How-To: Set-up Dockup Agent on AWS

Create a kubernetes cluster

First sign up for AWS account if you don't have one. You can do it by heading over to console page.

After you setup your AWS account, head over to kubernetes cluster page, and create a cluster called 'Dockup'. It's okay to go with all default values. Please make sure that you create a role which is attached to EKS

Cluster creation will take sometime. Please wait till it is complete.

Verify connection to cluster

This section is consolidated version of official aws documentation. You can refer this page for more details.

# Install AWS cli using pip:
> pip install awscli
# Install AWS iam authenticator. Using brew:
> brew install aws-iam-authenticator
# Install kubectl using brew:
> brew install kubectl
# Configure aws cli with client id and secret id. They can be generated via
# console
> aws configure
# Retrieve kubernetes configuration.
> aws eks --region us-east-1 update-kubeconfig --name Dockup
# Added new context arn:aws:eks:us-east-1:...:cluster/Dockup to ~/.kube/config
# Set the context for kubernetes
> kubectl config use arn:aws:eks:us-east-1:...:cluster/Dockup
# Ensure that you are able to list pods without any errors.
> kubectl get pod --all-namespaces

Now EKS is active without any nodes. You need to add worker nodes by following instructions as per documentation. Please make sure that security group, VPC and subnets match EKS settings. Note down NodeInstanceRole when stack creation is successful.

# Download, configure and update cloud formation template to connect to nodes
> curl -O
# Replace the <ARN of instance role (not instance profile)> snippet with
# the `NodeInstanceRole` value
> kubectl apply -f aws-auth-cm.yaml
# Check node statuses until any node reaches ready
> kubectl get nodes --watch
# Check pod status, and ensure they are running
> kubectl get pod --all-namespaces
# Install helm. Using homebrew:
> brew install kubernetes-helm
# Install helm on kubernetes cluster:
> helm init
> kubectl create serviceaccount --namespace kube-system tiller
> kubectl create clusterrolebinding tiller-cluster-rule --clusterrole=cluster-admin \
> kubectl patch deploy --namespace kube-system tiller-deploy \
-p '{"spec":{"template":{"spec":{"serviceAccount":"tiller"}}}}'
# You should be able to list helm installations without any error:
> helm list

Install dockup agent

Once you have kubectl and helm installed and configured, you need to install dockup agent. Fetch your Agent API key by going to

> helm repo add dockup
> helm install \
--set agent.dockupApiKey="<agent-api-key-from-settings>" \
--name=dockup-agent \

Once agent is installed, refresh and ensure that agent status is online.

Configure DNS

Next, you will install traefik which will be the loadbalancer for your kubernetes cluster. If you need auto-renewed Letsencrypt SSL, contact mentioning your DNS provider. The following command installs traefik without SSL.

> helm install \
--set rbac.enabled=true \
--name=traefik \
# Get "LoadBalancer Ingress" for traefik from the output of the following command.
> kubectl describe service traefik

Add a CNAME record in your DNS provider to point to the EXTERNAL_IP you got from the previous command.

For example, if you own "" and you want your deployment URLs to be "<random_urls>", then your base domain is and your DNS A record will have key: *.dockup and value: <value of EXTERNAL-IP>

Now you can go back to and update the base domain. In the above example, the value will be "". Once this is done, you are ready to start deploying!

Configure Agent

Configuring base domain

After installing agent, you would have configured DNS. You need to specify the key and domain name without wildcard for agent.

For example, if you own "" and you want your deployment URLs to be "<random_string>", then your base domain is and same should be configured in agent settings.

Configuring registry

Please write to for automatic management of ECR.