Deploying IBM Cloud Kubernetes Service template

Deploy and configure IBM Cloud Kubernetes Service template in Managed services.

Before you begin

• You must have administrator privileges for the deployment of IBM Cloud Kubernetes Service cluster and charts into that cluster. If you do not have the privileges, contact your administrator.

• If you need to initialize helm for clusters that were not created by Managed services, then see Deploying helm-tiller template to add the tiller deployment. For more information about Helm init, see public Helm documentation for Helm init.

Procedure

1. Log in to Managed services.
2. From Managed services menu, go to Library > Services.
3. In the Services pane, click Create a Service.
4. Enter your service name and category and click Create.
5. Go to the Composition tab, drag IBM Cloud Kubernetes Service from the Templates > IBM section. In Managed services IBM Cloud Kubernetes Service is available in the Library > Terraform templates page.
6. In the Basic information tab > Connections, select the IBM Cloud connection from the drop-down list. For more information about IBM Cloud connection, see Configuring an IBM Cloud connection. Use this connection to deploy IBM Cloud Kubernetes Service cluster from Managed services. If you are using a IBM Cloud Kubernetes cluster that is already available in IBM public cloud, see Configuring IBM Cloud Kubernetes Service.
7. Drag any Helm chart based on your requirement.
8. In the Basic information tab > Connection, pass the output of IBM Cloud Kubernetes Service Terraform template activity in the Connection field. The format to provide the value is ${templates.<activity id>.output.cluster_name. For example, ${templates.ibmcloudcf954ae7.output.cluster_name}. The cloud connection is auto-generated.

9. In the Parameters tab, select your namespace. The values vary based on your connection.

   Note: The image-repository is a read-only value. As long as the file is accessible by the cluster, it is usable by the Helm chart. You can deploy a chart that point to images in the public cloud repository. IBM Cloud Kubernetes Service has its own repository of Helm charts. If you want to integrate that, then go to IBM Cloud Pak® for Multicloud Management console and add the repository. To configure repositories in IBM Cloud Pak® for Multicloud Management, do the following steps:

   1. Go to IBM Cloud Pak® for Multicloud Management.
   2. Click Manage > Helm repositories. You can see a list of Helm repositories that are configured for IBM Cloud Pak® for Multicloud Management. You can click Sync repositories to add new repositories. For more information about configuring your repositories in IBM Cloud Pak® for Multicloud Management, see IBM Cloud Pak® for Multicloud Management IBM Documentation.

10. Save and deploy the service. It creates an IBM Cloud Kubernetes Service cluster and installs Helm tiller on the cluster automatically. This fully integrated and initialized cluster is available for Managed services to deploy Helm charts.

    Note: If the following error message is displayed in the activity logs, then ensure that the helm tiller is configured on their IBM Kubernetes Service cluster: Error: could not find tiller

11. Optionally, when you want to connect workloads in public and private cloud, then deploy VPN Helm charts into the IBM Cloud Kubernetes cluster. You can also create a new service to do the VPN Helm charts deployment and configure a secure connection between the external cluster and IBM Cloud Pak® for Multicloud Management.

    Note**: Workaround for strongSwan Helm Chart to work on IBM Cloud Pak® for Multicloud Management, see strongSwan Helm Chart on IBM Cloud Pak® for Multicloud Management section.

12. Manage clusters from Kubernetes dashboard:

    1. After you deploy the service, go to Deployed instances and open your service.
    2. Go to the Overview tab > Activity instances section.
    3. Click the instance name link. It takes you to the IBM Cloud console where the cluster is running.
    4. Click Kubernetes dashboard.
    5. Select your namespace. All of the active workloads are displayed. You can view running pods, discover and perform load balancing, scale deployments, and add replicas from the dashboard.