Skip to main content

Running an HA Teleport cluster using AWS, EKS, and Helm

In this guide, we'll use Teleport Helm charts to set up a high-availability Teleport cluster that runs on AWS EKS.

Have an existing Teleport cluster?

If you are already running Teleport on another platform, you can use your existing Teleport deployment to access your Kubernetes cluster. Follow our guide to connect your Kubernetes cluster to Teleport.

tip

Teleport Enterprise Cloud takes care of this setup for you so you can provide secure access to your infrastructure right away.

Get started with a free trial of Teleport Enterprise Cloud.

Prerequisites​

Teleport's charts require the use of Helm version 3. You can install Helm 3 by following these instructions.

Throughout this guide, we will assume that you have the helm and kubectl binaries available in your PATH:

$ helm version
# version.BuildInfo{Version:"v3.4.2"}

$ kubectl version
# Client Version: version.Info{Major:"1", Minor:"17+"}
# Server Version: version.Info{Major:"1", Minor:"17+"}
Best practices for production security

When running Teleport in production, you should adhere to the following best practices to avoid security incidents:

  • Avoid using sudo in production environments unless it's necessary.
  • Create new, non-root, users and use test instances for experimenting with Teleport.
  • Run Teleport's services as a non-root user unless required. Only the SSH Service requires root access. Note that you will need root permissions (or the CAP_NET_BIND_SERVICE capability) to make Teleport listen on a port numbered < 1024 (e.g. 443).
  • Follow the principle of least privilege. Don't give users permissive roles when more a restrictive role will do. For example, don't assign users the built-in access,editor roles, which give them permissions to access and edit all cluster resources. Instead, define roles with the minimum required permissions for each user and configure access requests to provide temporary elevated permissions.
  • When you enroll Teleport resources—for example, new databases or applications—you should save the invitation token to a file. If you enter the token directly on the command line, a malicious user could view it by running the history command on a compromised system.

You should note that these practices aren't necessarily reflected in the examples used in documentation. Examples in the documentation are primarily intended for demonstration and for development environments.

Choose a Kubernetes namespace and Helm release name​

note

Before starting, setting your Kubernetes namespace and Helm release name here will enable easier copy/pasting of commands for installation.

If you don't know what to put here, use teleport for both values.

Namespace: namespace

Release name: release-name

Step 1/7. Install Helm​

Teleport's charts require the use of Helm version 3. You can install Helm 3 by following these instructions.

Throughout this guide, we will assume that you have the helm and kubectl binaries available in your PATH:

$ helm version
# version.BuildInfo{Version:"v3.4.2"}

$ kubectl version
# Client Version: version.Info{Major:"1", Minor:"17+"}
# Server Version: version.Info{Major:"1", Minor:"17+"}

Step 2/7. Add the Teleport Helm chart repository​

Set up the Teleport Helm repository.

Allow Helm to install charts that are hosted in the Teleport Helm repository:

$ helm repo add teleport https://charts.releases.teleport.dev

Update the cache of charts from the remote repository so you can upgrade to all available releases:

$ helm repo update

Step 3/7. Set up AWS IAM configuration​

For Teleport to be able to manage the DynamoDB tables, indexes, and the S3 storage bucket it needs, you'll need to configure AWS IAM policies to allow access.

note

Add these IAM policies to your AWS account and then grant it to the role associated with your EKS node group(s).

DynamoDB IAM policy​

On startup, the Teleport Auth Service checks whether the DynamoDB table you have specified in its configuration file exists. If the table does not exist, the Auth Service attempts to create one.

The IAM permissions that the Auth Service requires to manage DynamoDB tables depends on whether you expect to create a table yourself or enable the Auth Service to create and configure one for you:

If you choose to manage DynamoDB tables yourself, you must take the following steps, which we will explain in more detail below:

  • Create a cluster state table.
  • Create an audit event table.
  • Create an IAM policy and attach it to the Teleport Auth Service's IAM identity.

Create a cluster state table​

The cluster state table must have the following attribute definitions:

NameType
HashKeyS
FullPathS

The table must also have the following key schema elements:

NameType
HashKeyHASH
FullPathRANGE

Create an audit event table​

The audit event table must have the following attribute definitions:

NameType
SessionIDS
EventIndexN
CreatedAtDateS
CreatedAtN

The table must also have the following key schema elements:

NameType
CreatedAtDateHASH
CreatedAtRANGE

Create and attach an IAM policy​

Create the following IAM policy and attach it to the Teleport Auth Service's IAM identity.

You'll need to replace these values in the policy example below:

Placeholder valueReplace with
us-west-2AWS region
1234567890AWS account ID
teleport-helm-backendDynamoDB table name to use for the Teleport backend
teleport-helm-eventsDynamoDB table name to use for the Teleport audit log (must be different to the backend table)
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "ClusterStateStorage",
            "Effect": "Allow",
            "Action": [
                "dynamodb:BatchWriteItem",
                "dynamodb:UpdateTimeToLive",
                "dynamodb:PutItem",
                "dynamodb:DeleteItem",
                "dynamodb:Scan",
                "dynamodb:Query",
                "dynamodb:DescribeStream",
                "dynamodb:UpdateItem",
                "dynamodb:DescribeTimeToLive",
                "dynamodb:DescribeTable",
                "dynamodb:GetShardIterator",
                "dynamodb:GetItem",
                "dynamodb:ConditionCheckItem",
                "dynamodb:UpdateTable",
                "dynamodb:GetRecords",
                "dynamodb:UpdateContinuousBackups"
            ],
            "Resource": [
                "arn:aws:dynamodb:us-west-2:1234567890:table/teleport-helm-backend",
                "arn:aws:dynamodb:us-west-2:1234567890:table/teleport-helm-backend/stream/*"
            ]
        },
        {
            "Sid": "ClusterEventsStorage",
            "Effect": "Allow",
            "Action": [
                "dynamodb:BatchWriteItem",
                "dynamodb:UpdateTimeToLive",
                "dynamodb:PutItem",
                "dynamodb:DescribeTable",
                "dynamodb:DeleteItem",
                "dynamodb:GetItem",
                "dynamodb:Scan",
                "dynamodb:Query",
                "dynamodb:UpdateItem",
                "dynamodb:DescribeTimeToLive",
                "dynamodb:UpdateTable",
                "dynamodb:UpdateContinuousBackups"
            ],
            "Resource": [
                "arn:aws:dynamodb:us-west-2:1234567890:table/teleport-helm-events",
                "arn:aws:dynamodb:us-west-2:1234567890:table/teleport-helm-events/index/*"
            ]
        }
    ]
}

Note that you can omit the dynamodb:UpdateContinuousBackups permission if disabling continuous backups.

S3 IAM policy​

On startup, the Teleport Auth Service checks whether the S3 bucket you have configured for session recording storage exists. If it does not, the Auth Service attempts to create and configure the bucket.

The IAM permissions that the Auth Service requires to manage its session recording bucket depends on whether you expect to create the bucket yourself or enable the Auth Service to create and configure it for you:

Note that Teleport will only use S3 buckets with versioning enabled. This ensures that a session log cannot be permanently altered or deleted, as Teleport will always look at the oldest version of a recording.

You'll need to replace these values in the policy example below:

Placeholder valueReplace with
your-sessions-bucketName to use for the Teleport S3 session recording bucket
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Sid": "BucketActions",
            "Effect": "Allow",
            "Action": [
                "s3:ListBucketVersions",
                "s3:ListBucketMultipartUploads",
                "s3:ListBucket",
                "s3:GetEncryptionConfiguration",
                "s3:GetBucketVersioning"
            ],
            "Resource": "arn:aws:s3:::your-sessions-bucket"
        },
        {
            "Sid": "ObjectActions",
            "Effect": "Allow",
            "Action": [
                "s3:GetObjectVersion",
                "s3:GetObjectRetention",
                "s3:GetObject",
                "s3:PutObject",
                "s3:ListMultipartUploadParts",
                "s3:AbortMultipartUpload"
            ],
            "Resource": "arn:aws:s3:::your-sessions-bucket/*"
        }
    ]
}

Step 4/7. Configure TLS certificates for Teleport​

We now need to configure TLS certificates for Teleport to secure its communications and allow external clients to connect.

Depending on the approach you use for provisioning TLS certificates, the teleport-cluster chart can deploy either a Kubernetes LoadBalancer or Kubernetes Ingress to handle incoming connections to the Teleport Proxy Service.

Determining an approach​

There are three supported options when using AWS. You must choose only one of these options:

ApproachAWS Load Balancer TypeKubernetes Traffic DestinationCan use an existing AWS LB?Caveats
Using cert-managerNetwork Load Balancer (NLB)LoadBalancerNoRequires a Route 53 domain and an Issuer configured with IAM permissions to change DNS records for your domain
Using AWS Certificate ManagerApplication Load Balancer (ALB)IngressYesRequires a working instance of the AWS Load Balancer controller installed in your Kubernetes cluster
Using your own TLS credentialsNetwork Load Balancer (NLB)LoadBalancerNoRequires you to independently manage the maintenance, renewal and trust of the TLS certificates securing Teleport's web listener

Using cert-manager​

You can use cert-manager to provision and automatically renew TLS credentials by completing ACME challenges via Let's Encrypt.

You can also use cert-manager with AWS Private Certificate Authority (PCA) in EKS using the aws-privateca-issuer plugin.

This method uses a Kubernetes LoadBalancer, which will provision an underlying AWS Network Load Balancer (NLB) to handle incoming traffic.

Using AWS Certificate Manager​

You can use AWS Certificate Manager to handle TLS termination with AWS-managed certificates.

This method uses a Kubernetes Ingress, which can provision an underlying AWS Application Load Balancer (ALB) to handle incoming traffic if one does not already exist. It also requires the installation and setup of the AWS Load Balancer controller.

You should be aware of these potential limitations and differences when using Layer 7 load balancers with Teleport:

  • Connecting to Kubernetes clusters at the command line requires the use of the tsh proxy kube or tsh kubectl commands and tsh proxy db/tsh db connect commands respectively. It is not possible to connect kubectl directly to Teleport listeners without the use of tsh as a proxy client in this mode.
  • Connecting to databases at the command line requires the use of the tsh proxy db or tsh db connect commands. It is not possible to connect database clients directly to Teleport listeners without the use of tsh as a proxy client in this mode.
  • The reason for both of these requirements is that Teleport uses X509 certificates for authentication, which requires that it terminate all inbound TLS traffic itself on the Teleport proxy. This is not directly possible when using a Layer 7 load balancer, so the tsh client implements this flow itself using ALPN connection upgrades.
  • The use of Teleport and tsh v13 or higher is required.
warning

Using ACM with an ALB also requires that your cluster has a fully functional installation of the AWS Load Balancer controller with required IAM permissions. This guide provides more details below.

Using your own TLS credentials​

With this approach, you are responsible for determining how to obtain a TLS certificate and private key for your Teleport cluster, and for renewing your credentials periodically. Use this approach if you would like to use a trusted internal certificate authority instead of Let's Encrypt or AWS Certificate Manager. This method uses a Kubernetes LoadBalancer and will provision an underlying AWS NLB.

Steps to follow​

Once you have chosen an approach based on the details above, select the correct tab below for instructions.

In this example, we are using multiple pods to create a High Availability Teleport cluster. As such, we will be using cert-manager to centrally provision TLS certificates using Let's Encrypt. These certificates will be mounted into each Teleport pod, and automatically renewed and kept up to date by cert-manager.

If you are planning to use cert-manager, you will need to add one IAM policy to your cluster to enable it to update Route53 records.

Route53 IAM policy​

This policy allows cert-manager to use DNS01 Let's Encrypt challenges to provision TLS certificates for your Teleport cluster.

You'll need to replace these values in the policy example below:

Placeholder valueReplace with
Z0159221358P96JYAUAA4Route 53 hosted zone ID for the domain hosting your Teleport cluster
{
    "Version": "2012-10-17",
    "Statement": [
        {
            "Effect": "Allow",
            "Action": "route53:GetChange",
            "Resource": "arn:aws:route53:::change/*"
        },
        {
            "Effect": "Allow",
            "Action": [
                "route53:ChangeResourceRecordSets",
                "route53:ListResourceRecordSets"
            ],
            "Resource": "arn:aws:route53:::hostedzone/Z0159221358P96JYAUAA4"
        }
    ]
}

Installing cert-manager​

If you do not have cert-manager already configured in the Kubernetes cluster where you are installing Teleport, you should add the Jetstack Helm chart repository which hosts the cert-manager chart, and install the chart:

$ helm repo add jetstack https://charts.jetstack.io
$ helm repo update
$ helm install cert-manager jetstack/cert-manager \
--create-namespace \
--namespace cert-manager \
--set installCRDs=true \
--set global.leaderElection.namespace=cert-manager \
--set extraArgs="{--issuer-ambient-credentials}" # required to automount ambient AWS credentials when using an Issuer

Once cert-manager is installed, you should create and add an Issuer.

You'll need to replace these values in the Issuer example below:

Placeholder valueReplace with
email@address.comAn email address to receive communications from Let's Encrypt
example.comThe name of the Route 53 domain hosting your Teleport cluster
us-west-2AWS region where the cluster is running
Z0159221358P96JYAUAA4Route 53 hosted zone ID for the domain hosting your Teleport cluster
cat << EOF > aws-issuer.yaml
apiVersion: cert-manager.io/v1
kind: Issuer
metadata:
  name: letsencrypt-production
  namespace: teleport
spec:
  acme:
    email: email@address.com
    server: https://acme-v02.api.letsencrypt.org/directory
    privateKeySecretRef:
      name: letsencrypt-production
    solvers:
    - selector:
        dnsZones:
          - "example.com"
      dns01:
        route53:
          region: us-west-2
          hostedZoneID: Z0159221358P96JYAUAA4
EOF

After you have created the Issuer and updated the values, add it to your cluster using kubectl:

$ kubectl create namespace namespace
$ kubectl label namespace teleport 'pod-security.kubernetes.io/enforce=baseline'
$ kubectl --namespace namespace create -f aws-issuer.yaml

Step 5/7. Set values to configure the cluster​

Before you can install Teleport in your Kubernetes cluster, you will need to create a secret that contains your Teleport license information.

The Teleport Auth Service reads a license file to authenticate your Teleport Enterprise account.

To obtain your license file, navigate to your Teleport account dashboard and log in. You can start at teleport.sh and enter your Teleport account name (e.g. my-company). After logging in you will see a "GENERATE LICENSE KEY" button, which will generate a new license file and allow you to download it.

License File Download

Create a secret from your license file. Teleport will automatically discover this secret as long as your file is named license.pem.

$ kubectl -n namespace create secret generic license --from-file=license.pem

Next, configure the teleport-cluster Helm chart to use the aws mode. Create a file called aws-values.yaml and write the values you've chosen above to it:

chartMode: aws
clusterName: teleport.example.com                 # Name of your cluster. Use the FQDN you intend to configure in DNS below.
proxyListenerMode: multiplex
aws:
  region: us-west-2                # AWS region
  backendTable: teleport-helm-backend # DynamoDB table to use for the Teleport backend
  auditLogTable: teleport-helm-events             # DynamoDB table to use for the Teleport audit log (must be different to the backend table)
  auditLogMirrorOnStdout: false                   # Whether to mirror audit log entries to stdout in JSON format (useful for external log collectors)
  sessionRecordingBucket: your-sessions-bucket  # S3 bucket to use for Teleport session recordings
  backups: true                                   # Whether or not to turn on DynamoDB backups
  dynamoAutoScaling: false                        # Whether Teleport should configure DynamoDB's autoscaling.
highAvailability:
  replicaCount: 2                                 # Number of replicas to configure
  certManager:
    enabled: true                                 # Enable cert-manager support to get TLS certificates
    issuerName: letsencrypt-production            # Name of the cert-manager Issuer to use (as configured above)
# If you are running Kubernetes 1.23 or above, disable PodSecurityPolicies
podSecurityPolicy:
  enabled: false
note

If using an AWS PCA with cert-manager, you will need to ensure you set highAvailability.certManager.addCommonName: true in your values file. You will also need to get the certificate authority certificate for the CA (aws acm-pca get-certificate-authority-certificate --certificate-authority-arn <arn>), upload the full certificate chain to a secret, and reference the secret with tls.existingCASecretName in the values file.

Install the chart with the values from your aws-values.yaml file using this command:

$ helm install release-name teleport/teleport-cluster \
  --create-namespace \
  --namespace namespace \
  -f aws-values.yaml
note

You cannot change the clusterName after the cluster is configured, so make sure you choose wisely. You should use the fully-qualified domain name that you'll use for external access to your Teleport cluster.

Once the chart is installed, you can use kubectl commands to view the deployment (example using cert-manager):

$ kubectl --namespace namespace get all

NAME                                 READY   STATUS    RESTARTS   AGE
pod/teleport-auth-57989d4cbd-4q2ds   1/1     Running   0          22h
pod/teleport-auth-57989d4cbd-rtrzn   1/1     Running   0          22h
pod/teleport-proxy-c6bf55cfc-w96d2   1/1     Running   0          22h
pod/teleport-proxy-c6bf55cfc-z256w   1/1     Running   0          22h

NAME                        TYPE           CLUSTER-IP     EXTERNAL-IP                         PORT(S)                                                                     AGE
service/teleport            LoadBalancer   10.40.11.180   xxxxx.elb.us-east-1.amazonaws.com   443:30258/TCP                                                               22h
service/teleport-auth       ClusterIP      10.40.8.251    <none>                              3025/TCP,3026/TCP                                                           22h
service/teleport-auth-v11   ClusterIP      None           <none>                              <none>                                                                      22h
service/teleport-auth-v12   ClusterIP      None           <none>                              <none>                                                                      22h

NAME                             READY   UP-TO-DATE   AVAILABLE   AGE
deployment.apps/teleport-auth    2/2     2            2           22h
deployment.apps/teleport-proxy   2/2     2            2           22h

NAME                                       DESIRED   CURRENT   READY   AGE
replicaset.apps/teleport-auth-57989d4cbd   2         2         2       22h
replicaset.apps/teleport-proxy-c6bf55cfc   2         2         2       22h

Step 6/7. Set up DNS​

You'll need to set up a DNS A record for teleport.example.com. In our example, this record is an alias to an ELB.

Using Application Access?

Teleport assigns a subdomain to each application you configure for Application Access. For example, if you enroll Grafana as a resource, Teleport assigns the resource to the grafana.teleport.example.com subdomain.

If you host the Teleport cluster on your own network, you should update your DNS configuration to account for application subdomains. You can update DNS in one of two ways:

  • Create a single DNS address (A) or canonical name (CNAME) record using wildcard substitution for the subdomain name. For example, create a DNS record with the name *.teleport.example.com.
  • Create a separate DNS address (A) or canonical name (CNAME) record for each application subdomain.

Modifying DNS ensures that the certificate authority—for example, Let's Encrypt—can issue a certificate for each subdomain and that clients can verify Teleport hosts regardless of the application they are accessing.

If you use the Teleport cloud platform, no DNS updates are needed because your Teleport cluster automatically provides the subdomains and signed TLS certificates for your applications under your tenant address.

Here's how to do this in a hosted zone with AWS Route 53:

# Change these parameters if you altered them above
$ NAMESPACE='namespace'
$ RELEASE_NAME='release-name'

# DNS settings (change as necessary)
$ MYZONE_DNS='example.com'
$ MYDNS='teleport.example.com'
$ MY_CLUSTER_REGION='us-west-2'

# Find AWS Zone ID and ELB Zone ID
$ MYZONE="$(aws route53 list-hosted-zones-by-name --dns-name="${MYZONE_DNS?}" | jq -r '.HostedZones[0].Id' | sed s_/hostedzone/__)"
$ MYELB="$(kubectl --namespace "${NAMESPACE?}" get "service/${RELEASE_NAME?}-proxy" -o jsonpath='{.status.loadBalancer.ingress[*].hostname}')"
$ MYELB_NAME="${MYELB%%-*}"
$ MYELB_ZONE="$(aws elbv2 describe-load-balancers --region "${MY_CLUSTER_REGION?}" --names "${MYELB_NAME?}" | jq -r '.LoadBalancers[0].CanonicalHostedZoneId')"

# Create a JSON file changeset for AWS.
$ jq -n --arg dns "${MYDNS?}" --arg elb "${MYELB?}" --arg elbz "${MYELB_ZONE?}" \
    '{
        "Comment": "Create records",
        "Changes": [
          {
            "Action": "CREATE",
            "ResourceRecordSet": {
              "Name": $dns,
              "Type": "A",
              "AliasTarget": {
                "HostedZoneId": $elbz,
                "DNSName": ("dualstack." + $elb),
                "EvaluateTargetHealth": false
              }
            }
          },
          {
            "Action": "CREATE",
            "ResourceRecordSet": {
              "Name": ("*." + $dns),
              "Type": "A",
              "AliasTarget": {
                "HostedZoneId": $elbz,
                "DNSName": ("dualstack." + $elb),
                "EvaluateTargetHealth": false
              }
            }
          }
      ]
    }' > myrecords.json

# Review records before applying.
$ cat myrecords.json | jq
# Apply the records and capture change id
$ CHANGEID="$(aws route53 change-resource-record-sets --hosted-zone-id "${MYZONE?}" --change-batch file://myrecords.json | jq -r '.ChangeInfo.Id')"

# Verify that change has been applied
$ aws route53 get-change --id "${CHANGEID?}" | jq '.ChangeInfo.Status'
# "INSYNC"

Step 7/7. Create a Teleport user​

Create a user to be able to log into Teleport. This needs to be done on the Teleport auth server, so we can run the command using kubectl:

$ kubectl --namespace namespace exec deploy/release-name-auth -- tctl users add test --roles=access,editor

User "test" has been created but requires a password. Share this URL with the user to complete user setup, link is valid for 1h:
https://teleport.example.com:443/web/invite/91cfbd08bc89122275006e48b516cc68

NOTE: Make sure teleport.example.com:443 points at a Teleport proxy that users can access.

Load the user creation link to create a password and set up multi-factor authentication for the Teleport user via the web UI.

High Availability​

In this guide, we have configured two replicas. This can be changed after cluster creation by altering the highAvailability.replicaCount value using helm upgrade as detailed below.

Upgrading the cluster after deployment​

To make changes to your Teleport cluster after deployment, you can use helm upgrade.

Helm defaults to using the latest version of the chart available in the repo, which will also correspond to the latest version of Teleport. You can make sure that the repo is up to date by running helm repo update.

Here's an example where we set the chart to use 2 replicas:

Edit your aws-values.yaml file from above and make the appropriate changes:

highAvailability:
  replicaCount: 2

Upgrade the deployment with the values from your aws-values.yaml file using this command:

$ helm upgrade release-name teleport/teleport-cluster \
  --namespace namespace \
  -f aws-values.yaml
note

To change chartMode, clusterName, or any aws settings, you must first uninstall the existing chart and then install a new version with the appropriate values.

Then perform a cluster upgrade with the new values:

$ helm upgrade release-name teleport/teleport-cluster \
  --namespace namespace \
  -f aws-values.yaml

Uninstalling Teleport​

To uninstall the teleport-cluster chart, use helm uninstall <release-name>. For example:

$ helm --namespace namespace uninstall release-name

Uninstalling cert-manager​

If you want to remove the cert-manager installation later, you can use this command:

$ helm --namespace cert-manager uninstall cert-manager

Troubleshooting​

AWS quotas​

If your deployment of Teleport services brings you over your default service quotas, you can request a quota increase from the AWS Support Center. See Amazon's AWS service quotas documentation for more information.

For example, when using DynamoDB as the backend for Teleport cluster state, you may need to request increases for read/write quotas.

Next steps​

Now that you have deployed a Teleport cluster, read the Manage Access section to get started enrolling users and setting up RBAC.

See the high availability section of our Helm chart reference for more details on high availability.

Read the cert-manager documentation.