Kubeslice Worker Operator

The Kubeslice Worker Operator, also known as Slice Operator manages the lifecycle of KubeSlice worker cluster-related custom resource definitions (CRDs). The kubeslice-worker operator uses Kubebuilder, a framework for building Kubernetes APIs using CRDS.

Get Started

It is strongly recommended that you use a released version.

Please refer to our documentation on:

• Get Started on KubeSlice
• Install KubeSlice

Install kubeslice-worker on a Kind Cluster

Prerequisites

Before you begin, make sure the following prerequisites are met:

• Docker is installed and running on your local machine.
• A running kind cluster.
• kubectl is installed and configured.
• You have prepared the environment to install kubeslice-controller on the controller cluster and worker-operator on the worker cluster. For more information, see Prerequisites.

Build and Deploy a Worker Operator on a Kind Cluster

To download the latest Worker Operator docker image, click here.

docker pull aveshasystems/worker-operator:latest

Setting up Your Helm Repo

If you have not added avesha helm repo yet, add it.

helm repo add avesha https://kubeslice.github.io/charts/

Upgrade the avesha helm repo.

helm repo update

Get Secrets from the Controller Cluster (if it's not already done)

The following command will get the relevant secrets from the controller cluster and copy them to the secrets folder. Additionally, it will return the secrets so that we can use them to populate the helm chart values.

deploy/controller_secret.sh [controller_cluster_context] [project_namespace] [worker_cluster_name]

Example

deploy/controller_secret.sh gke_avesha-dev_us-east1-c_xxxx kubeslice-cisco my-awesome-cluster

Build Docker Images

1. Clone the latest version of worker-operator from the master branch.

   git clone https://github.com/kubeslice/worker-operator.git
   cd worker-operator

2. Edit the VERSION variable in the Makefile to change the docker tag to be built. The image is set as docker.io/aveshasystems/worker-operator:$(VERSION) in the Makefile. Modify this if required.

   make docker-build

Run the Local Image on a Kind Cluster

1. You can load the Worker Operator on your kind cluster using the following command:

   kind load docker-image <my-custom-image>:<unique-tag> --name <cluster-name>

   Example:

   kind load docker-image aveshasystems/worker-operator:1.2.1 --name kind

2. Check the loaded image in the cluster. Modify the node name if required.

   docker exec -it <node-name> crictl images

   Example:

   docker exec -it kind-control-plane crictl images

Deploy the Worker Operator on a Cluster

Create a chart values file called yourvaluesfile.yaml. Refer to values.yaml to create yourvaluesfile.yaml and update the operator image subsection to use the local image.

From the sample:

operator:
  image: docker.io/aveshasystems/worker-operator
  tag: 0.2.3

Change it to:

operator:
  image: <my-custom-image> 
  tag: <unique-tag>

Deploy the Updated Chart

make chart-deploy VALUESFILE=yourvaluesfile.yaml

Verify the Installation

Verify the installation by checking the status of pods belonging to the kubeslice-system namespace.

kubectl get pods -n kubeslice-system

Example output

NAME                                     READY   STATUS    RESTARTS   AGE
jaeger-65c6b7f5dd-frxtx                  1/1     Running   0          49s
kubeslice-netop-g4hqd                    1/1     Running   0          49s
kubeslice-operator-6844b47cf8-c8lv2      2/2     Running   0          48s
mesh-dns-65fd8585ff-nlp5h                1/1     Running   0          48s
nsm-admission-webhook-7b848ffc4b-dhn96   1/1     Running   0          48s
nsm-kernel-forwarder-fd74h               1/1     Running   0          49s
nsm-kernel-forwarder-vvrp6               1/1     Running   0          49s
nsmgr-62kdk                              3/3     Running   0          48s
nsmgr-7dh2w                              3/3     Running   0          48s
prefix-service-76bd89c44f-2p6dw          1/1     Running   0          48s

Uninstall the Worker Operator

For more information, see deregister the worker cluster.

helm uninstall kubeslice-worker -n kubeslice-system

License

Apache License 2.0