Creating a Kubernetes Configuration File

Set up a Kubernetes configuration file for each OKE cluster that you work with. Your Kubernetes configuration file enables you to access OKE clusters using the kubectl command and the Kubernetes Dashboard.

Kubernetes configuration files organize information about clusters, users, namespaces, and authentication mechanisms. You can define contexts to easily switch between clusters and namespaces. The kubectl tool uses Kubernetes configuration files to find the information it needs to choose a cluster and communicate with the API server of a cluster.

Installing the Kubernetes Command Line Tool

Install and configure the Kubernetes command line tool kubectl. The kubectl tool enables you to perform operations on OKE clusters such as deploy applications, inspect and manage cluster resources, and view logs.

To install kubectl, see https://kubernetes.io/docs/tasks/tools/. The kubectl version must be within one minor version of the OKE cluster Kubernetes version. For example, a v1.29 client can communicate with v1.28, v1.29, and v1.30 control planes. See Supported Versions of Kubernetes.

For more information, including a complete list of kubectl operations, see the Command line tool (kubectl) reference page.

Creating a Kubernetes Configuration File

Use the OCI CLI to create your Kubernetes configuration file.

Tip:

The Quick Start button on a cluster details page in the Compute Web UI shows how to create a Kubernetes configuration file, and provides the OCID of the cluster.

Get the OCID of the cluster: oci ce cluster list
Run the command to create the configuration file.

The --cluster-id option is the only required option.

The default value of the --file option is ~/.kube/config. If you already have a file at the specified location and you want to replace it, use the --overwrite option. To maintain more than one configuration file, select a different file by using the KUBECONFIG environment variable or the --kubeconfig option.

The value of the --kube-endpoint option must be PUBLIC_ENDPOINT.

If provided, the value of the --token-version option must be 2.0.0.

Example:

Use the following command to configure a Kubeconfig file for the specified cluster using the public endpoint:
```
$ oci ce cluster create-kubeconfig --cluster-id ocid1.cluster.unique_ID \
--file $HOME/.kube/config --kube-endpoint PUBLIC_ENDPOINT
New config written to the Kubeconfig file /home/username/.kube/config
```
Use the following command to set your KUBECONFIG environment variable to the Kubeconfig file that you created or updated in the preceding command:
```
$ export KUBECONFIG=$HOME/.kube/config
```
The following command shows the content of your new YAML configuration file:
```
$ kubectl config view
```
If you run the command again with a different cluster OCID, the new information is merged with the existing information. The following message is displayed:
```
Existing Kubeconfig file found at /home/username/.kube/config and new config merged into it
```

Verify Your Cluster Access

Run the following command to confirm that you can access your cluster:

$ kubectl cluster-info

Every Kubernetes namespace contains at least one ServiceAccount: the default ServiceAccount for that namespace, which is named default. If you do not specify a ServiceAccount when you create a Pod, the OKE service automatically assigns the ServiceAccount named default in that namespace.

An application running inside a Pod can access the Kubernetes API using automatically mounted service account credentials.