Skip to content
/
AWS K8s Deployment Exampl...

AWS Kubernetes Services (EKS) Deployment Example

This how-to guide is an example AWS Kubernetes deployment. This is not the only way to deploy a Kubernetes cluster.

Security

Follow your organization's security requirements for Kubernetes clusters.

Setup Summary

The following Azure resources are used in this example deployment.

  • AWS CLI
  • EKSCTL
  • AWS S3 Bucket

Install AWS CLI

The AWS Command-Line Interface (CLI) allows you to interact with AWS resources from a terminal or command prompt.

Select OS

Select your operating system to view installation instructions. Note: Instructions for Windows systems will be available soon.

  1. Install curl using Brew.

    # Install Brew
/bin/bash -c "$(curl -fsSL https://raw.githubusercontent.com/Homebrew/install/HEAD/install.sh)"

# Install Curl
brew install curl

  2. Install the AWS CLI.

    curl "https://awscli.amazonaws.com/awscli-exe-linux-x86_64.zip" -o "awscliv2.zip"
unzip awscliv2.zip
sudo ./aws/install

  3. Confirm the installation by checking the app version.

    aws --version

Configure AWS CLI

You must authenticate to your AWS account to access AWS resources using the AWS CLI.

Region

This example uses the us-east-1 region. This is the default region for Model Scanner.

  1. Configure the AWS CLI. (Note: This step has you manually entering your data. Using a configuration file may be more secure.)

    aws configure

  2. Enter your:

    a. AWS Access Key ID b. AWS Secret Access Key c. Default Region d. Default Output Format

Amazon S3

Default Bucket and Default Region
  • The default bucket name for Model Scanner is hl-modelscan. To change the default bucket name, set the HL_MODEL_SCAN_BUCKET environment variable in the yaml file.
  • This example uses the us-east-1 region. This is the default region for Model Scanner.

  1. Create an S3 bucket for Model Scanner.

    aws s3 mb s3://hl-modelscan

  2. Create the IAM user. Save the output for later use.

    aws iam create-user --user-name HLModelScanner

  3. Run the following command to write an IAM policy document to allow the Model Scanner to access the S3 bucket. Replace <bucket> with the S3 bucket name you created earlier. This outputs a file named hl-modelscanner-s3-policy.json to the current folder.

    echo -e '{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": [
                "s3:GetObject",
                "s3:GetObjectVersion",
                "s3:ListBucket",
                "s3:PutObject",
                "s3:DeleteObject"
            ],
            "Resource": "arn:aws:s3:::<bucket>/*"
        },
        {
            "Effect": "Allow",
            "Action": [
                "s3:ListBucket"
            ],
            "Resource": "arn:aws:s3:::<bucket>"
        }
    ]
}' > hl-modelscanner-s3-policy.json

  4. Attach the IAM policy for the S3 bucket to the previously created user.

  5. Generate Access Keys for the user.

    aws iam create-access-key --user-name HLModelScanner

  6. Save the AccessKeyId and SecretAccessKey fields from the aws iam create-access-key response. They will be used later as the AWS_ACCESS_KEY_ID and AWS_SECRET_ACCESS_KEY when deploying the Model Scanner.

Install EKSCTL

EKSCTL is a command line tool that helps you create an EKS cluster, including the supporting resources for that cluster. It is used for this demonstration to help simplify the process.

Select OS

Select your operating system to view installation instructions. Note: Instructions for Windows systems will be available soon.

  1. Install the Weaveworks Tap and Weaveworks EKSCTL.

    brew tap weaveworks/tap
brew install weaveworks/tap/eksctl

Create EKS Cluster

  1. Creating an EKS cluster using EKSCTL only requires one command. Replace <cluster-name> with the name you want to give the cluster. Replace <aws-region> with the region you want to deploy the cluster in. Creating the cluster can take some time.

    eksctl create cluster --name <cluster-name> --region <aws-region>

Connect Local Kubectl to EKS Cluster

  1. Use the following command to connect your local kubectl instance to your AWS EKS cluster. Replace <aws-region> and <cluster-name> with the information you used in the previous step.

    aws eks update-kubeconfig --region <aws-region> --name <cluster-name>

Set Default Storage Class

A default storage class is needed for some of the Model Scanner pods, like the Redis pods. Setting a default storage class allows clusters to be created without needing to be assigned to the storage class.

GP2 Storage Class

When creating an EKS cluster using EKSCTL, a GP2 storage class is also created.

  1. Get a list of the StorageClasses in the cluster.

    kubectl get storageclass

  2. Make the GP2 storage class the default. This allows pods to be automatically assigned to the storage resource.

    kubectl patch storageclass gp2 -p '{"metadata": {"annotations":{"storageclass.kubernetes.io/is-default-class":"true"}}}'

Enable IAM OIDC Provider

To use AWS Identity and Access Management (IAM) roles for service accounts, an IAM OIDC provider must exist for your cluster's OIDC issuer URL.

  1. Create an IAM OIDC identity provider for your cluster with the following command. Replace <aws-region> with the region you want to deploy the cluster in. Replace <cluster-name> with the name of the cluster.

    eksctl utils associate-iam-oidc-provider --region=<aws-region> --cluster=<cluster-name> --approve

Create AWS EBS CSI Driver IAM Role

The Amazon EBS CSI driver Amazon EKS add-on is a Kubernetes Container Storage Interface (CSI) plugin that provides Amazon EBS storage for your cluster.

  1. Use the following command to create an IAM role and attach the managed policy to it. Replace <aws-region> with the region you want to deploy the cluster in. Replace <cluster-name> with the name of the cluster.

    eksctl create iamserviceaccount \
    --region <aws-region> \
    --name ebs-csi-controller-sa \
    --namespace kube-system \
    --cluster <cluster-name> \
    --attach-policy-arn arn:aws:iam::aws:policy/service-role/AmazonEBSCSIDriverPolicy \
    --approve \
    --role-only \
    --role-name AmazonEKS_EBS_CSI_DriverRole

Create AWS EBS CSI Addon

Create AWS EBS CSI add-ons to manage your EKS cluster.

  1. Replace <cluster-name> with the name of the cluster.

    eksctl create addon --name aws-ebs-csi-driver --cluster <cluster-name> --service-account-role-arn arn:aws:iam::$(aws sts get-caller-identity --query Account --output text):role/AmazonEKS_EBS_CSI_DriverRole --force
Previous page
Next page