Recently, during migration from one host to another, we had a idea to run the docker in swarm mode, so i future, when we needed more capacity, we could just add more nodes “without the complexity of kubernetes”. However, after starting the working docker-compose in the docker swarm mode, i noticed that the traefik was not picking up the labels from the containers and i couldn’t really figure out why.
Plesk default page instead of domain content after enabling Cloudflare proxy

Recently i faced an interesting issue - after enabling Cloudflare proxy on domain hosted on Plesk (with Litespeed webserver, but that is not important), the domain returned Plesk default page instead of the domain, even through in domain logs i could see the requests.
How NOT to setup a RWX storage for Kubernetes cluster - SMB share on Mikrotik router

Some time ago I decided to setup seperate homelab network in my house, as my previous homelab server is more like a production homeserver which I don’t want to disrupt. It is not much, but I wanted to build something small, energy-efficient and from the parts I already had.
But back to the topic that made me curious - I kinda needed some persistent storage for pods in lab. Then I remembered that the Mikrotik (RB951G-2HnD) router i’m using have a USB port and I saw an option in winbox to create SMB share. I think you can image where it is going.
MicroSD Card not visible on boot in linux
I’m still using my chromebook with coreboot as a portable thinclient. Due to small built-in storage (only 16GB), i’m also using microSD card as a secondary storage for larger files and project. However i was strugeling with microSD card not being mounted at boot (waiting for device), my solution was to just physically eject and insert the card at boot, but it is pretty inconvient and is probably destroying the card pins.
Solving issue with K3s System Upgrade exec /bin/upgrade.sh: exec format error
Recently, I was trying to upgrade my toy K3s cluster using System Upgrade Plans operator. Applied example manifest, control-plane upgraded without issues, but pod for upgrading worker node was constantly restarting with error in logs exec /bin/upgrade.sh: exec format error
.
I would kinda expect that issue on ARM system (although we come a long way in the last couple of years!), but my nodes are running on x86 CPU, so what is happening?