If you want to follow along with the hands-on examples in this tutorial, you'll need a Kubernetes cluster and Helm to get started. You'll find instructions in the preparation section.
Before we get to the hands-on part, let's review different ways an attacker might compromise a Kubernetes cluster:
Kubernetes attack vectors¶
We won't have time to explore all of these in detail but we will be covering some of the most important topics.
- An attacker might take advantage of vulnerabilities in your application code. You can go a long way to address this by scanning container images for vulnerabilities.
- If an attacker gets a foothold within a container, you want to prevent themi from escaping the container to access the host or other components, by configuring container images with security in mind, and enforcing policies for runtime and networking behavior.
- The Kubernetes components and APIs offer more potential surfaces for attack, so you want to check your Kubernetes settings that might leave your deployment vulnerable.
- At the root of many security issues lies human error, as well as bad actors. You can limit human access to the cluster, making auditing easier, and enhancing security by employing GitOps.
With that, you had a look at the big picture, so let's move on to setting up our hands-on environment.