Skip to content

Accessing a Cluster

Clusters created or joined using the DCE 5.0 Container Management platform can be accessed not only through the UI interface but also through CloudShell and kubectl.

To access the cluster, the user should have the Cluster Admin permission or higher.

Access via CloudShell

  1. On the Cluster List page, select the cluster that needs to be accessed through CloudShell, click the ... operation icon on the right, and click Console in the drop-down list.

  2. Execute the kubectl get node command on the CloudShell console to verify the connectivity between CloudShell and the cluster. The console will return the node information under the cluster.

Now you can access and manage the cluster through CloudShell.

Access via kubectl

When accessing and managing cloud clusters through local nodes, the following requirements need to be met:

  • Network interconnection between local nodes and cloud clusters.
  • The cluster certificate has been downloaded to the local node.
  • The kubectl tool has been installed on the local node. For detailed installation methods, refer to Installing kubectl.

  • On the Cluster List page, select the cluster that needs to download the certificate, click ... on the right, and click Get Certificate in the pop-up menu.

  • Select the certificate validity period and click Download Certificate.

  • Open the downloaded cluster certificate and copy the content of the certificate to the config file of the local node.

    By default, the kubectl tool will look for a file named config from the $HOME/.kube directory of the local node. This file stores the access credentials of the relevant cluster, and kubectl can connect to the cluster with this configuration file.

  • Run the following command on the local node to verify the connectivity of the cluster:

    kubectl get pod -n default
    

    The expected output is something like:

    NAME READY STATUS RESTARTS AGE
    dao-2048-2048-58c7f7fc5-mq7h4 1/1 Running 0 30h
    

Now you can access and manage the cluster locally with kubectl.

Comments