First part of a new long term home project coming in. An #Ubiquiti PoE+ switch to power a small #Kubernetes cluster built using #raspberrypi nodes. Going to blog about every step once it has been completed. But it is going to be a few quarters long project doing bit by bit
Achievement unlocked: Write a #kubernetes node autoscaler in an afternoon that scales up and down with demand. Completely hacked together in #php to my particular requirements. (Absolutely not built for the long term.) So now I have two #k8s nodes behind me that are off 💙.
With some tweaking to the autoscaling, it's pretty clear it only sometimes scales up when needed. Got a bit batch of World of #Warcraft datamining scheduling for tomorrow, so will see how that impacts node usage. Overall I'm pleased with the makeshift node auto scaling. #kubernetes#k8s
I managed to avoid #Kubernetes for 10 years, but it’s finally caught up to me, so I hope I’m a Kubernetes god after going through all this required (by job) Kubernetes training.
As my old NUC was showing its age, and didn't suffice for my #homelab needs anymore, I decided to build a new one. And because I prefer running all my #selfhosted in containers, but abhor fucking around with #docker and docker-compose, it's a single-node #k8s cluster, using #k3s, just like my old server. One big difference is that the new server has a decent amount of drives for storage. I decided to set up #zfs to manage that, and zfs is all it's cracked up to be.
Kubernetes stuff is all like "Kubestock is a kube control node for funneling proxyd from your kubesits to your kubesats, allowing connections with kubenite and spintwiddle twanks, to keep your yart flow in line with your kubehats."
@wyri unpopular opinion: Sometimes it’s just time to say goodbye.
There are just way better alternatives to #PHP for #K8s platforms. The core concept of configuration and runtime management was build for something different, out of different reasons.
Choosing Drupal as part of your Composable Architecture solution has been a developer topic for a while. If you’re still catching up, our blog post will give you short, informative insights into what this is and why the combination is a knock-out. Learn more: https://www.amazee.io/blog/post/composable-architecture-and-drupal
I forgot that the #Yuzu dedicated multiplayer lobbies, which I've deployed in my #k8s cluster and using for online games would no longer work - cos eventho it's self-hosted, it does send POST requests to yuzu-emu's API which ofc no longer exists. Dk, maybe it still works but requires a direct connection rather than be able to browse/locate through the lobby list?
I'm curious to know how others version control their private configuration files / docs / wiki. For now, I've created a #Gitea#Docker stack on my home network. Instead of copying the files I want to track into another folder and pushing from there, I've created hard links. Any other suggestions? #git#config#versioncontrol#linux@gitea
@5am I use @gitea (self hosted) and keep all my #k8s (k3s, actually) and whatnot in different projects. I use #1password’s op command to inject secrets into the various #yaml files. I use just to automate common tasks. It all works great.
#pv-migrate is an amazing #Kubernetes tool that is absolutely essential to all #homelab (s) and cluster admins that allows you to easily and securely copy the data from one persistent volume claim to another in the same namespace, a different namespace, or heck even a different cluster.
One thing to note about it tho, what I learned just today after years of using it is that it does not support #SELinux found on #RHEL based clusters (which also means I just learned that my former company's clusters prolly weren't using SELinux... oops). I wouldn't recommend removing SELinux entirely tho, simply toggle SELinux from enforcing (1) to permissive (0) when you're about to migrate your data using pv-migrate. Once you're done, just toggle SELinux back on and you're golden.
#Kubernetes/#K8S Q: I've been having an issue all this while I haven't quite been able to tackle. How do I properly mount a #Samba/#SMB/#CIFS share in a #Docker container on Kubernetes?
I definitely don't want a method that does any "pass through" outside of the container such as mounting said share on the Kubernetes node then passing it to the container, since that seems quite hacky and the deployment/pod could easily be reassigned to a different node.