You deploy a Kubernetes cluster on an OpenStack or SDDC provider. VMware Integrated OpenStack with Kubernetes supports exclusive and shared cluster types.

Prerequisites

  • Verify that the cloud provider you deployed is active. See Adding a Cloud Provider.

  • Determine the type of cluster you want to add.

Procedure

  1. Login to VMware Integrated OpenStack with Kubernetes.
  2. On the Clusters home page, click Deploy a New Cluster or click +NEW to add a cluster to a set of existing clusters.
  3. On the Intro page, click Next.

    Alternatively, you can click Choose File to upload a JSON file containing cluster configuration information. The information automatically populates fields in the subsequent wizard screens.

  4. Highlight the infrastructure provider and click Next.
  5. On the Configure node group page, configure the master and worker node groups.
    1. Specify the number of master nodes in the cluster. The default node profile is selected. To select a different node profile, uncheck the box.
      Note:

      Because etcd servers coexist on master nodes, you must specify an odd number of nodes to support high availability and fault tolerance.

      A table appears listing a selection of node profiles.

    2. Specify the number of worker nodes in the cluster. The default node profile is selected. To select a different node profile, uncheck the box.

      A table appears listing a selection of node profiles.

      Note:

      If the group includes multiple nodes, and you want to use a worker node configured with an extra network in the ENS transport zone, select the node profile labeled node_profile=high_performance. To use this network in a pod deployment with multiple NICs, you must assign your pod to this node by adding the following field setting to your pod configuration:

      nodeSelector:
          node_profile: high_performance
    3. Click Next.

    For more information about node profiles, see Customizing Your Cluster.

  6. Configure the cluster.
    1. Provide a cluster name.
    2. (optional) Specify the DNS servers.
    3. (optional) Specify the IP address or FQDN for the Log Insight server configured to receive logs from the Kubernetes cluster.
    4. Select the Shared Cluster or Exclusive Cluster type.
      • For a shared cluster, specify a namespace.

      • For an exclusive cluster, specify a user and group.

    5. Provide information to create an NSX-T external network IP pool with addresses for the cluster.
      Note:

      If the infrastructure provider is configured with NSX-T networking, every Kubernetes cluster should use its own pool. NSX-T Backend.

      • For the external IP pool, specify a CIDR value with a network address and not a host address. For example, 4.3.0.0/16.

      • Specify the range of addresses to be allocated for the IP pool.

      • (optional) Specify the gateway for the IP pool.

      • (optional) Specify the DNS server IP.

  7. Click Next.
  8. Review the Configuration Summary and click Finish to add the cluster