Deploying HariKube operator to manage dynamic database topology

Before diving in, make sure you have a running Kubernetes cluster. If you need help setting one up, check out our tutorial. →

Installing Essential Kubernetes Add-ons

To get started, let’s install two popular open-source add-ons: Cert-Manager for automated certificate management, and the Prometheus Operator for monitoring and alerting. These add-ons are widely used and will help your cluster run smoothly.

1
2

kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.16.3/cert-manager.yaml
kubectl apply -f https://github.com/prometheus-operator/prometheus-operator/releases/download/v0.77.1/stripped-down-crds.yaml

Deploying the HariKube Operator

HariKube images aren’t public yet. If you’d like to try them, request a trial version on the Get Started page.

Start by authenticating your local Docker client with the private registry at registry.harikube.info. This step is essential for pulling images from the registry.

1

docker login registry.harikube.info

Next, pull the HariKube Operator image from our registry:

1

docker pull registry.harikube.info/harikube/operator:beta-v1.0.0

If you’re using Kind for your cluster, load the image into your Kind node:

1

kind load docker-image -n harikube-cluster registry.harikube.info/harikube/operator:beta-v1.0.0

Now, deploy the HariKube Operator to your cluster. This operator will manage your custom database routing policies and automate topology changes.

1

kubectl apply -f https://harikube.info/manifests/harikube-operator-beta-v1.0.0.yaml

Configuring the Operator and Registering a Custom Resource

Let’s create your first topology configuration. This tells the HariKube Operator how to route data for a specific custom resource.

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13

apiVersion: harikube.info/v1
kind: TopologyConfig
metadata:
  name: topologyconfig-shirts
  namespace: default
spec:
  targetSecret: default/topology-config
  backends:
  - name: shirts
    endpoint: sqlite:///db/shirts.db?_journal=WAL&cache=shared
    customresource:
      group: stable.example.com
      kind: shirts

This TopologyConfig custom resource instructs the operator to manage all shirts resources, storing their data in a dedicated SQLite database (shirts.db) inside the HariKube Middleware container.

Apply the configuration:

1

kubectl apply -f topology-shirts.yml

Defining and Deploying a New Application Resource

To see HariKube in action, let’s define a new custom resource type. We’ll use a simple example: a Shirt resource.

1

kubectl apply -f https://raw.githubusercontent.com/kubernetes/website/main/content/en/examples/customresourcedefinition/shirt-resource-definition.yaml

This command registers the Shirt custom resource definition (CRD) in your cluster. Now, Kubernetes can manage Shirt objects just like native resources.

Let’s create an actual Shirt instance:

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12

cat | kubectl apply -f - <<EOF
apiVersion: stable.example.com/v1
kind: Shirt
metadata:
  name: example1
  labels:
    # Disables Kubernetes Controller Manager operations on this instance
    skip-controller-manager-metadata-caching: "true"
spec:
  color: blue
  size: S
EOF

Once applied, the Kubernetes API server will accept the new Shirt object. The HariKube Middleware, guided by your TopologyConfig, will store this instance in the dedicated SQLite database. This demonstrates the full workflow: from defining a custom resource to automated, isolated database storage.

To verify everything is working, check the database for your new Shirt resource:

1

docker run -it --rm -v harikube_db:/data alpine/sqlite /data/shirts.db "select name from kine"

And that’s it! You’ve just set up dynamic database topology management for custom resources in Kubernetes. With HariKube, you can enjoy lower latency, higher throughput, true data isolation, virtually unlimited storage, and a much simpler development experience. HariKube supports both flat and hierarchical topologies, so you can organize your databases however you like.

Thank you for reading! If you have questions or ideas, please share them—we’d love to hear from you.