Version v0.5 of the documentation is no longer actively maintained. The site that you are currently viewing is an archived snapshot. For up-to-date documentation, see the latest version.

Securing Your Clusters

How to secure Kubeflow clusters using VPC service controls and private GKE

This guide describes how to secure Kubeflow using VPC service Controls and private GKE.

Together these two features signficantly increase security and mitigate the risk of data exfiltration

  • VPC Service Controls allow you to define a perimeter around GCP services

    • Kubeflow uses VPC Service Controls to prevent applications running on GKE from writing data to GCP resources outside the perimeter.
  • Private GKE removes public IP addresses from GKE nodes making them inaccessible from the public internet

    • Kubeflow uses IAP to make Kubeflow web apps accessible from your browser.

VPC Service Controls allow you to restrict which Google services are accessible from your GKE/Kubeflow clusters. This is an important part of security and in particular mitigating the risks of data exfiltration.

For more information refer to the VPC Service Control Docs.

Creating a private Kubernetes Engine cluster means the Kubernetes Engine nodes won’t have public IP addresses. This can improve security by blocking unwanted outbound/inbound access to nodes. Removing IP addresses means external services (including GitHub, PyPi, DockerHub etc…) won’t be accessible from the nodes. Google services (BigQuery, Cloud Storage, etc…) are still accessible.

Importantly this means you can continue to use your Google Container Registry (GCR) to host your Docker images. Other Docker registries (for example DockerHub) will not be accessible. If you need to use Docker images hosted outside Google Container Registry you can use the scripts provided by Kubeflow to mirror them to your GCR registry.

Before you start

Before installing Kubeflow ensure you have installed the following tools:

You will need to know your gcloud organization id and project number; you can get it via gcloud

export PROJECT=<your GCP project id>
export ORGANIZATION_NAME=<name of your organization>
export ORGANIZATION=$(gcloud organizations list --filter=DISPLAY_NAME=${ORGANIZATION_NAME} --format='value(name)')
export PROJECT_NUMBER=$(gcloud projects describe kubeflow-dev --format='value(projectNumber)')
  • Projects are identified by names, ids and numbers for more info see here

Enable VPC Service Controls In Your Project

  1. Enable VPC service controls

    export PROJECT=<Your project>
    gcloud services enable \
  2. Check if you have an access policy object already created

    gcloud beta access-context-manager policies list \
    • An access policy is a GCP resource object that defines service perimeters. There can be only one access policy object in an organization, and it is a child of the Organization resource.
  3. If you don’t have an access policy object create one

    gcloud beta access-context-manager policies create \
    --title "default" --organization=${ORGANIZATION}
  4. Save the Access Policy Object id as an environment variable so that in can be used in subsequent commands

    export POLICYID=$(gcloud beta access-context-manager policies list --organization=${ORGANIZATION} --limit=1 --format='value(name)')
  5. Create a service perimeter

    gcloud beta access-context-manager perimeters create KubeflowZone \
        --title="Kubeflow Zone" --resources=projects/${PROJECT_NUMBER} \,, \
        --project=${PROJECT} --policy=${POLICYID}
    • Here we have created a service perimeter with the name KubeflowZone.

    • The perimeter is created in PROJECT_NUMBER and restricts access to GCS( and BigQuery( and GCR(

    • Placing GCS and BigQuery in the perimeter means that access to GCS and BigQuery resources owned by this project is now restricted; by default access from outside the perimeter will be blocked

    • More than one project can be added to the same perimeter

  6. Create an access level to allow Google Container Builder to access resources inside the permiter

    • Create a members.yaml file with the following contents
       - members:      
         - serviceAccount:${PROJECT_NUMBER}
         - user:<your email>
    • Google Container Builder is used to mirror Kubeflow images into the perimeter
    • Adding your email allows you to access the GCP services inside the perimeter from outside the cluster

      • This is convenient for building and pushing images and data from your local machine.
    • For more information refer to the docs.

  7. Create the access level

    gcloud beta access-context-manager levels create kubeflow \
       --basic-level-spec=members.yaml \
       --policy=${POLICYID} \
       --title="Kubeflow ${PROJECT}"
    • The name for the level can’t have any hyphens
  8. Bind Access Level to a Service Perimeter

    gcloud beta access-context-manager perimeters update KubeflowZone \
     --add-access-levels=kubeflow \
  9. Setup container registry for GKE private clusters (for more info see instructions)

    1. Create a managed private zone

      export ZONE_NAME=kubeflow
      export NETWORK=<Network you are using for your cluster>
      gcloud beta dns managed-zones create ${ZONE_NAME} \
       --visibility=private \
       --networks=${PROJECT}/global/networks/${NETWORK} \
       --description="Kubeflow DNS" \ \
    2. Start a transaction

      gcloud dns record-sets transaction start \
       --zone=${ZONE_NAME} \
    3. Add a CNAME record for *

      gcloud dns record-sets transaction add \
       --name=* \
       --type=CNAME \
       --zone=${ZONE_NAME} \
       --ttl=300 \
    4. Add an A record for the restricted VIP

       gcloud dns record-sets transaction add \ \
         --type=A \
         --zone=${ZONE_NAME} \
         --ttl=300 \
    5. Commit the transaction

       gcloud dns record-sets transaction execute \
        --zone=${ZONE_NAME} \

Deploy Kubeflow with Private GKE

  1. Set user credentials. You only need to run this command once:

    gcloud auth application-default login
  2. Copy non-GCR hosted images to your GCR registry

    1. Clone the Kubeflow source

      git clone git_kubeflow      
    2. Use Google Cloud Builder(GCB) to replicate the images

      cd git_kubeflow/scripts/gke
      PROJECT=<PROJECT> make copy-gcb
    • This is needed because your GKE nodes won’t be able to pull images from non GCR registries because they don’t have public internet addresses

    • gcloud may return an error even though the job is submited successfully and will run successfully see kubeflow/kubeflow#3105

    • You can use the Cloud console to monitor your GCB job.

  3. Follow the instructions for creating an OAuth client

  4. Create environment variables for IAP OAuth access

     export CLIENT_ID=<CLIENT_ID from OAuth page>
     export CLIENT_SECRET=<CLIENT_SECRET from OAuth page>
  5. Download a kfctl release from the Kubeflow releases page.

  6. Unpack the tar ball:

    tar -xvf kfctl_<release tag>_<platform>.tar.gz
    • Optional Add the kfctl binary to your path.
    • If you don’t add the kfctl binary to your path then in all subsequent steps you will need to replace kfctl with the full path to the binary.
  7. Initialize the directory containing your Kubeflow deployment config files

    export PROJECT=<your GCP project>
    export KFAPP=<your choice of application directory name>
    # Default uses IAP.
    kfctl init ${KFAPP} --platform gcp --project ${PROJECT}
    cd ${KFAPP}
    kfctl generate all -V
    • ${KFAPP} - the name of a directory where you want Kubeflow configurations to be stored. This directory is created when you run kfctl init. If you want a custom deployment name, specify that name here. The value of this variable becomes the name of your deployment. The value of this variable cannot be greater than 25 characters. It must contain just the directory name, not the full path to the directory. The content of this directory is described in the next section.
    • ${PROJECT} - the name of the GCP project where you want Kubeflow deployed.
    • When you run kfctl init you need to choose to use either IAP or basic authentication, as described below.
    • kfctl generate all attempts to fetch your email address from your credential. If it can’t find a valid email address, you need to pass a valid email address with flag --email <your email address>. This email address becomes an administrator in the configuration of your Kubeflow deployment.
  8. Enable private clusters by editing ${KFAPP}/gcp_configs/cluster-kubeflow.yaml and updating the following two parameters:

    privatecluster: true
    gkeApiVersion: v1beta1
  9. Remove components which are not useful in private clusters:

    cd ${KFAPP}/ks_app
    ks component rm cert-manager
  10. Create the deployment:

    cd ${KFAPP}
    kfctl apply platform
    • If you get an error legacy networks not supported, follow the troubleshooting guide to create a new network.

      cd ${KFAPP}/gcp_configs
      gcloud --project=${PROJECT} deployment-manager deployments create ${KFAPP}-network --config=network.yaml
      • Then edit gcp_config/cluster.jinja to add a field network in your cluster
         name: {{ CLUSTER_NAME }}
         network: <name of the new network>
      • To get the name of the new network run
      gcloud --project=${PROJECT} compute networks list
      • The name will contain the value ${KFAPP}
  11. Update iap-ingress component parameters:

    cd ${KFAPP}/ks_app
    ks param set iap-ingress privateGKECluster true
  12. Obtain an HTTPS certificate for your ${FQDN} and create a Kubernetes secret with it.

    • You can create a self signed cert using kube-rsa
      go get
      kube-rsa ${FQDN}
    • The fully qualified domain is the host field specified for your ingress; you can get it by running

      cd ${KFAPP}/ks_app
      ks param list | grep hostname
    • Then create your Kubernetes secret

      kubectl create secret tls --namespace=kubeflow envoy-ingress-tls --cert=ca.pem --key=ca-key.pem
  13. Update the various ksonnet components to use images instead of Docker Hub images:

    cd ${KFAPP}/ks_app
  14. Apply all the Kubernetes resources:

    cd ${KFAPP}
    kfctl apply k8s
  15. Wait for Kubeflow to become accessible and then access it at

    • ${FQDN} is the host associated with your ingress

      • You can get it by running kubectl get ingress
    • Follow the instructions to monitor the deployment

    • It can take 10-20 minutes for the endpoint to become fully available

Next steps