Spot Ocean: The easy button for Kubernetes version and patch management

As a recovering practitioner, one of the things I was always less than thrilled to do was patching. I dreaded having to run updates, ensure everything was configured correctly, and finding windows of time to reboot without knocking an application offline. Just like your cloud workloads — everything changes when it has to be done at scale.

Kubernetes has always been built around the DevOps concepts of rapid development, delivery, and iteration. As Kubernetes has solidified its position as “the operating system of the cloud,” Automation becomes even more crucial to day-to-day CloudOps practitioners; the Kubernetes project maintainers know this – which is why they follow DevOps principles around their releases throughout the calendar year.

With the next major release of Kubernetes slated for April 17, Kubernetes version 1.30 will have a myriad of changes and enhancements, and with them come the consideration of how to approach upgrade management. This planning becomes even more crucial when considering there is usually a four-to-eight-week lag time between an official Kubernetes upstream release and when the hyperscalers support the new versions and features. Therefore, it can be easy to put aside or even forget about critical upgrades to your core infrastructure.

And if there is one thing I hear repeatedly as I meet with customers and other practitioners, it is the ongoing struggle to balance operational needs during that four-to-eight-week window, as opposed to running updates and preparing for a new Kubernetes release.

What if that process could become more automated and less burdensome while improving the resilience and availability of your workloads?

This is what Ocean from Spot by NetApp was purpose-built to help with.

Let’s explore the imperatives of having procedures in place to effectively upgrade and patch Kubernetes and why you should consider automation to offload those tasks.

Why keeping Kubernetes up to date is critical

• Security: If your Kubernetes cluster is your fortress, and each new version is an upgrade to the strength of your wall, skipping or missing if you skip or miss an important upgrade you could be opening the front gate to an intruder who can potentially exfiltrate your data. Just because a vulnerability is identified and fixed doesn’t mean your infrastructure is protected.
• Performance: As Kubernetes evolves and grows, so do the impacts on your infrastructure. If you decide not to upgrade, you’re essentially making the choice to run a marathon with a ball and chain around your ankle. Sure, you’ll finish (eventually) but it be much harder — and for what? Because you were able to kick out another project faster?
• Futureproofing: As sysadmins/cloud admins, we tend to forget how important futureproofing infrastructure really is. Kubernetes and the cloud native landscape are always changing. Not upgrading Kubernetes versions to the newest releases is like refusing to buy a new wardrobe since the 1980s. Sure, you think you look cool, but everyone else thinks you’re stuck in the past.
• Budget and financial management: When you use EKS from AWS, you pay $0.10 per cluster hour to run EKS. If you run a version in “extended support,” the price will jump to $0.60 per hour. Assuming 730 hours in a billing month, your cost at scale to use EKS goes from $73.00 per month to $438.00 per month per cluster. Depending on how many clusters you have, that unnecessary cost could jump exorbitantly before you know it.

Spot Ocean: The solution to Kubernetes update automation

1. Update frequency: Ocean can check daily, weekly, or on a custom time scale that aligns with your business.
2. Set it and forget it: With Ocean, it’s quick to set up auto-upgrading on your cluster. Automate the upgrades so you or your team can focus on higher priority, higher value tasks.
3. Infrastructure as Code support: Spot has worked to ensure that we can meet customers no matter where they are on their cloud journey, and this includes supporting customers with whatever Infrastructure as Code tools they use.
4. Slow your (cluster) roll: With Virtual Node Groups (VNGs), you can roll your cluster using isolated node groups instead of your entire cluster, which increases uptime and availability of your Kubernetes workloads.
5. AI- and ML-powered automation: Spot’s recommendation engine is powered by years of AI and ML data to help determine the infrastructure your workload needs. Sleep easy at night knowing your workload will scale when it is needed, and only for as long as you need it.

How do I start?

At the beginning of the article, I know many of you groaned when you saw it was about considerations for upgrading and maintaining versions. But what if automating the patching and upgrading could be painless?

Because of the shared responsibility model that is cloud computing, upgrading your base Kubernetes version also means upgrading the version of Kubernetes your nodes will be running. To assist CloudOps teams with this, we released support for automatic version and patching of your AMIs for AWS EKS and ECS.

This release is true to Spot Ocean’s primary mission: Making your K8s management at scale easier and more cost-effective. We automate the repetitive complex tasks that otherwise ruin your day, week, or month — like upgrading your Kubernetes infrastructure.

Are you not currently using Spot Ocean to manage your Kubernetes upgrades? Learn how our CloudOps tools can be a game changer for your cloud and business operations. Meet with one of our infrastructure solutions architects or sign up for a risk-free 14-day trial from Spot to see how we can take your workload management to the next level.