step 5: Create AWS roles.

  10. (Optional) Choose Validate to make sure that the YAML code is valid before committing.

  11. Choose Commit.

  12. In the Commit workflow dialog box, enter the following:

    1. For Commit message, remove the text and enter:

      Add first workflow
    2. For Repository, choose codecatalyst-eks-source-repository.

    3. For Branch name, choose main.

    4. Choose Commit.

    You have now created a workflow. A workflow run starts automatically because of the trigger defined at the top of the workflow. Specifically, when you committed (and pushed) the workflow.yaml file to your source repository, the trigger started the workflow run.

To view the workflow run progress
  1. In the navigation pane of the CodeCatalyst console, choose CI/CD, and then choose Workflows.

  2. Choose the workflow you just created, codecatalyst-eks-workflow.

  3. Choose BuildBackend to see the build progress.

  4. Choose DeployToEKS to see the deployment progress.

    For more information about viewing run details, see Viewing workflow run status and details.

To verify the deployment
  1. Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.

  2. On the left, near the bottom, choose Load Balancers.

  3. Select the load balancer that was created as part of your Kubernetes deployment. If you're not sure which load balancer to choose, look for the following tags under the Tags tab:

    • kubernetes.io/service-name

    • kubernetes.io/cluster/ekstutorialcluster

  4. With the correct load balancer selected, choose the Description tab.

  5. Copy and paste the DNS name value into your browser's address bar.

    The 'Hello, World!' webpage appears in your browser, indicating that you successfully deployed your application.

Step 9: Make a change to your source files

In this section, you make a change to the index.html file in your source repository. This change causes the workflow to build a new Docker image, tag it with a commit ID, push it to Amazon ECR, and deploy it to Amazon ECS.

To change the index.html
  1. Go to your Dev Environment.

  2. At the terminal prompt, change to your source repository:

    cd /projects/codecatalyst-eks-source-repository
  3. Pull the latest workflow changes:

    git pull
  4. Open codecatalyst-eks-source-repository/public-html/index.html.

  5. On line 14, change the Hello, World! text to Tutorial complete!.

  6. Add, commit, and push:

    git add . git commit -m "update index.html title" git push

    A workflow run starts automatically.

  7. (Optional) Enter:

    git show HEAD

    Note the commit ID for the index.html change. This commit ID will be tagged to the Docker image that will be deployed by the workflow run that you just started.

  8. Watch the deployment progress:

    1. In the CodeCatalyst console, in the navigation pane, choose CI/CD, and then choose Workflows.

    2. Choose codecatalyst-eks-workflow to view the latest run.

    3. Choose BuildBackend, and DeployToEKS to see the workflow run progress.

  9. Verify that your application was updated, as follows:

    1. Open the Amazon EC2 console at https://console.aws.amazon.com/ec2/.

    2. On the left, near the bottom, choose Load Balancers.

    3. Select the load balancer that was created as part of your Kubernetes deployment.

    4. Copy and paste the DNS name value into your browser's address bar.

      The 'Tutorial Complete!' webpage appears in your browser, indicating that you successfully deployed a new revision of your application.

  10. (Optional) In AWS, switch to the Amazon ECR console and verify that the new Docker image was tagged with the commit ID from step 7 of this procedure.

Clean up

You should clean up your environment so that you're not charged unnecessarily for the storage and compute resources used by this tutorial.

To clean up
  1. Delete your cluster:

    1. In the Dev Environment terminal, enter:

      eksctl delete cluster --region=us-west-2 --name=codecatalyst-eks-cluster

      Where:

      • us-west-2 is replaced with your Region.

      • codecatalyst-eks-cluster is replaced with the name of the cluster you created.

      After 5-10 minutes, the cluster and associated resources are deleted, including but not limited to AWS CloudFormation stacks, nodes groups (in Amazon EC2), and load balancers.

    Important

    If the eksctl delete cluster command doesn't work, you may need to refresh your AWS credentials or your kubectl credentials. If you're not sure which credentials to refresh, refresh the AWS credentials first. To refresh your AWS credentials, see How do I fix "Unable to locate credentials" and "ExpiredToken" errors?. To refresh your kubectl credentials, see How do I fix "Unable to connect to the server" errors?.

  2. In the AWS console, clean up as follows:

    1. In Amazon ECR, delete codecatalyst-eks-image-repo.

    2. In IAM Identity Center, delete:

      1. codecatalyst-eks-user

      2. codecatalyst-eks-permission-set

    3. In IAM, delete:

      • codecatalyst-eks-build-role

      • codecatalyst-eks-deploy-role

      • codecatalyst-eks-build-policy

      • codecatalyst-eks-deploy-policy

  3. In the CodeCatalyst console, clean up as follows:

    1. Delete codecatalyst-eks-workflow.

    2. Delete codecatalyst-eks-environment.

    3. Delete codecatalyst-eks-source-repository.

    4. Delete your Dev Environment.

    5. Delete codecatalyst-eks-project.

In this tutorial, you learned how to deploy an application to an Amazon EKS service using a CodeCatalyst workflow and a Deploy to Kubernetes cluster action.