vRealize Automation 7.5 and Pivotal Container Service Integration

Over the past month I’ve been exploring the Application Platform and Cloud Native space, deploying NSX-T, Pivotal Container Service (PKS) and Istio in the lab as I prepared to co-present with Annie Lin (Application Platform Architect) at vForum Sydney. Must say, it’s been pretty awesome to get my hands dirty with something different! Especially trying to wrap my head around Service Mesh and what this means for developers and operations teams.

Although not related to what I was doing for vForum, with the release of vRealize Automation (vRA) 7.5 which includes native support for creating and managing PKS clusters, I thought I’ll share a quick walk-through on how to configure a PKS endpoint, create and import existing clusters as I couldn’t find anything intuitive besides VMware’s official documentation.

The integration allows users to consume and manage PKS clusters through a self-service portal or a common API. This might be useful for users that don’t care about the PKS CLI, the portal also allows users to download the kubectl config.


  1. vRA 7.5
    • Container administrator privilege
  2. PKS 1.2
    • PKS API address
    • PKS credentials
    • UAA address

Adding PKS Endpoint

  1. Login to vRA and navigate to “Containers” > “Identity Management”

  2. Click “+Credentials”, enter PKS credentials and save
  3. Navigate to “PKS Endpoints” and click “New Endpoint”
  4. Enter “Name”, “UAA Address”, “PKS Address” and select the credentials created in the previous step. Note: the UAA port is 8443 and PKS address is 9021
  5. Click “Test Connection” and accept any certificate warning

  6. Click “Plan Assignments” and map the appropriate business group to plan

  7. Click “Save”

Importing Existing PKS Clusters

Naturally, because PKS has been around much longer than vRA 7.5 you’ll probably have existing clusters. Here’s how to import them.

  1. Navigate to “PKS Clusters” and click “Add Cluster”
  2. Select your endpoint from the “PKS Endpoint” dropdown list
  3. Tick the cluster name to import, select the “Connect By” preference and click “Add”. In my case because the cluster VIP was added to DNS I selected “Master Host Name”

  4. After successfully adding you should see it appear in the “PKS Clusters” page

  5. You can download the “kubeconfig” or view the cluster summary by clicking on it

  6. Click on “Nodes” to view their status

Creating New PKS Clusters

  1. Navigate to “PKS Clusters” and click “New Cluster”
  2. Select your endpoint from the “PKS Endpoint” dropdown list
  3. Enter the “Cluster Name”, this is the short name
  4. Select a plan from the “Plan” dropdown list
  5. Enter the “Master Host Name”, this is the long name (FQDN) which is equivalent to “–external-hostname” when using the PKS CLI to create a new cluster
  6. Select “Connect By” preference and click “Add”. Note: If you connect by using “Master Host Name” a DNS entry must exist for the cluster VIP, otherwise select “Master Node IP”

  7. The new cluster should be created shortly. To view the status, navigate to “Requests” or if you have access to the PKS CLI you can run the command to view the status pks cluster [clustername]

  8. After the new cluster has successfully finished creating it will appear in the “PKS Clusters” page. You can then download “kubeconfig”, view the cluster or nodes summary as per previous step when we imported an existing cluster

That’s it! Apologies for the page layout, we are still fine tuning the styles/themes.