Charmed Apache Kafka K8s Documentation - How to deploy on EKS

charm
, , ,
1

How to deploy on EKS

Amazon Elastic Kubernetes Service (EKS) is a popular, fully automated Kubernetes service. To access the EKS Web interface, go to console.aws.amazon.com/eks/home.

Summary

Install EKS and Juju tooling

Install Juju and the kubectl CLI tools (that will be used for managing the Kubernetes cluster) via snap:

sudo snap install juju --channel 3.5/stable
sudo snap install kubectl --classic

Follow the installation guides for:

  • eksctl - the Amazon EKS CLI
  • AWS CLI - the Amazon Web Services CLI

To check they are all correctly installed, you can run the commands below.

juju version
Sample output: 
3.5.2-genericlinux-amd64

kubectl version --client
Sample output: 
Client Version: v1.28.2
Kustomize Version: v5.0.4-0.20230601165947-6ce0bf390ce3

eksctl info
Sample output: 
eksctl version: 0.159.0
kubectl version: v1.28.2

aws --version
Sample output: 
aws-cli/2.13.25 Python/3.11.5 Linux/6.2.0-33-generic exe/x86_64.ubuntu.23 prompt/off

Authenticate

Create an IAM account (or use legacy access keys) and login to AWS:

> aws configure
AWS Access Key ID [None]: SECRET_ACCESS_KEY_ID
AWS Secret Access Key [None]: SECRET_ACCESS_KEY_VALUE
Default region name [None]: eu-west-3
Default output format [None]:

Verify that the CLI tool is correctly authenticating

aws sts get-caller-identity
Sample output: 
{
    "UserId": "1234567890",
    "Account": "1234567890",
    "Arn": "arn:aws:iam::1234567890:root"
}

Create a new EKS cluster

Export the deployment name for further use:

export JUJU_NAME=eks-$USER-$RANDOM

This following examples in this guide will use the location eu-west-3 and K8s v.1.27 - feel free to change this for your own deployment.

Sample `cluster.yaml`: 
apiVersion: eksctl.io/v1alpha5
kind: ClusterConfig

metadata:
    name: ${JUJU_NAME}
    region: eu-west-3
    version: "1.27"
iam:
  withOIDC: true

addons:
- name: aws-ebs-csi-driver
  wellKnownPolicies:
    ebsCSIController: true

nodeGroups:
    - name: ng-1
      minSize: 3
      maxSize: 5
      iam:
        attachPolicyARNs:
        - arn:aws:iam::aws:policy/AmazonEKSWorkerNodePolicy
        - arn:aws:iam::aws:policy/AmazonEKS_CNI_Policy
        - arn:aws:iam::aws:policy/AmazonEC2ContainerRegistryReadOnly
        - arn:aws:iam::aws:policy/AmazonSSMManagedInstanceCore
        - arn:aws:iam::aws:policy/AmazonS3FullAccess
      instancesDistribution:
        maxPrice: 0.15
        instanceTypes: ["m5.xlarge", "m5.2xlarge"] # At least two instance types should be specified
        onDemandBaseCapacity: 0
        onDemandPercentageAboveBaseCapacity: 50
        spotInstancePools: 2

Bootstrap EKS cluster with the following command:

eksctl create cluster -f cluster.yaml
Sample `cluster.yaml`: 
...
2023-10-12 11:13:58 [ℹ]  using region eu-west-3
2023-10-12 11:13:59 [ℹ]  using Kubernetes version 1.27
...
2023-10-12 11:40:00 [✔]  EKS cluster "eks-taurus-27506" in "eu-west-3" region is ready

Bootstrap Juju on EKS

Add Juju K8s clouds:

juju add-k8s $JUJU_NAME

Bootstrap Juju controller:

juju bootstrap $JUJU_NAME

Deploy Charms

Create a new Juju model, if needed:

juju add-model <MODEL_NAME>

(Optional) Increase the debug level if you are troubleshooting charms:

juju model-config logging-config='<root>=INFO;unit=DEBUG'

Then, Charmed Apache Kafka can be deployed as usual:

juju deploy zookeeper-k8s -n3 --channel 3/stable
juju deploy kafka-k8s -n3 --channel 3/stable
juju integrate kafka-k8s zookeeper-k8s

We also recommend to deploy a Data Integrator for creating an admin user to manage the content of the Kafka cluster:

juju deploy data-integrator admin --channel edge \
  --config extra-user-roles=admin \
  --config topic-name=admin-topic

And integrate it with the Kafka application:

juju integrate kafka-k8s admin

For more information on Data Integrator and how to use it, please refer to the how-to manage applications user guide.

Display deployment information

Display information about the current deployments with the following commands:

kubectl cluster-info
Sample output: 
Kubernetes control plane is running at https://AAAAAAAAAAAAAAAAAAAAAAA.gr7.eu-west-3.eks.amazonaws.com
CoreDNS is running at https://AAAAAAAAAAAAAAAAAAAAAAA.gr7.eu-west-3.eks.amazonaws.com/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

eksctl get cluster -A
Sample output: 
NAME            REGION      EKSCTL CREATED
eks-marc-9587	eu-west-3	True

kubectl get node
Sample output: 
NAME                                           STATUS   ROLES    AGE     VERSION
ip-192-168-1-168.eu-west-3.compute.internal    Ready    <none>   5d22h   v1.27.16-eks-a737599
ip-192-168-45-234.eu-west-3.compute.internal   Ready    <none>   3h25m   v1.27.16-eks-a737599
ip-192-168-85-225.eu-west-3.compute.internal   Ready    <none>   5d22h   v1.27.16-eks-a737599

Clean up

Always clean EKS resources that are no longer necessary - they could be costly!

To clean the EKS cluster, resources and juju cloud, run the following commands:

juju destroy-controller $JUJU_NAME --yes --destroy-all-models --destroy-storage --force
juju remove-cloud $JUJU_NAME

List all services and then delete those that have an associated EXTERNAL-IP value (e.g. load balancers):

kubectl get svc --all-namespaces
kubectl delete svc <service-name>

Next, delete the EKS cluster (As described on the Deleting an Amazon EKS cluster page):

eksctl get cluster -A
eksctl delete cluster <cluster_name> --region eu-west-3 --force --disable-nodegroup-eviction

Finally, remove AWS CLI user credentials (to avoid forgetting and getting exposed to a risk of leaking credentials):

rm -f ~/.aws/credentials