Do you know about using Kubernetes Debug containers? They're really useful for troubleshooting well-built, locked-down images that are running in your cluster. I was thinking it would be nice if k9s had this feature, and lo and behold, it has a plugin! I just had to add that snippet to my ${HOME}/.config/k9s/plugins.yaml, run k9s, find the pod, press enter to get into the pod's containers, select a container, and press Shift-D. The debug-container plugin uses the nicolaka/netshoot image, which has a bunch of useful tools on it. Easy debugging in k9s!

I got a Kubernetes survey from The Linux Foundation today and it had a question that asked how Kubernetes impacted my professional life. Good question. Here's my answer:

Kubernetes renewed my love for infrastructure and DevOps! As cloud platforms grew more popular, my job function changed; instead of managing services, I was managing "managed" services. I was losing the ability to actually build infrastructure with free, open source software (FOSS), and I was forced to become an expert in proprietary solutions. I switched to FOSS years ago because the idea of writing software for anyone in the world to use and modify - for free - was really inspiring! When I discovered Kubernetes, I was able to work with FOSS infrastructure again! Once I have compute, storage, and networking from the cloud platform, I can install Kubernetes, and actually manage my own services again!

I've been working with Kubernetes since 2015 and I've mangled with handcrafted manifests including almost duplicate manifests for staging/production environments, played around with stuff like Cue, built lots glue (mostly shell script) to automate manifest-handling and -generation and I also enjoy parts of Kustomize. When Helm started to appear it seemed like a terrible hack, especially since it came with the Tiller-dependency to handle Helm-managed state-changes inside of the clusters. And while they dropped Tiller (thankfully), I still haven't made my peace with Helm.

Go-templating it awful to read, a lot of Helm charts don't really work out of the box, charts can be fed values that aren't shown via helm show values ./chart, debugging HelmChart $namespace/$release-$chartname is not ready involves going over multiple logs spread over different parts of the cluster and I could go on and on. And yet, almost every project that goes beyond offering Dockerfile+docker-compose.yaml just releases a Helm Chart for their app.

Am I the only one who is annoyed by Helm? Have I been using it wrongly? Is there something I've been missing?

In case you're a Helm maintainer: Please don't take it personally, my issue is not with the people behind Helm!

Hello world!

I want to release to internet my custom immutable rolling-release extreme-simple Linux distribution for Kubernetes deployments.

I was using this distribution for about the last 6 years on production environments (currently used by a few startups and two country's public services). I really think that it could be stable enough to be public published to internet before 2024.

I'm asking for advice before the public release, as licensing, community building, etc.

A few specs about the distribution:

• Rolling release. Just one file (currently less than ~40Mb) that can be bootable from BIOS or UEFI (+secure boot) environments. You can replace this file by the next release or use the included toolkit to upgrade the distribution (reboot/kexec it). Mostly automated distribution releases by each third-party releases (Linux, Systemd, Containerd, KubeAdm, etc).

• HTTP setup. The initial setup could be configured with a YAML file written anywhere in a FAT32 partition or through a local website installer. You can install the distribution or configure KubeAdm (control-plane & worker) from the terminal or the local website.

• Simple, KISS. Everything must be simple for the user, this must be the most important aspect for the distribution. Just upstream software to run a production ready Kubernetes cluster.

• No money-driven. This distribution must be public, and it must allow to be forked at any time by anyone.

A bit of background:

I was using CoreOS before Redhat bought them. I like the immutable distro and A/B release aspect of the CoreOS distribution. After the Redhat acquisition, the CoreOS distribution was over-bloated. I switched to use my own distribution, built with Buildroot. A few years later, I setup the most basic framework to create a Kubernetes cluster without any headache. It's mostly automated (bots checking for new third-party releases as Linux, Systemd, Containerd, KubeAdm, etc; building, testing & signing each release). I already know that building a distribution is too expensive, because of that I programmed few bots that made this job for me. Now days, I only improve the toolkits, and approve the Git requests from thats bots.

Thank you for your time!

Looking for the best way to learn kubernetes, given that I have plenty of years of engineering (Java, python) and a solid experience with AWS.

any format works - paid/free courses, working through articles, getting started guides, etc..