Kubernetes Version 1.23 Eases Management Headaches

The latest version of Kubernetes promises to make it easier to observe and manage individual pods within clusters using a set of temporary containers.

Rey Lejano, a field engineer for SUSE who led the Kubernetes 1.23 release team, says the PodSpec.EphemeralContainer capability will simplify troubleshooting and debugging of Kubernetes pods within a cluster by making it easier to temporarily deploy IT management tools.

Version 1.23 of Kubernetes also includes a command, dubbed kubectl debug, that makes it simpler to launch those temporary containers within a pod. The difference between these and other containers is that temporary containers don’t provide access to ports or other resource requests because they are short-lived.

Other management capabilities include a kubectl event command that makes it easier to monitor the cluster’s overall state and solve problems. This command, available in alpha, makes it simpler to view all events related to a particular resource, watch for specific events in the cluster and filter events by their status or type in a specific namespace.

A horizontal pod autoscaler (HPA), a central component of Kubernetes that automatically scales the number of pods in use based on metrics, is now also generally available. In addition, there is now a proposal to create a custom resource definition (CRD) validation expression language to validate CRDs.

In general, Lejano says right now the technical oversight committee (TOC) for Kubernetes is focused on making it simpler to manage and maintain Kubernetes clusters after they are deployed in a production environment. Those capabilities can be invoked either via the command line interface (CLI) or at a higher level of abstraction to manage Kubernetes provided by an IT vendor, notes Lejano.

It’s unlikely that the TOC itself would ever create its own abstraction layer, but there are multiple initiatives underway within the Cloud Native Computing Foundation (CNCF) to build those abstractions, adds Lejano.

In the longer term, Lejano says IT teams should also expect to see machine learning algorithms being more widely used to simplify the management of fleets of Kubernetes clusters.

In addition, early next year the Kubernetes community will make good on a promise to deprecate Dockershim support in version 1.24 of Kubernetes. The TOC, along with Docker Inc., has been signaling its intention to move away from Docker Engine in favor of an engine for Kubernetes based on Containerd. This move is meant to provide a more efficient runtime exposed via the container runtime interface (CRI). The Dockershim project made the original Docker Engine compatible with the CRI as defined by the Kubernetes TOC. It is now a separate open source project being maintained by Mirantis and Docker, Inc.

Regardless of how organizations approach Kubernetes management, it’s certain that in 2022, more IT teams will be exposed to Kubernetes than ever before. The challenge—both now and in the future—is finding a way to make all those Kubernetes pods and clusters a lot simpler to manage than they currently are.

Mike Vizard

Mike Vizard is a seasoned IT journalist with over 25 years of experience. He also contributed to IT Business Edge, Channel Insider, Baseline and a variety of other IT titles. Previously, Vizard was the editorial director for Ziff-Davis Enterprise as well as Editor-in-Chief for CRN and InfoWorld.

Mike Vizard has 1615 posts and counting. See all posts by Mike Vizard