Attention: This version of Event Streams has reached End of Support. For more information about supported versions, see the support matrix.

Helm commands fail when running as Team Administrator


Running a Helm command as a Team Administrator fails with the following error message:

Error: pods is forbidden: User "<username>" cannot list resource "pods" in API group "" in the namespace "kube-system"

Where <username> is the name of the user that logged in with the command cloudctl login.


Running any Helm command requires the user to have authorization to the kube-system namespace. The Team Administrator does not have access to the kube-system namespace.

Resolving the problem

The Helm Tiller service can also be configured to use a NodePort to bypass the default proxied connection and allowing users to use Helm CLI without requiring access to the kube-system namespace.

The default port is 31514. To retrieve the port value if not using the default number, log in as the Cluster Administrator by using the cloudctl login command, and run the following command:

kubectl get services -n kube-system tiller-deploy -o jsonpath='{.spec.ports[0].nodePort}'

The port number for the <tiller-nodeport> is displayed. Provide the port number to the Team Administrator.

The Team Administrator can then use the port number to run Helm commands in one of the following ways:

  • Export the HELM_HOST variable, for example:
     export HELM_HOST=<cluster-ip>:<tiller-nodeport>

    The Team Administrators can then run Helm commands.

  • Use the --host option when running Helm commands, and include the <cluster-ip>:<tiller-nodeport> value, for example:
     helm install --tls --host <cluster-ip>:<tiller-nodeport> install/delete/list ...

For more information about the workaround, see the IBM Cloud Private documentation.