The Hitchhiker’s Guide to Kubernetes Add-Ons: How to Simplify Operations and Scale with Confidence

Please note that the text may have slight differences or mistranscription from the audio recording.

Introduction

Udi Hofesh: Hi everyone, and welcome to our webinar. We’re going to wait a couple of minutes to let everyone join in. I see some familiar names—welcome back! And if this is your first Komodor webinar, we’re happy to have you here.

Amit Bar Oz: Yeah, great to see so many people joining! We have a really interesting session planned today. We’ll be diving into Kubernetes add-ons, breaking down some of the most important ones, and discussing how to manage them effectively. We’ll also give you a live demo on how Komodor simplifies add-on operations.

Itiel Shwartz: Sorry for being a little late! I got caught up, but thanks for the warm welcome. Really excited to be here and talk about this topic—it’s something I deal with every day, and I know a lot of folks struggle with managing add-ons properly.

Udi Hofesh: No worries at all! Before we dive in, just a quick housekeeping note—if you have any questions, drop them in the chat. We’ll have a Q&A at the end. Also, everyone who registered will receive a recording and the deck after the session. Let’s get started!

What Are Kubernetes Add-ons?

Udi Hofesh: A lot of people think Kubernetes is a complete solution on its own, but add-ons are what really make it powerful. They extend Kubernetes to provide automation, security, networking, and other critical functions.

Itiel Shwartz: Exactly. Kubernetes is designed to be extensible. With CRDs, controllers, and Helm, teams can integrate external capabilities easily. That’s why there’s such a massive ecosystem of add-ons available today.

Amit Bar Oz: Yeah, I like to think of Kubernetes as a foundation, and add-ons as the specialized tools that enhance it. The challenge, though, is that as you scale, managing these add-ons becomes more complex. It’s easy to install them, but maintaining them properly is another story.

Udi Hofesh: That’s a great point. Let’s break down some of the key add-ons and discuss their use cases, challenges, and best practices.

Cert-Manager

Itiel Shwartz: Cert-Manager is a widely used add-on that automates SSL certificate provisioning and renewal. It simplifies security management by handling certificates for your services automatically.

Amit Bar Oz: Yeah, it’s super useful, but when it breaks, it can be a nightmare. A lot of teams don’t even realize there’s a problem until their certificates expire, and suddenly, everything stops working.

Udi Hofesh: That’s why it’s critical to have proactive monitoring in place. Ensuring high availability for the controller and setting up alerts for certificate expiration can save you a lot of headaches.

External-DNS

Amit Bar Oz: External-DNS dynamically manages DNS records based on Kubernetes resources. It makes mapping services to domain names much easier, especially for teams managing multiple environments.

Itiel Shwartz: The biggest issue I see with External-DNS is API rate limits from DNS providers. Another common mistake is accidentally deleting production DNS records, which can take services offline.

Udi Hofesh: Definitely. A best practice is to separate DNS zones, set reasonable TTL values, and implement ownership policies to prevent accidental deletions.

Helm

Itiel Shwartz: Helm is basically the package manager for Kubernetes. It standardizes deployment, making it easier to install, upgrade, and manage applications.

Amit Bar Oz: It’s a great tool, but it’s not always as simple as it seems. Customizing Helm charts for production requires careful planning, and if you’re not careful, updates can break things.

Udi Hofesh: That’s where Helm Dashboard comes in handy. It gives you visibility into your Helm releases and helps troubleshoot issues. Komodor also provides deep insights into Helm deployments, making updates safer.

Argo Workflows

Amit Bar Oz: Argo Workflows is a Kubernetes-native workflow engine. It’s commonly used for CI/CD pipelines and data processing tasks.

Itiel Shwartz: Troubleshooting failed workflows is one of the biggest challenges, especially for data engineers who may not have deep Kubernetes expertise.

Udi Hofesh: To manage this effectively, it’s best to use workflow templates, implement retry strategies, and actively monitor workflow execution in real time.

Karpenter & Cluster Autoscaler

Amit Bar Oz: These tools help scale Kubernetes nodes dynamically based on workload demand. They’re essential for optimizing resource usage and cost efficiency.

Itiel Shwartz: But if they’re not configured correctly, you can end up either over-provisioning and spending too much or under-provisioning and running into capacity issues.

Udi Hofesh: The key is to optimize resource requests and use multiple node groups to balance cost and performance effectively.

Kyverno

Amit Bar Oz: Kyverno is a policy engine for Kubernetes that helps enforce security and compliance rules across clusters.

Itiel Shwartz: It’s easy to deploy, but poorly designed policies can cause performance bottlenecks or create conflicts with existing configurations.

Udi Hofesh: A good approach is to start with simple policies, store them in version control, and regularly monitor policy violations to prevent unexpected issues.

Add-on Challenges & Best Practices

Amit Bar Oz: Managing add-ons at scale comes with its own set of challenges—operational overhead, dependency conflicts, and even unexpected downtime if something breaks.

Itiel Shwartz: Visibility is one of the biggest pain points. Many teams struggle to understand how add-ons interact with their applications and infrastructure.

Udi Hofesh: That’s where Komodor comes in. We help teams visualize, detect, investigate, and remediate add-on issues, ensuring a more reliable and streamlined Kubernetes experience.

Komodor’s Solution

Itiel Shwartz: At Komodor, we treat add-ons as first-class citizens. Our platform provides:

• Detection & Monitoring: Real-time insights into add-on health
• Automated Troubleshooting: AI-powered root cause analysis
• Optimization Recommendations: Helping you run add-ons efficiently

Udi Hofesh: Let’s take a look at how Komodor makes managing add-ons easier with a quick demo.

(Demo showcasing Komodor’s capabilities)

Q&A & Closing Remarks

Udi Hofesh: Thanks, everyone, for joining us today! The recording and slides will be sent to all registrants. If you have any more questions, feel free to reach out. See you next time!

Itiel Shwartz & Amit Bar Oz: Thanks, everyone!