examples/chart/teleport-cluster/README.md
This chart sets up a Teleport cluster composed of at least 1 Proxy instance and 1 Auth instance. When applicable, the chart will default to 2 pods to provide high-availability.
Ingress requires ALPN support.
Exposing Teleport through a Service with type LoadBalancer is still recommended
because its the most flexible and least complex setup.To install Teleport in a separate namespace and provision a web certificate using Let's Encrypt, run:
$ helm install teleport/teleport-cluster \
--set acme=true \
--set [email protected] \
--set clusterName=teleport.example.com\
--create-namespace \
--namespace=teleport-cluster \
./teleport-cluster/
Finally, configure the DNS for teleport.example.com to point to the newly created LoadBalancer.
Note: this guide uses the built-in ACME client to get certificates.
In this setup, Teleport nodes cannot be replicated. If you want to run multiple
Teleport replicas, you must provide a certificate through tls.existingSecretName
or by installing cert-manager and setting the highAvailability.certManager.* values.
The first user can be created by executing a command in one of the auth pods.
kubectl exec it -n teleport-cluster statefulset/teleport-cluster-auth -- tctl users add my-username --roles=editor,auditor,access
The command should output a registration link to finalize the user creation.
helm uninstall --namespace teleport-cluster teleport-cluster
See https://goteleport.com/docs/admin-guides/deploy-a-cluster/helm-deployments/ for guides on setting up HA Teleport clusters in EKS or GKE, plus a comprehensive chart reference.
Please read CONTRIBUTING.md before raising a pull request to this chart.