cluster-vsphere

This repository contains the Helm chart used for deploying CAPV clusters using CAPI. The chart only contains the CAPV-specific components; the core CAPI compnents are deployed via the shared giantswarm/cluster chart.

IP address management (IPAM) for workload clusters

When creating a new vSphere cluster, a user can put an empty string to .connectivity.network.controlPlaneEndpoint.host and at the same time specify the .connectivity.network.controlPlaneEndpoint.ipPoolName. In this case, the cluster will be created in the paused: true state and a post-install job will be spawned. The goal of this job is to acquire the new IP address and assign it to .spec.controlPlaneEndpoint.host of newly created clusters and other places where it's needed (kube-vip static pod definition and certSANs). Only then the cluster is unpaused.

The abovementioned mechanism relies on IpAddressClaim and IpAddress CRDs. These are part of the Cluster API spec and can be reconciled for instance by cluster-api-ipam-provider-in-cluster. So if you want to use it, make sure this app is also installed in the management cluster.

Tagging VMs

It is possible to apply vSphere tags to the virtual machines deployed by this chart.

• First find the ID of the tags to apply.

Powershell:
get-tag -name <tag name> -category <tag category> | select -ExpandProperty id

govc:
govc tags.info -c <tag category> -json | jq -r ".[] | select(.name == \"<tag name>") | .id"

The ID should look like this: urn:vmomi:InventoryServiceTag:a5242a12-87e4-4954-b357-d711c99c91e5:GLOBAL

• Set the values where you want to apply the tag.

    global:
      controlPlane:
        machineTemplate:
          tagIDs:
          - "urn:vmomi:InventoryServiceTag:827c9b5a-e6e5-4c3f-8349-29c083395a7f:GLOBAL"
      nodePools:
        <nodepool name>:
          tagIDs:
          - "urn:vmomi:InventoryServiceTag:827c9b5a-e6e5-4c3f-8349-29c083395a7f:GLOBAL"

Using a VM template per location

When there are clusters in different geographical locations under the same datacenter entity, it is recommended to use a VM template stored in that cluster to avoid cloning over a WAN connection.

In this case:

• Set e.g. global.providerSpecific.templateSuffix=paris
• Upload the VM image template to the Paris cluster and rename it to add the -paris suffix.

Using Kamaji as the control plane provider

By default the control plane is provisioned as vSphere VMs via KubeadmControlPlane. Alternatively, the control plane can be hosted by Kamaji, which runs the control plane components as pods on the management cluster. The kamaji app must be installed in the management cluster beforehand.

To enable it, set the following values:

    cluster:
      providerIntegration:
        resourcesApi:
          controlPlaneResource:
            enabled: true
            provider: kamaji
        controlPlane:
          resources:
            controlPlane:
              api:
                group: controlplane.cluster.x-k8s.io
                kind: KamajiControlPlane
                version: v1alpha1

With these values applied, the chart renders a KamajiControlPlane resource instead of a KubeadmControlPlane, and no control plane VMs are created in vSphere.