kube-goat A deliberately vulnerable Kubernetes cluster

This document details the process of getting started with Kube-Goat in a local environment using Kind or a public cloud provider (GCP and AWS).

Getting Started

• Kind
  • Prerequisites
  • Install Kind
  • Create KubeGoat Cluster
  • Deploy Resources
  • Delete Cluster
• KOPS on GCP
  • Install Prerequisites
    • KOPS
    • GCloud SDK
    • kubectl
  • Launch Cluster
  • Delete Cluster

Kind

kind (kubernetes-in-docker) can be used to bring up local clusters anywhere you have Docker installed (e.g. Windows, MacOS, Linux) with the added benefit of not having to deal with virtualization.

Prerequisites

Docker:

On Linux: curl -fsSL https://get.docker.com
On MacOS: brew install docker

Kubectl

On Linux: curl -LO https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/linux/amd64/kubectl
On MacOS: brew install kubectl

[Optional] Go

Install Kind

You can download & install pre-compiled versions of kind from their GitHub Releases.

Alternatively, you may build kind by running go get -u -v sigs.k8s.io/kind

Create KubeGoat Cluster

1. kind create cluster --name=kubegoat --config=config.yaml
2. export KUBECONFIG="$(kind get kubeconfig-path --name="kubegoat")"

Deploy Resources

In the manifests directory run the following command:

kubectl create -f app -f dashboard -f namespace -f secrets

Delete Cluster

kind delete cluster --name="kubegoat"

KOPS on GCP

These commands can be run in Google Cloud Shell after a GCP account is created and activated. They rely on the gcloud and kubectl command-line utilities to be installed which Cloud Shell already gives us.

Install Prerequisites

KOPS

Kops is used to bootstrap a cluster. It relies on a central configuration file. First, install the kops binary:

wget -O kops https://github.com/kubernetes/kops/releases/download/$(curl -s https://api.github.com/repos/kubernetes/kops/releases/latest | grep tag_name | cut -d '"' -f 4)/kops-linux-amd64 && \
chmod +x ./kops && \
sudo mv ./kops /usr/local/bin/

Ensure Kops is installed successfully:

kops version

GCloud SDK

Visit https://cloud.google.com/sdk/ for instructions on installing the gcloud CLI. Once you are done installing GCloud SDK, you must run, gcloud init, this will configure your gcloud with your existing GCP project.

kubectl

From the official Kubernetes kubectl release:

wget -O kubectl https://storage.googleapis.com/kubernetes-release/release/$(curl -s https://storage.googleapis.com/kubernetes-release/release/stable.txt)/bin/darwin/amd64/kubectl
chmod +x ./kubectl
sudo mv ./kubectl /usr/local/bin/kubectl

Launch Cluster

In the examples/kops directory, cd into your cloud provider of choice and run the following command (this example is for GCP)”

Ensure you are authenticated to the GCP account for Kube-Goat

gcloud auth login

Run the following script to spin up Kube-Goat in GCP. Name the Project and Bucket as you wish.

./kops_gcp_setup.sh <GCP-Project-ID> <GCP-Bucket-ID>

This will go through the necessary steps to create a kube-goat cluster in your GCP account.

This cluster runs using two Compute VMs and a single bucket for data storage. It is not free! You can always sign up with GCP to get $300 in credit for testing purposes

The cluster should now be up and running. Go to Compute Engine -> VM Instances to view the cluster nodes. If you are wondering how you got your kubectl configured to the this cluster, KOPS does that for you. It exports a kubecfg file for a cluster from the state store to your ~/.kube/config

kubectl get pods --all-namespaces

Delete Cluster

Use kops to delete the running cluster:

kops delete cluster kube-goat.k8s.local --yes

Optionally, you can delete the entire GCP project:

gcloud projects delete <project_id> -q

---