Understanding Kubernetes Visibility

Kubernetes is an open-source system for automating deployment, scaling, and management of containerized applications. A Kubernetes deployment creates a Kubernetes cluster which consists of at least one control plane node and a set of worker nodes that run containers. The worker nodes host the Pods that handle application workloads. The Kubernetes control plane administers the worker nodes and the Pods in a cluster. The core of Kubernetes' control plane is the API server through which users, different parts of the Kubernetes cluster, and external components interact.The Kubernetes API lets you query the state of objects in Kubernetes. For more information, see Kubernetes Documentation.

An increasing number of applications today use Kubernetes for container orchestration. The cloud-native application environments built using Kubernetes are highly distributed, and the majority of traffic consists of East-West flows between containers or pods. From a network perspective, there is very less visibility into these flows, which can make troubleshooting application performance issues difficult.

NetVisor OS version 7.0.0 introduces the Kubernetes visibility feature which gives you full visibility into East-West traffic flows between containers inside a Kubernetes cluster, without external network TAP infrastructure. This feature allows you to correlate containers with applications, to know on which hosts they reside and how they are connected to the fabric.

You can track and monitor containers (for information including IP, name, image, and location in the fabric) and Kubernetes cluster components (kube-controller and kube-scheduler). This granular information helps to perform root cause analysis of application availability and performance issues in a container environment. NetVisor OS implements a Kubernetes client that subscribes to the Kubernetes API server for real time notifications of events and changes.

By configuring the Kubernetes visibility feature, you can track historical data regarding a Kubernetes cluster which helps you to identify and fix availability issues even for containers that existed in the past. The feature lets you trace all available shortest paths between two Pods or nodes. For fine-grained Pod to Pod traffic analytics, you can also enable vPort table creation and connection statistics.