Skip to main content

Better Together: CAPI, Sveltos, and Cyclops for Automated Development Environments

· 11 min read
Eleni Grosdouli
Eleni Grosdouli
DevOps Consulting Engineer at Cisco Systems

Introduction​

Are you ready to simplify how your Platform team spins up and down development environments while improving DevX? In this post, we demonstrate how CAPI, Sveltos, and Cyclops work together to automatically create Kubernetes environments while allowing developers to interact and manage their applications in a user-friendly environment. It is not magic, it is the power of Sveltos combined with the right tooling!

title image reading "Sveltos, CAPI, Cyclops Examaple"

For the demo, I kept the costs to the minimum using Civo Cloud for the management cluster and Akamai/Linode for the CAPI deployments. Cilium is the preferred CNI for this setup.

Lab Setup​

+-----------------+-------------------------+--------------------------+
|   Cluster Name  |          Type           |         Version          |
+-----------------+-------------------------+--------------------------+
|      mgmt       |   Civo Medium Standard  |      1.30.5-k3s1         |
+-----------------+-------------------------+--------------------------+

+-----------------+-------------------------+
|    Deployment   |         Version         |
+-----------------+-------------------------+
|    Sveltos      |          v0.51.1        |
|   clusterctl    |          v1.9.6         |
|    Cilium       |          v1.16.4        |
+-----------------+-------------------------+

GitHub Resources​

The showcase repository is available here.

Prerequisites​

  1. A Kubernetes management cluster
  2. Akamai/Linode Cloud account
  3. Helm installed
  4. kubectl installed
  5. clusterctl installed

Scenario​

The idea is to demonstrate how Sveltos automates the deployment of different environments using blueprint configurations with the required global add-ons and applications while allowing users and developers to work in an intuitive UI to easily adjust the number of replicas, configure rolling updates, and monitor deployment progress. We combine the power of CAPI and Sveltos to look out for updates within a ConfigMap and once a change is detected, Sveltos spins up or down CAPI clusters on the Akamai/Linode cloud. When the CAPI cluster is ready, Sveltos deploys the required global configuration using a ClusterProfile like Kyverno, Grafana Prometheus, and Cyclops. We want to provide DevOps engineers and/or seasonal engineers the tools and the understanding to create and manage the lifecycle of a fleet of clusters in a meaningful and manageable way following the GitOps approach while allowing developers to focus on their main task of developement.

Install Sveltos Management Cluster​

On the management cluster, let's install Sveltos and Sveltos Dashboard. We will use the Helm chart installation for both components as it is easier to deploy and maintain. For Sveltos installation details, check out the link.

Sveltos​

$ helm repo add projectsveltos https://projectsveltos.github.io/helm-charts
$ helm repo update

$ helm install projectsveltos projectsveltos/projectsveltos -n projectsveltos --create-namespace
$ helm list -n projectsveltos

Sveltos Dashboard​

$ helm install sveltos-projectsveltos projectsveltos/sveltos-dashboard -n projectsveltos
$ helm list -n projectsveltos

Sveltos Label Management Cluster​

Sveltos can also manage add-ons and applications in the management cluster. The SveltosCluster instance mgmt in the mgmt namespace represents the management cluster which is automatically created by Sveltos upon installation. However, to manage this cluster, we will add the cluster label type=mgmt and Sveltos will be able to match it with a specific ClusterProfile. More details about the Sveltos resources at a later section.

$ kubectl label sveltoscluster -n mgmt mgmt type=mgmt
note

This is an important step. If the label is not included, Sveltos will not be able to match any clusters with the label set to type=mgmt.

Linode Environment​

Before we continue, ensure a Linode API token with the right permissions is available. Then export the required variables.

export LINODE_TOKEN=xxxxxxxxx
export LINODE_REGION=de-fra-2
export LINODE_CONTROL_PLANE_MACHINE_TYPE=g6-standard-2
export LINODE_MACHINE_TYPE=g6-standard-2
export CONTROL_PLANE_MACHINE_COUNT=1
export WORKER_MACHINE_COUNT=3

For the Linode regions and machine types, have a look here and here.

tip

g6-standard-2 is the smallest type I could spin up a CAPI cluster. Feel free to use a larger image if you prefer.

CAPI Preparation - Management Cluster​

Initialise Provider​

Looking at the official documentation, we need to initialise the Linode Provider for the required resources to be available at the management cluster.

$ clusterctl init --infrastructure linode-linode --addon helm
Fetching providers
Installing cert-manager version="v1.16.3"
Waiting for cert-manager to be available...
Installing provider="cluster-api" version="v1.9.6" targetNamespace="capi-system"
Installing provider="bootstrap-kubeadm" version="v1.9.6" targetNamespace="capi-kubeadm-bootstrap-system"
Installing provider="control-plane-kubeadm" version="v1.9.6" targetNamespace="capi-kubeadm-control-plane-system"
Installing provider="infrastructure-linode-linode" version="v0.8.6" targetNamespace="capl-system"
Installing provider="addon-helm" version="v0.3.1" targetNamespace="caaph-system"

Your management cluster has been initialized successfully!

You can now create your first workload cluster by running the following:

  clusterctl generate cluster [name] --kubernetes-version [version] | kubectl apply -f -

Create CAPI Cluster Configuration​

As long as CAPI is installed in the management cluster, we can continue and create the CAPI cluster configuration. Use the command below.

$ clusterctl generate cluster capi-sveltos01 --kubernetes-version v1.32.0 > capi-sveltos01.yaml

The capi-sveltos01.yaml file contains all Kubernetes resources required to create a CAPI cluster in the Linode cloud. The configuration can be updated based on your needs.

Sveltos Resources - Management Cluster​

To allow Sveltos to automatically spin up and down development environments, we will create and apply a ConfigMap which keeps track of the users needing a development environment. Then, we will convert the capi-sveltos01.yaml file into a Sveltos Template and later on create two Sveltos ClusterProfiles to control the creation/deletion of CAPI clusters alongside the global add-on and deployments on these clusters.

ConfigMap existing-users​

apiVersion: v1
kind: ConfigMap
metadata:
  name: existing-users
  namespace: default
$ kubectl apply -f existing_users.yaml

CAPI Linode Configuration​

apiVersion: v1
data:
  capi-linode.yaml: |
    {{- if index (getResource "Users") "data" }}
      {{- range $key, $value := (getResource "Users").data }}
        apiVersion: v1
        kind: Secret
        metadata:
          labels:
            clusterctl.cluster.x-k8s.io/move: "true"
          name: {{ $key }}-credentials
          namespace: default
        stringData:
          apiToken: xxx
          dnsToken: xxx
        apiVersion: kind.x-k8s.io/v1alpha4
        kind: Cluster
        name: kind-{{ $key }}
        networking:
          disableDefaultCNI: true
        nodes:
        - role: control-plane
          kubeadmConfigPatches:
          - |
            kind: InitConfiguration
            nodeRegistration:
              kubeletExtraArgs:
                node-ip: "172.0.0.3"
        containerdConfigPatches:
        - |-
          [plugins."io.containerd.grpc.v1.cri".registry.mirrors."localhost:5000"]
            endpoint = ["http://kind-registry:5000"]
        ...
kind: ConfigMap
metadata:
  name: deploy-linode-capi
  namespace: default
  annotations:
    projectsveltos.io/template: ok

As the Linode CAPI configuration is long, check out the GitHub repository reference for the complete example template. Effectively, we instruct Sveltos to iterate over the list of users within the existing-users ConfigMap and either create or remove a CAPI cluster. If a unique user is added, a new CAPI cluster will get deployed. If a user is removed from the ConfigMap, the CAPI cluster will get removed immediately! Take advantage of Sveltos power!

$ kubectl apply -f deploy_capi_linode.yaml
tip

The Cilium Helm Chart configuration was updated to support v1.16.4.

ClusterProfile - Linode CAPI Deployment​

apiVersion: config.projectsveltos.io/v1beta1
kind: ClusterProfile
metadata:
  name: deploy-linode-capi
spec:
  clusterSelector:
    matchLabels:
      type: mgmt
  templateResourceRefs:
  - resource:
      apiVersion: v1
      kind: ConfigMap
      name: existing-users
      namespace: default
    identifier: Users
  policyRefs:
  - name: deploy-linode-capi
    kind: ConfigMap
    namespace: default

The ClusterProfile gets deployed on the management cluster as the matchLabels are set to type: mgmt. Only the management cluster fits this rule. The ClusterProfile deploys a CAPI linode cluster once a new user is added to the ConfigMap with the name existing-users.

ClusterProfile - Global Deployments CAPI Linode Cluster​

apiVersion: config.projectsveltos.io/v1beta1
kind: ClusterProfile
metadata:
  name: deploy-capi-linode-managed
spec:
  clusterSelector:
    matchExpressions:
    - { key: type, operator: In, values: [ staging, production ] }
  syncMode: Continuous
  helmCharts:
  - repositoryURL:    https://kyverno.github.io/kyverno/
    repositoryName:   kyverno
    chartName:        kyverno/kyverno
    chartVersion:     v3.3.3
    releaseName:      kyverno-latest
    releaseNamespace: kyverno
    helmChartAction:  Install
  - repositoryURL:    https://prometheus-community.github.io/helm-charts
    repositoryName:   prometheus-community
    chartName:        prometheus-community/prometheus
    chartVersion:     26.0.0
    releaseName:      prometheus
    releaseNamespace: prometheus
    helmChartAction:  Install
  - repositoryURL:    https://grafana.github.io/helm-charts
    repositoryName:   grafana
    chartName:        grafana/grafana
    chartVersion:     8.6.4
    releaseName:      grafana
    releaseNamespace: grafana
    helmChartAction:  Install
  - repositoryURL: https://cyclops-ui.com/helm
    repositoryName: cyclops-ui
    chartName: cyclops-ui/cyclops
    chartVersion: 0.14.5
    releaseName: cyclops
    releaseNamespace: cyclops
    helmChartAction: Install
  policyRefs:
  - name: deploy-cyclops-template
    namespace: default
    kind: ConfigMap

The ClusterProfile installs Kyverno, Grafana Prometheus alongside Cyclops as Helm Charts and once the CAPI cluster is successfully deployed. For Cyclops, we have a custom template located in a ConfigMap named deploy-cyclops-template. Check out the Git repository for more details.

$ kubectl apply -f clusterprofile_linode.yaml,clusterprofile_linode_managed.yaml,cm_deploy_cyclops_template.yaml

Create CAPI Cluster​

To create a CAPI environment, we need to update the existing_users.yaml file and include the below entry.

data:
  user01: staging
$ kubectl apply -f existing_users.yaml

Once the file is saved and applied, Sveltos will notice the change and will start deploying a CAPI cluster with the name user01. The deployment might take some time. Be patient.

Validation​

kubectl Management Cluster​

Use the below commands to validate the status of the CAPI cluster.

$ kubectl get clusters
$ kubectl get kubeadmcontrolplane
$ kubectl describe kubeadmcontrolplane user01

kubectl CAPI Linode Cluster​

$ clusterctl get kubeconfig user01 > kube_user01.yaml # Collect user01 kubeconfig from management cluster
$ export KUBECONFIG=kube_user01.yaml # Connect to CAPI cluster

$ helm list -A
NAME                                     	NAMESPACE  	REVISION	UPDATED                                	STATUS  	CHART                                	APP VERSION
ccm-linode-1743419479                    	kube-system	1       	2025-03-31 11:11:20.876697746 +0000 UTC	deployed	ccm-linode-v0.4.25                   	v0.4.25    
cilium-1743419479                        	kube-system	1       	2025-03-31 11:11:20.427167074 +0000 UTC	deployed	cilium-1.16.4                        	1.16.4     
cyclops                                  	cyclops    	1       	2025-03-31 11:12:25.991307579 +0000 UTC	deployed	cyclops-0.14.5                       	v0.18.1    
grafana                                  	grafana    	1       	2025-03-31 11:12:24.155170354 +0000 UTC	deployed	grafana-8.6.4                        	11.3.1     
kyverno-latest                           	kyverno    	1       	2025-03-31 11:12:12.901992248 +0000 UTC	deployed	kyverno-3.3.3                        	v1.13.1    
linode-blockstorage-csi-driver-1743419479	kube-system	1       	2025-03-31 11:11:20.734904185 +0000 UTC	deployed	linode-blockstorage-csi-driver-v1.0.3	v1.0.3     
prometheus                               	prometheus 	1       	2025-03-31 11:12:20.376079741 +0000 UTC	deployed	prometheus-26.0.0                    	v3.0.0

Sveltos Dashboard​

CAPI clusters are automatically detected by Sveltos. If the right labels are applied to the CAPI cluster, Sveltos can match them and deploy add-ons and applications.

To login to the Sveltos Dashboard, have a look here. After deployment, we have the below view.

title image reading "Sveltos UI " title image reading "Sveltos UI " title image reading "Sveltos UI " title image reading "Sveltos UI "

Cyclops UI​

The developers can work on specific applications based on their needs in a user-friendly UI. For this demonstration, we imported a custom Helm Chart to the Cyclops setup.

title image reading "Cyclops UI" title image reading "Cyclops UI" title image reading "Cyclops UI" title image reading "Cyclops UI"

GitOps Integration​

Looking at the next steps, we can integrate ArgoCD into the deployment and include a GitOps approach towards Sveltos resource deployed at the management cluster. This requires additional steps to be taken, however, there might be a part 2 of this blog. The advantage of integrating ArgoCD is that every time a user updates the existing_users ConfigMap with new details, ArgoCD will synchronise with the repository and automatically push the updates to the management cluster. Then, Sveltos will notice the changes and take action.

Conclusion: CAPI, Sveltos and Cyclops Better Together​

The approach simplifies both development and platform administration. The Sveltos use case focuses on automating the creation of development environments across on-prem and cloud platforms. When combined with Cyclops, it offers a streamlined and intuitive interface, enabling developers to simplify complex operations while efficiently orchestrating applications. This not only makes platform administration easier but also provides developers and users with a user-friendly Kubernetes experience and global control over operations and user permissions.

Resources​

✉️ Contact​

We are here to help! Whether you have questions, or issues or need assistance, our Slack channel is the perfect place for you. Click here to join us us.

👏 Support this project​

Every contribution counts! If you enjoyed this article, check out the Projectsveltos GitHub repo. You can star 🌟 the project if you find it helpful.

The GitHub repo is a great resource for getting started with the project. It contains the code, documentation, and many more examples.

Thanks for reading!