Confluent for Kubernetes is now Generally Available! You can find the docs at https://docs.confluent.io/operator/current/overview.html and the examples repo at https://github.com/confluentinc/confluent-kubernetes-examples.

[OLD] Confluent Operator Early Access

The next version of Confluent Operator offers a Kubernetes-native experience, including:

• First class Kubernetes custom resource definitions (CRDs)
• Config overrides for all Confluent components
• Auto generated certificates
• Kubernetes tolerations, pod/node affinity support

Check out the product in action: Kubernetes-Native DevOps with Confluent Operator

Get early access by registering interest here: Confluent Operator Early Access Registration

This repository is structured into Concepts and Scenario workflows:

• Read Concepts to understand how Confluent on Kubernetes works, and should be architected.
• Use Scenario workflows to deploy and manage Confluent on Kubernetes for various use cases.

Scenario workflows

Clone the following repo to get the files needed for each workflow:

git clone git@github.com:confluentinc/operator-earlyaccess.git

The documentation is organized as scenario workflows:

• Quickstart
  • Deploy simple non-secure Confluent Platform
• Secure deployment
  • Production recommended secure deployment
  • Deploy Confluent Platform with encryption and authentication
  • Deploy Confluent Platform with RBAC authorization
• External Access
  • Deploy Confluent Platform with external access through Load Balancer
  • Deploy Confluent Platform with external access through NodePorts
  • Deploy Confluent Platform with external access using static host-based routing
  • Deploy Confluent Platform with external access using static port-based routing
• Multi-AZ deployment
  • Deploy Confluent Platform to a multi-zone Kubernetes cluster, with Kafka rack awareness configured
• Connect to Confluent Cloud
  • Deploy Confluent Platform to connect and monitor multiple Connect and ksqlDB clusters

Pre-requisites

This Confluent Operator Early Access is compatible with Confluent Platform 6.0.

To use this Confluent Operator Early Access, you’ll need:

• A Kubernetes cluster - any CNCF conformant version
• Helm 3 installed on your local machine
• Kubectl installed on your local machine

Set up the Kubernetes cluster

Set up the Kubernetes cluster for this tutorial.

1. Add or get access to a Kubernetes cluster.

2. Create the namespace and set it to the current namespace. In this tutorial, we will deploy Confluent Platform in the confluent namespace.

   kubectl create namespace confluent
   

   kubectl config set-context --current --namespace=confluent
   

Configure Early Access credentials

1. For this Early Access program, you will have received an API key (associated with your email address) to the Confluent JFrog Artifactory.

2. Save the API key and your email for easy access while running this tutorial:

   export USER=<your email address>
   export EMAIL=<your email address>
   export APIKEY=<your API key>
   

3. Create a Kubernetes secret with the registry credentials:

   kubectl create secret docker-registry confluent-registry \
     --docker-server=confluent-docker-internal-early-access-operator-2.jfrog.io \
     --docker-username=$USER \
           --docker-password=$APIKEY \
           --docker-email=$EMAIL
   

4. The Confluent Operator itself is packaged as a Helm Chart. Add a Helm repo:

   helm repo add confluentinc_earlyaccess \
     https://confluent.jfrog.io/confluent/helm-early-access-operator-2 \
     --username $USER \
     --password $APIKEY
   

   helm repo update
   

Download Confluent Operator tutorial package

Download the tutorial package from the Git Hub repo:

git clone git@github.com:confluentinc/operator-earlyaccess.git

Remove Confluent Operator tutorial environment

When you finish running the scenarios in the Confluent Operator tutorial packages, remove the namespace, the secret, and the Helm repo with the following command:

kubectl delete secret docker-registry

helm repo remove confluentinc_earlyaccess

kubectl delete namespace confluent