Launch EKS

DO NOT PROCEED with this step unless you have validated the IAM role in use by the Cloud9 IDE. You will not be able to run the necessary kubectl commands in the later modules unless the EKS cluster is built using the IAM role.

Challenge:

How do I check the IAM role on the workspace?

Expand here to see the solution

Create an EKS cluster

eksctl version must be 0.38.0 or above to deploy EKS 1.20, click here to get the latest version.

Create an eksctl deployment file (tinyhats.yaml) use in creating your cluster using the following syntax:

cat << EOF > tinyhats.yaml
---
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig

metadata:
  name: tinyhats
  region: ${AWS_REGION}
  version: "1.20"

availabilityZones: ["${AZS[0]}", "${AZS[1]}"]

managedNodeGroups:
- name: nodegroup
  desiredCapacity: 2
  instanceType: t3.2xlarge
  ssh:
    enableSsm: true

# To enable all of the control plane logs, uncomment below:
# cloudWatch:
#  clusterLogging:
#    enableTypes: ["*"]

secretsEncryption:
  keyARN: ${MASTER_ARN}
EOF

Later on in this workshop, we will be installing Pixie to monitor our EKS cluster. Pixie requires a minimum of 2GB of RAM per node and recommends a minimum of 8GB per node. When deploying in a real world environment, if you see constant churning of your Pixie pods, you may wish to increase the amount of RAM allocated to each of your nodes.

Next, use the file you created as the input for the eksctl cluster creation.

eksctl create cluster -f tinyhats.yaml

Launching EKS and all the dependencies will take approximately 25 minutes