Compliance automation is a commonly overlooked area of Kubernetes observability. The question is: how do you automate compliance to a security framework that isn’t well understood by DevSecOps teams to begin with? This lack of understanding contributes to mismanaged compliance efforts and in a worst-case scenario, audit exposures and organizational risk. This talk will walk through an example of how to 1) map compliance controls to specific Kubernetes technical configuration 2) automate the assessment of those controls 3) visualize the assessment results. DevSecOps teams will better understand how to incorporate compliance automation alongside security automation.

"Companies of various sizes are building their applications on Kubernetes because it provides significant operational benefits like autoscaling, self-healing, extensibility, and declarative deployment style. However, the operational benefits are only a starting point down the path of building a secure and observable platform that enables the continuous delivery of application workloads. This session shows how to build a fully operational platform, leveraging platform-oriented building blocks to address network security and observability.

As more production workloads migrated to the cloud, the need for Intrusion Detection Systems(IDS) grew to meet compliance and security needs. With the number of workloads in each cluster, IDS needs to be efficient to not take up the shared resources. Techniques such as packet inspection and web application firewalls provide a solid defense against threats and by leveraging the cluster's network control pane, we are able to selectively choose vulnerable workloads and provide an easy way to trace back to the origin of the attack.

Learn how eBPF will bring a richer picture of what's going on in your cluster, without changing your applications. With eBPF we can safely collect information from deep within your applications, wherever they interact with the kernel. For example, collecting detailed socket statistics to root-cause network issues, or pinpointing the precise binary inside a container that made a particular request for your audit trail. This allows for insights into the behavior (and security) of the system that previously would have needed every process to be (manually) instrumented.