Field Notes From KubeCon #4: Ensuring Reliability of Production-Ready Kubernetes Operators Using EnvTest
Field Notes from KubeCon India 2024 on Reliability of Production-Ready Kubernetes Operators
Field Notes from KubeCon India 2024 on Reliability of Production-Ready Kubernetes Operators
Hi Ishan here again! Delivering you the field notes I took from all the sessions I attended during KubeCon Day 1 and 2 — so you don’t miss a thing!
After grabbing a quick coffee, I headed to the session led by Aniruddha Basak and Rayan Das. The title alone promised a deep dive into Kubernetes operators—a crucial area for anyone working with cloud-native tech.
Aniruddha kicked things off by explaining what a Kubernetes operator is. "Think of it as automating what a human operator would do to deploy and maintain complex applications," he said. Kubernetes operators manage applications using custom resources and controllers.
Rayan added, "At its core, a controller watches the cluster’s desired state, compares it to the current state, and reconciles the differences.” This constant feedback loop ensures that Kubernetes systems stay healthy.
Their example? A Kubernetes operator managing external secrets by automatically syncing sensitive data—a real-world use case that immediately clicked with the audience.
They didn’t sugarcoat the struggles of testing Kubernetes operators. Traditional testing methods—unit, integration, and end-to-end (E2E)—each have pros and cons. Rayan shared candidly, "Most people hate writing E2E tests because they are slow, flaky, and a nightmare to maintain.”
Aniruddha nodded. "They often use fake Kubernetes clients that don't behave like real clusters, leading to blind spots in production." The limitations were clear.
To overcome these hurdles, they introduced EnvTest, a Go library designed for testing Kubernetes controllers. EnvTest spins up lightweight Kubernetes API components like etcd and the Kubernetes API server, allowing realistic tests without spinning up a full cluster.
"It’s like getting the benefits of integration testing without the usual overhead," Aniruddha explained. Rayan shared a live demo, showing how tests could run faster while still interacting with real Kubernetes components.
They wrapped up the session with actionable best practices:
This session was a goldmine for anyone working with Kubernetes operators. It made me rethink how we approach testing at Facets.cloud. Using EnvTest could help us scale faster and reduce bugs in production.
As I left the room, I couldn’t help but appreciate how deep dives like this push the entire Kubernetes ecosystem forward. Looking forward to sharing more from KubeCon in the next installment!
.png)
