No More Backdoors: Know Who Has Access to What, Right Now
Jun 13
Register Today
Teleport logoTry For Free
Home > Teleport Academy > Authentication and Privileges

Authenticating AWS EKS Kubernetes Clusters with GitHub SSO

Posted 4th Apr 2023 by Travis Rodgers


In this tutorial, we’ll look at how to authenticate AWS EKS Kubernetes clusters with GitHub single sign-on (SSO). Amazon EKS is Amazon’s managed Kubernetes service that makes it easy and efficient for you to run Kubernetes on AWS and on-premises. The convenience of having Amazon manage the availability, performance, and scalability of nodes in addition to direct integration with AWS networking and security has made AWS EKS the preferred way of running Kubernetes for many organizations.

In addition, these organizations are adopting SSO solutions to centralize identity management for better security, experience and efficiency. This post will provide instructions on how to utilize GitHub SSO to secure access to AWS EKS Kubernetes clusters. Sitting in between these two services will be Teleport, an open-source, identity-native infrastructure access platform that allows engineers and machines to securely connect to any infrastructure resource in the world, including EKS clusters. Teleport also provides us with, amongst other types of auth connectors like SAML and OIDC, a GitHub SSO connector so that we can bridge these two services together.

Key takeaways:

You’ll learn how to:

  • Quickly provision an Amazon EKS cluster
  • Enroll and protect access to EKS with Teleport
  • Configure RBAC for Kubernetes
  • Configure GitHub SSO and map users and groups
  • Audit access to the Kubernetes API including kubectl exec session recordings

Prerequisites: This article assumes that you are familiar with, and have deployed, a Teleport cluster and can log into it.

Step 1 - Provisioning an Amazon EKS cluster

First, you need to launch an EKS cluster. Here, we’ll launch two Ubuntu nodes with version 1.23 of Kubernetes. You may need to replace the parameters such as the region and node type. Download the CLI, eksctl, and configure AWS CLI for your AWS account to follow along.

touch eks-config
export KUBECONFIG=$PWD/eks-config

eksctl create cluster \
	--name cluster1 \
	--region us-west-2 \
	--version 1.23 \
	--nodegroup-name ng-workers \
	--node-type t3.medium \
	--nodes 2 \
	--nodes-min 2 \
	--nodes-max 3 \
	--node-ami-family Ubuntu2004 \

When the provisioning is complete, the kubeconfig contents are written to the eks-config file in the current directory. You can list the nodes and verify the access to the cluster.

Step 2 - Registering Amazon EKS Kubernetes cluster with Teleport

Next, we can easily and securely manage our EKS Kubernetes cluster access with Teleport, which also conveniently provides us with a GitHub SSO connector which we’ll look at in the next section.

Similar to other configurations, such as servers, Teleport expects an agent to run within the target cluster. The teleport-kube-agent can be installed through a Helm chart by pointing it to the Teleport proxy server endpoint.

Before proceeding further, we’ll need to get the token responsible for validating the agent.

First, make sure you log in to Teleport as a user with roles editor and access.

Next, use the tctl command to create a token based on the Kubernetes role and save to a TOKEN variable.

TOKEN=$(tctl nodes add --roles=kube --ttl=1h --format=json | jq -r '.[0]')

The next step is to add Teleport’s repo and update Helm, which provides us access to the Helm chart.

helm repo add teleport
helm repo update

Finally, set the variables as seen below and run the helm install command with the specified parameters to deploy the teleport-kube-agent to your EKS cluster.

helm install teleport-agent teleport/teleport-kube-agent \
  --set kubeClusterName=${CLUSTER?} \
  --set proxyAddr=${PROXY_ADDR?} \
  --set authToken=${TOKEN?} \
  --create-namespace \
  --namespace=teleport \
  --version 12.1.1

Wait for a few minutes, and check if the agent pod is up and running within the teleport namespace.

You can also verify the logs with the below command:

kubectl logs teleport-agent-0 -n teleport

Step 3 - Configuring Teleport RBAC to allow access to Amazon EKS Kubernetes cluster

The cluster is now registered, but not accessible. For our user (or any user) to access the cluster, we’ll need to configure Teleport’s RBAC.

Kubernetes uses an identity embedded within the kubeconfig file to access the cluster. Let’s create a new Teleport role adding that identity to the kubernetes_users section for the user to assume the role.

First, let’s get the current user from the kubeconfig file.

kubectl config view  -o jsonpath="{.contexts[?(\"$(kubectl config current-context)\")].context.user}"

By default, EKS creates a user based on the format Let’s tell Teleport that this user will assume the role by creating an RBAC definition in a file by name of kube-access.yaml and applying it to create the new teleport role named ‘kube-access’ with the user listed under kubernetes_user.

kind: role
  name: kube-access
version: v5
      '*': '*'
    - viewers
    - [email protected]
  deny: {}

Apply your changes to create the role:

tctl create -f kube-access.yaml

Now assign the kube-access role to your Teleport user by first retrieving your local user’s configuration and outputting to a file:

tctl get users/$(tsh status -f json | jq -r '.active.username') > user.yaml

Then editing this file, user.yaml, adding kube-access to the list of existing roles:

   - access
   - auditor
   - editor
+  - kube-access

Now apply your changes:

tctl create -f user.yaml

Log out of your Teleport cluster and log in again to assume the new role.

Now that Teleport RBAC is configured, you can authenticate to your Kubernetes cluster via Teleport. But to interact with your Kubernetes cluster, you will also need to configure authorization within Kubernetes.

To configure authorization within your Kubernetes cluster, you need to create Kubernetes RoleBindings or ClusterRoleBindings that grant permissions to the subjects listed in kubernetes_users and kubernetes_groups in the Teleport role.

For example, you can grant some limited read-only permissions to the viewers group used in the kube-access role defined above:

Create a file called viewers-bind.yaml with the following contents:

kind: ClusterRoleBinding
  name: viewers-crb
- kind: Group
  # Bind the group "viewers", corresponding to the kubernetes_groups we assigned our "kube-access" role above
  name: viewers
  kind: ClusterRole
  # "view" is a default ClusterRole that grants read-only access to resources
  # See:
  name: view

Apply the ClusterRoleBinding with kubectl:

kubectl apply -f viewers-bind.yaml

Log out of Teleport and log in again.

List connected clusters using tsh kube ls and switch between them using tsh kube login:

Step 4 - Configuring GitHub as SSO provider for Amazon EKS Kubernetes cluster

Now that we have access to our Amazon EKS Cluster via kubectl, let’s create a GitHub Single Sign On (SSO) connection within Teleport so you can automatically map teams in your Github organization to users and roles in Teleport.

As a prerequisite you’ll need a GitHub organization with at least one team. This organization must not have external SSO set up, or Teleport will refuse to create the GitHub authentication connector. Follow the two steps below:

1. Create a GitHub OAuth app:

Within GitHub, create an OAuth app. Instructions for creating this can be found here:

Creating an OAuth App

When you do so ensure that your OAuth App's "Authentication callback URL" is the following:


PROXY_ADDRESS must be the public address of the Teleport Proxy Service.

2. Create a GitHub authentication connector

Define a GitHub authentication connector by creating a file called github.yaml with the following content:

kind: github
version: v3
  # Connector name that will be used with `tsh --auth=github login`
  name: github
  # Client ID of your GitHub OAuth App
  client_id: <client-id>
  # Client secret of your GitHub OAuth App
  client_secret: <client-secret>
  # Connector display name that will be shown on the Web UI login screen
  display: GitHub
   # URL of your GitHub Enterprise Server instance (if applicable)
  # endpoint_url: https://<github-enterprise-server-address>
  # Callback URL that will be called after successful authentication
  redirect_url: https://<proxy-address>/v1/webapi/github/callback
  # Mapping of org/team memberships onto allowed roles
    - organization: octocats # GitHub organization name
      team: admins # GitHub team name within that organization
      # Maps octocats/admins to the "access" Teleport role
        - access

The values of client_id, client_secret, and redirect_url come from the GitHub OAuth App you created earlier.

Teleport will request only the read:org OAuth scope. Read more about OAuth scopes in GitHub's documentation.

Finally, once you’ve filled in the appropriate values, create the connector using tctl:

tctl create github.yaml

Step 5 - Accessing EKS Kubernetes clusters through GitHub SSO Identity Provider

Now let’s test the entire flow by signing into Teleport as a GitHub user and then using the tsh CLI to list the registered Kubernetes clusters.

First, let’s log into Teleport with GitHub as the authentication.

tsh login --auth=github

This opens up a browser window for entering your credentials. Once entered, you should get a prompt to authorize the access:

Once authorized, the CLI confirms the identity with relevant data about the user:

Now let’s list our Kubernetes clusters:

And finally, let’s login so that we can view pods:

And remember that our role only allows us the view permission — this is why we are able to list pods. Let’s make sure that we aren’t able to do anything else:

This results in error due to the lack of permission to create resources.

Since the action violates the access policy defined by the RBAC, it is also visible in Teleport’s audit logs where it gives me the timestamp, user, action (POST request) and more:

We can define fine-grained RBAC policies that map the users from the GitHub groups to Teleport users to Kubernetes roles. This gives ultimate control to cluster administrators and DevOps teams to allow or restrict access to Kubernetes resources.

Auditing kubectl execs

In addition to cluster activity, Teleport will also record every kubectl exec session that can then be played back on demand in a video player. This allows auditors to see keystrokes that a user made for the entire duration of the session. Teleport can then store these recorded sessions in a local filesystem or an external source such as an AWS S3 bucket.


In this tutorial, we have learned how to leverage GitHub as an SSO provider for Teleport to define access policies for Amazon EKS-based Kubernetes clusters.

Other related posts:

Try using Teleport to connect GitHub with your EKS cluster, by signing up for our 14-day Teleport Cloud trial.