Skip to main content

Practical Use Cases

Use this page when you need to connect the UI to concrete host-side outcomes.

Bring Up A Local SNO Cluster​

Use this when you want a local OpenShift SNO environment on one KVM host without losing the install logic in a one-off shell session.

  1. Open the Cockpit plugin on the host.
  2. Choose the single-node control plane shape.
  3. Supply cluster identity, pull secret, SSH key, DNS, VIP, and static node networking.
  4. Review generated install-config.yaml, agent-config.yaml, guest plan, static-network-configs.yaml, and the virt-install plan.
  5. Launch deployment from the same workspace.

The UI preserves the generated installer inputs and keeps the deployment step tied to the host model that will run it.

Bring Up A Compact Cluster​

Use this when you need a three-control-plane local OpenShift footprint and still want a guided local workflow.

  1. Use the create flow.
  2. Select 3 control plane nodes.
  3. Set the host sizing and bridge model.
  4. Review the generated node plan, discovery plan, and static host networking YAML.
  5. Deploy from the final review step.

The plugin keeps cluster intent, generated YAML, and VM plan review in one operator surface instead of scattering them across local files and terminal history.

Review Generated Artifacts​

The dangerous part of local OpenShift bring-up is deploying with the wrong network values, VIPs, or guest shape.

  1. Stay in the create workflow until the review step.
  2. Inspect rendered install-config.yaml.
  3. Inspect rendered agent-config.yaml.
  4. Inspect static-network-configs.yaml, the guest plan, and the virt-install plan.
  5. Deploy only after the values match the intended host plan.

The review step turns the plugin into a verification surface, not just a form.

Return Later To See What Exists​

After deployment, use the cluster list in Cockpit to find cluster type, creation time, version, provider identity, and next actions.

  1. Return to the cluster list.
  2. Filter the fleet view by name or cluster type.
  3. Open the cluster overview page for one cluster.
  4. Use the overview details, notices, and actions instead of reconstructing state from libvirt manually.

Reprovision Or Destroy​

Local labs and proof-of-concept environments are often reprovisioned after destructive testing. Use the same tool that created the cluster for cleanup.

  1. Return to cluster inventory or cluster overview.
  2. Choose Reprovision to recreate the cluster from saved configuration, or choose Destroy to remove it.
  3. Let the local backend handle the lifecycle path it already understands.

Creation, reprovision, and cleanup stay in one operational boundary, reducing drift between the install path and the lifecycle actions.