Teleport
Teleport EKS Auto-Discovery
Version preview- Older Versions
EKS Auto-Discovery can automatically discover any EKS cluster and enroll it in Teleport if its tags match the configured labels.
Teleport cluster auto-discovery involves two components:
- The Teleport Discovery Service that watches for new clusters or
changes to previously discovered clusters.
It dynamically registers each discovered cluster as a
kube_cluster
resource in your Teleport cluster. It does not need connectivity to the clusters it discovers. - The Teleport Kubernetes Service that monitors the dynamic
kube_cluster
resources registered by the Discovery Service. It proxies communications between users and the cluster.
This guide presents the Discovery Service and Kubernetes Service running in the same process, however both can run independently and on different machines.
For example, you can run an instance of the Kubernetes Service in the same private network as the clusters you want to register with your Teleport cluster, and an instance of the Discovery Service in any network you wish.
How it works
The Teleport Discovery Service scans configured cloud providers, including AWS, for Kubernetes clusters that match specified filtering labels, creating dynamic resources within Teleport for any new clusters identified. The Teleport Kubernetes Service monitors these dynamic resources, forwarding requests to the corresponding Kubernetes clusters. Both services require access to the AWS API to perform their functions.
Additionally, the Kubernetes Service needs direct access to the target clusters and the necessary permissions to forward requests.
Prerequisites
-
A running Teleport cluster version 17.0.0-dev or above. If you want to get started with Teleport, sign up for a free trial or set up a demo environment.
-
The
tctl
admin tool andtsh
client tool.Visit Installation for instructions on downloading
tctl
andtsh
.
- An AWS account with permissions to create and attach IAM policies.
- A host to run the Teleport Discovery and Kubernetes services.
- One or more EKS clusters running.
Starting with Teleport v15.3.8, the Discovery Service can self-bootstrap access to EKS clusters by automatically creating and managing Access Entries for each discovered cluster. This contrasts with earlier versions of EKS Auto-Discovery, where agents could not access a cluster without having pre-existing access configured.
Step 1/3. Set up AWS IAM credentials
Create and attach the following AWS IAM policy to the identity of the instance running the Teleport Discovery Service:
{
"Version": "2012-10-17",
"Statement": [
{
"Sid": "EKSDiscovery",
"Effect": "Allow",
"Action": [
"eks:DescribeCluster",
"eks:ListClusters"
],
"Resource": "*"
},
{
"Sid": "EKSManageAccess",
"Effect": "Allow",
"Action": [
"eks:AssociateAccessPolicy",
"eks:CreateAccessEntry",
"eks:DeleteAccessEntry",
"eks:DescribeAccessEntry",
"eks:TagResource",
"eks:UpdateAccessEntry"
],
"Resource": "*"
}
]
}
Statement | Purpose |
---|---|
EKSDiscovery | Discover EKS clusters and fetch additional details about them. |
EKSManageAccess | Automatically set up Teleport access for discovered EKS clusters. |
You can use a list of ARNs and narrow the scope of the permissions to specific regions or EKS clusters instead of using a wildcard. The resource ARN has the following format:
arn:{Partition}:eks:{Region}:{Account}:cluster/{ClusterName}
The permissions in the EKSManageAccess
statement are optional because the
Discovery Service will discover EKS clusters even when it cannot ensure that the
Teleport Kubernetes Service has access to the clusters it discovers.
If you omit any of the EKSManageAccess
permissions, then it is your
responsibility to ensure that the Teleport Kubernetes Service can access each
EKS cluster.
Step 2/3. Configure EKS cluster authorization
If you are running Teleport Discovery v15.3.8 or later and the IAM role used by the Discovery Service has the necessary permissions to create and update Access Entries, you can skip this section. The service can self-bootstrap the required permissions automatically.
When the Kubernetes Service uses an IAM role that is different from the one that
creates the clusters, you need to configure the mapping between the Teleport IAM
Role and the Kubernetes RBAC permissions by editing the aws-auth
Configmap
on
each of the discovered clusters.
To forward requests to the Kubernetes cluster, the Teleport Kubernetes Service
requires cluster-wide permissions to Impersonate
RBAC users and groups, to
create SelfSubjectAccessReviews
and SelfSubjectRulesReviews
, and, finally,
read access to Pods
.
If your Kubernetes cluster does not have an RBAC group with the required
permissions, you can create the ClusterRole
, ClusterRoleBinding
, and the
mapping by following the Creating RBAC group guide.
If your cluster already has an RBAC group that satisfies the required permissions,
you can reuse it and map it into the IAM Role used by the Teleport Kubernetes
Service. For simplicity, it is also possible to map the Teleport IAM role onto
a built-in Kubernetes RBAC group like system:masters
, but not recommended in
production.
Connect to your target cluster with your credentials and create the following
resources using kubectl
.
ClusterRole
Create the ClusterRole
RBAC definition with the required permissions for Teleport
Kubernetes Service to forward requests to the cluster.
apiVersion: rbac.authorization.k8s.io/v1
kind: ClusterRole
metadata:
name: teleport
rules:
- apiGroups:
- ""
resources:
- users
- groups
- serviceaccounts
verbs:
- impersonate
- apiGroups:
- ""
resources:
- pods
verbs:
- get
- apiGroups:
- "authorization.k8s.io"
resources:
- selfsubjectaccessreviews
- selfsubjectrulesreviews
verbs:
- create
ClusterRoleBinding
Link the previously created ClusterRole
into a teleport
RBAC group.
kind: ClusterRoleBinding
apiVersion: rbac.authorization.k8s.io/v1
metadata:
name: teleport
roleRef:
apiGroup: rbac.authorization.k8s.io
kind: ClusterRole
name: teleport
subjects:
- kind: Group
name: teleport
apiGroup: rbac.authorization.k8s.io
IAM mapping
If you cluster includes the aws-auth
config map, edit the configmap/aws-auth
in the kube-system
namespace and append the following to mapRoles
. Replace
{teleport_aws_iam_role}
with the appropriate IAM role that Teleport Kubernetes
Service will use. This step will link the Teleport IAM role into the Kubernetes
RBAC group teleport
, allowing Teleport Kubernetes Service to forward requests
into the cluster:
apiVersion: v1
data:
mapRoles: |
- groups:
- teleport
rolearn: {teleport_aws_iam_role} # e.g. arn:aws:iam::222222222222:role/teleport-role
username: teleport
Otherwise, create an EKS access entry to link the arn:aws:iam::222222222222:role/teleport-role to the Kubernetes Group
teleport
we created in the previous step:
aws eks create-access-entry \ --cluster-name eks-cluster \ --region eu-west-1 \ --principal-arn arn:aws:iam::222222222222:role/teleport-role \ --kubernetes-groups teleport
{ ...}
At this point, the Teleport IAM role already has the minimum permissions to forward requests to the cluster.
To associate the Teleport IAM role with an existing Kubernetes RBAC group,
edit the configmap/aws-auth
in the kube-system
namespace and append
the following to mapRoles
.
apiVersion: v1
data:
mapRoles: |
...
- groups:
- {rbac_group}
rolearn: {teleport_aws_iam_role} # e.g. arn:aws:iam::222222222222:role/teleport-role
username: teleport
Please replace {teleport_aws_iam_role}
with the appropriate IAM role that
Teleport Kubernetes Service is using and {rbac_group}
with the existing Kubernetes
RBAC Group that satisfies the required permissions.
At this point, the Teleport IAM role already has the minimum permissions to forward requests to the cluster.
Granting the system:masters
group to the IAM role associated with the Teleport
service means granting administrator-level permissions on the Kubernetes cluster.
To follow least privilege principle we do not recommend using this method in production.
If your cluster contains an aws-auth
config map, you can use this to associate
the Teleport IAM role with the system:masters
RBAC group. Edit the
configmap/aws-auth
in the kube-system
namespace and append the following to
mapRoles
:
apiVersion: v1
data:
mapRoles: |
...
- groups:
- system:masters
rolearn: {teleport_aws_iam_role} # e.g. arn:aws:iam::222222222222:role/teleport-role
username: teleport
Replace {teleport_aws_iam_role}
with the appropriate IAM role that
Teleport Kubernetes Service is using.
Otherwise, create an EKS access entry and Access Policy to link the arn:aws:iam::222222222222:role/teleport-role to the cluster wide
policy arn:aws:eks::aws:cluster-access-policy/AmazonEKSClusterAdminPolicy
(equivalent of cluster-admin
ClusterRole
):
aws eks create-access-entry \ --cluster-name eks-cluster \ --region eu-west-1 \ --principal-arn arn:aws:iam::222222222222:role/teleport-role
{ ...}aws eks associate-access-policy \ --cluster-name eks-cluster \ --region eu-west-1 \ --principal-arn arn:aws:iam::222222222222:role/teleport-role--policy-arn "arn:aws:eks::aws:cluster-access-policy/AmazonEKSClusterAdminPolicy" \ --access-scope type=cluster
{ ...}
At this point, the Teleport IAM role already has the minimum permissions to forward requests to the cluster.
If you provision your EKS clusters using tools such as terraform
, eksctl
or
Cloudformation
, you can use them to automatically configure the aws-auth
Configmap
or access entry
and create the ClusterRole
and ClusterRoleBinding
resources during cluster provisioning.
Step 3/3. Configure Teleport to discover EKS clusters
Get a join token
Teleport EKS Auto-Discovery requires a valid Teleport auth token for the Discovery and
Kubernetes services to join the cluster. Generate one by running the following
command against your Teleport Auth Service and save it in /tmp/token
on the
machine that will run Kubernetes Discovery:
tctl tokens add --type=discovery,kube
Configure the Teleport Kubernetes and Discovery Services
Discovery Service exposes a configuration parameter - discovery_service.discovery_group
-
that allows you to group discovered resources into different sets. This parameter
is used to prevent Discovery Agents watching different sets of cloud resources
from colliding against each other and deleting resources created by another services.
When running multiple Discovery Services, you must ensure that each service is configured
with the same discovery_group
value if they are watching the same cloud resources
or a different value if they are watching different cloud resources.
It is possible to run a mix of configurations in the same Teleport cluster meaning that some Discovery Services can be configured to watch the same cloud resources while others watch different resources. As an example, a 4-agent high availability configuration analyzing data from two different cloud accounts would run with the following configuration.
- 2 Discovery Services configured with
discovery_group: "prod"
polling data from Production account. - 2 Discovery Services configured with
discovery_group: "staging"
polling data from Staging account.
Enabling EKS Auto-Discovery requires that the discovery_service.aws
section
include at least one entry and that discovery_service.aws.types
include eks
.
It also requires configuring the kubernetes_service.resources.tags
to use the same
labels configured at discovery_service.aws.tags
or a subset of them to make
the Kubernetes Service listen to the dynamic resources created by the Discovery
Service.
version: v3
teleport:
join_params:
token_name: "/tmp/token"
method: token
proxy_server: teleport.example.com:443
auth_service:
enabled: off
proxy_service:
enabled: off
ssh_service:
enabled: off
discovery_service:
enabled: "yes"
discovery_group: "aws-prod"
aws:
- types: ["eks"]
regions: ["*"]
tags:
"env": "prod" # Match EKS cluster tags where tag:env=prod
kubernetes_service:
enabled: "yes"
resources:
- labels:
"env": "prod" # Match Kubernetes Cluster labels specified earlier
Start the Kubernetes and Discovery Services
Grant the Kubernetes and Discovery Services access to credentials that it can use to authenticate to AWS.
- If you are running the Kubernetes and Discovery Services on an EC2 instance, you may use the EC2 Instance Metadata Service method
- If you are running the Kubernetes and Discovery Services in Kubernetes, you can use IAM Roles for Service Accounts (IRSA)
- Otherwise, you must use environment variables
Teleport will detect when it is running on an EC2 instance and use the Instance Metadata Service to fetch credentials.
The EC2 instance should be configured to use an EC2 instance profile. For more information, see: Using Instance Profiles.
Refer to IAM Roles for Service Accounts (IRSA) to set up an OIDC provider in AWS and configure an AWS IAM role that allows the pod's service account to assume the role.
Teleport's built-in AWS client reads credentials from the following environment variables:
AWS_ACCESS_KEY_ID
AWS_SECRET_ACCESS_KEY
AWS_DEFAULT_REGION
When you start the Kubernetes and Discovery Services, the service reads environment variables from a
file at the path /etc/default/teleport
. Obtain these credentials from your
organization. Ensure that /etc/default/teleport
has the following content,
replacing the values of each variable:
AWS_ACCESS_KEY_ID=00000000000000000000
AWS_SECRET_ACCESS_KEY=0000000000000000000000000000000000000000
AWS_DEFAULT_REGION=<YOUR_REGION>
Teleport's AWS client loads credentials from different sources in the following order:
- Environment Variables
- Shared credentials file
- Shared configuration file (Teleport always enables shared configuration)
- EC2 Instance Metadata (credentials only)
While you can provide AWS credentials via a shared credentials file or shared
configuration file, you will need to run the Kubernetes and Discovery Services with the AWS_PROFILE
environment variable assigned to the name of your profile of choice.
If you have a specific use case that the instructions above do not account for, consult the documentation for the AWS SDK for Go for a detailed description of credential loading behavior.
Configure the Kubernetes and Discovery Services to start automatically when the host boots up by creating a systemd service for it. The instructions depend on how you installed the Kubernetes and Discovery Services.
On the host where you will run the Kubernetes and Discovery Services, enable and start Teleport:
sudo systemctl enable teleportsudo systemctl start teleport
On the host where you will run the Kubernetes and Discovery Services, create a systemd service configuration for Teleport, enable the Teleport service, and start Teleport:
sudo teleport install systemd -o /etc/systemd/system/teleport.servicesudo systemctl enable teleportsudo systemctl start teleport
You can check the status of the Kubernetes and Discovery Services with systemctl status teleport
and view its logs with journalctl -fu teleport
.
Once the Kubernetes and Discovery Services start, EKS clusters matching the tags and regions specified in the AWS section will be added to the Teleport cluster automatically.
Troubleshooting
Discovery Service troubleshooting
First, check if any Kubernetes clusters have been discovered.
To do this, you can use the tctl get kube_cluster
command and check if the
expected Kubernetes clusters have already been registered with your Teleport
cluster.
If some Kubernetes clusters do not appear in the list, check if the Discovery Service selector labels match the missing Kubernetes cluster tags or look into the Discovery Service logs for permission errors.
Check that the Discovery Service is running with credentials for the correct AWS account. It can discover resources in another AWS account, but it must be configured to assume a role in the other AWS account if that's the case.
Check if there is more than one Discovery Services running:
tctl inventory status --connected
If you are running multiple Discovery Services, you must ensure that each
service is configured with the same discovery_group
value if they are watching
the same cloud Kubernetes clusters or a different value if they are watching different
cloud Kubernetes clusters.
If this is not configured correctly, a typical symptom is kube_cluster
resources being intermittently deleted from your Teleport cluster's registry.
Kubernetes Service troubleshooting
If the tctl get kube_cluster
command returns the discovered clusters, but the
tctl kube ls
command does not include them, check that you have set the
kubernetes_service.resources
section correctly.
kubernetes_service:
enabled: "yes"
resources:
- labels:
"env": "prod"
If the section is correctly configured, but clusters still do not appear or return authentication errors, please check if permissions have been correctly configured in your target cluster or that you have the correct permissions to list Kubernetes clusters in Teleport.