Debezium Management Platform

Debezium Management Platform (Debezium Orchestra) aims to provide means to simplify the deployment of Debezium to various environments in highly opinionated manner. The goal is not to provide total control over environment specific configuration. To achieve this goal the platform uses a data-centric view on Debezium components.

Disclaimer: This project is still in early development stage and should not be used in production.

Platform Architecture

The platform is composed of two main components:

1. Conductor: The back-end component which provides a set of APIs to orchestrate and control Debezium deployments.
2. Stage: The front-end component which provides a user interface to interact with the Conductor.

Conductor Architecture

The conductor component itself is composed of several subcomponents:

1. API Server: The main entry point for the platform. It provides a set of APIs to interact with the platform.
2. Watcher: Component responsible for the actual communication with deployment environment (e.g. Debezium Operator in K8s cluster).

Installation

You can install the platform through helm chart. For instructions refer to the README

Run the example

If you don't have already a Kubernetes cluster up, you can use one of the commons tools to have a local K8s cluster

• minikube
• kind

If you are using kind locally, expose ports 80 and 443 on the control-plane node so the ingress can be reached from your host:

kind: Cluster
apiVersion: kind.x-k8s.io/v1alpha4
nodes:
- role: control-plane
  extraPortMappings:
    - containerPort: 80
      hostPort: 80
      listenAddress: "127.0.0.1"
      protocol: TCP
    - containerPort: 443
      hostPort: 443
      listenAddress: "127.0.0.1"
      protocol: TCP

kind create cluster --config kind-ingress.yaml

The prerequisite is to install an ingress controller.

For example, on kind you can install ingress-nginx with:

helm upgrade --install ingress-nginx ingress-nginx \
  --repo https://kubernetes.github.io/ingress-nginx \
  --namespace ingress-nginx \
  --create-namespace \
  --set controller.hostPort.enabled=true

For this example, considering a local setup, we will use the /etc/hosts to resolve the domain.

The following script will map the K8s IP to the specified domain into the /etc/hosts.

export DEBEZIUM_PLATFORM_DOMAIN=platform.debezium.io
sudo ./examples/update_hosts.sh

NOTE: If you are using minikube on Mac, you need also to run the minikube tunnel command. For more details see this and this.

NOTE: If you are using Windows, add 127.0.0.1 platform.debezium.io to C:\Windows\System32\drivers\etc\hosts.

Create a dedicated namespace

kubectl create ns debezium-platform

and then install debezium-platform through helm

cd helm && 
helm dependency build &&
helm install debezium-platform . -n debezium-platform -f ../examples/example.yaml &&
cd ..

after all pods are running you should access the platform UI from http://platform.debezium.io/

To finish the example we will create a PostgreSQL, that will be used as source database, and a kafka cluster, used as destination in our example pipeline.

# Deploy the source database

kubectl create -n debezium-platform -f examples/k8s/database/001_postgresql.yml

Install the Strimzi operator

helm repo add strimzi https://strimzi.io/charts/ &&
helm repo update strimzi &&
helm install strimzi-operator strimzi/strimzi-kafka-operator --version 0.45.1 --namespace debezium-platform

# Deploy the kafka cluster

kubectl create -n debezium-platform -f examples/k8s/kafka/001_kafka.yml

# Create a test pipeline
#
# The script uses the `http` command from HTTPie.

./examples/seed.sh platform.debezium.io 80 examples/payloads/

And that's all.

You should have a test pipeline configured to move data from PostgreSQL to Kakfa.