Skip to content

Getting Started for Service Providers#

Quick start guide for platform engineers and administrators building solutions on SCO.

Sarah is a platform engineer at CloudCo who needs to offer managed services to her organization's development teams. SCO is deployed on their Kubernetes infrastructure, providing Virtual Machines and Hosted Kubernetes Clusters as flagship services. This guide walks through the essential steps to get started with SCO as a service provider.

Prerequisites#

Before you begin, ensure you have:

  • Kubernetes cluster (tested on OpenShift 4.14+ on bare metal)
  • SCO installed using ksp up CLI
  • Access to the SCO admin console
  • kubectl or oc configured with admin access to the cluster
  • Basic understanding of Kubernetes and CRDs
  • Familiarity with composition tools (Crossplane, KRO, or custom operators)

Step 1: Understanding Built-in Solutions#

SCO comes with flagship solutions out of the box:

Virtual Machine Solution#

The VirtualMachine solution leverages OpenShift Virtualization to provide VM-as-a-Service.

API Group: compute.cloud.stakater.com Kind: VirtualMachine

Users can provision VMs by creating a custom resource:

apiVersion: compute.cloud.stakater.com/v1
kind: VirtualMachine
metadata:
  name: my-dev-vm
spec:
  parameters:
    instanceType: o1.medium  # 2 vCPU, 4GB RAM
    connection: private
    sshPublicKey: "ssh-rsa AAAAB3Nza..."
    cloudInit:
      userData: |
        #cloud-config
        packages:
          - nginx

OpenShift Hosted Cluster Solution#

The OpenShiftCluster solution provides hosted OpenShift clusters on demand.

API Group: kubernetes.cloud.stakater.com Kind: OpenShiftCluster

Users can provision full OpenShift clusters:

apiVersion: kubernetes.cloud.stakater.com/v1
kind: OpenShiftCluster
metadata:
  name: dev-cluster
spec:
  parameters:
    clusterName: dev-cluster
    compute:
      cpu: 8
      memory: 32Gi
    nodePool:
      replicas: 3

Verify Available Solutions#

Sarah can check what solutions are available:

oc get xrds

Expected output shows the built-in solutions:

NAME                                                   ESTABLISHED   OFFERED   AGE
virtualmachines.vm.cloud.stakater.com                  True          True      10d
openshiftclusters.openshiftcluster.cloud.stakater.com  True          True      10d

Step 2: Publish to the Marketplace#

After installing SCO, the built-in solutions are automatically published to the marketplace. Cloud users can immediately browse and provision VMs and OpenShift clusters from their projects.

To publish additional custom solutions, see Publishing APIs for the full workflow.

Step 3: Onboard an Organisation#

Organisations provide isolation for different customers or business units. Each organisation gets its own isolated identity realm and project namespace.

Organisations are created through the platform onboarding workflow, which provisions the virtual API workspace, identity realm, and all required configuration automatically.

To manually onboard a new organisation, apply the organisation onboarding claim:

apiVersion: infrastructure.stakater.com/v1alpha1
kind: XOrgOnboarding
metadata:
  name: acme-corp
spec:
  providerConfigRef:
    name: kubernetes-provider
  kcpRootProviderConfigRef:
    name: kcp-root-provider
  parameters:
    organizationName: acme-corp
    adminEmail: admin@acmecorp.example.com

kubectl apply -f org-onboarding.yaml

This provisions:

  • An isolated identity realm for the organisation
  • A virtual API workspace hierarchy for the organisation and its projects
  • Initial admin credentials sent to the specified email

Verify the organisation was created:

kubectl get organizations

Step 4: Verify Cloud Users Can Access#

Once the organisation is created:

  1. Cloud users can log in at the organisation's console URL
  2. They can request projects from their organisation administrator
  3. The VirtualMachine and OpenShiftCluster solutions appear in their project marketplace
  4. They can provision services using kubectl, the console, Terraform, or GitOps

Verify the built-in solutions are published and available:

kubectl get xrds

Expected output:

NAME                                                          ESTABLISHED   OFFERED   AGE
virtualmachines.compute.cloud.stakater.com                    True          True      1d
openshiftclusters.kubernetes.cloud.stakater.com               True          True      1d

What You've Accomplished#

Sarah has now:

✅ Understood the built-in VM and OpenShift cluster solutions ✅ Created an organization for cloud users ✅ Verified solutions are available in the marketplace ✅ Set up cloud users to provision infrastructure and platform services

Next Steps#

Now that you have the basics, explore more advanced topics:

Common Tasks#

Add Custom Solutions: Create additional service offerings like databases (PostgreSQL, Redis), message queues (Kafka), or storage solutions.

Manage Multiple Organizations: Create separate organizations for different customers or business units.

Configure RBAC: Define who can create solutions and manage organizations.

Monitor Usage: Track VM, cluster, and project resource usage across organizations.

Customize VM Instance Types: Modify available instance types and sizing options.

What's Next?#