Another year, another Hacktoberfest. This year, I created a basic ray tracer and also contributed to a Prometheus exporter.

This blog post documents the steps I took to create a Kubernetes cluster according to Kelsey Hightower's "Kubernets The Hard Way", but with LXC.

HashiCorp Vault is a tool for centralized secrets management. In my opinion, the best way to learn a technology is to use it. For this reason, we are going to build a demo system in this blog post.

This article will describe how to monitor batch jobs with Prometheus. The defining characteristic of batch jobs is their limited lifetime. Monitoring ephemeral jobs with Prometheus is suboptimal due to Prometheus pull-based architecture.

Sometimes I need to create certificates for testing purposes. For example, I set up an etcd cluster by hand recently. This happens rarely enough that I need to look up the openssl syntax every time.

Sometimes, issues can only be fixed in a layer below the abstractions you rely on usually. Understanding the whole stack as much as possible is a great advantage when a bug pops up. For this reason, I decided to implement a simple Linux syscall to become familiar with the Linux kernel’s code organization. That said, I don’t aspire to become a kernel hacker. The new syscall printk simply prints a given null-terminated message into the kernel log.

When I started this blog at the end of 2018, I didn’t want to spend much time on the blog’s setup, but rather get started writing as soon as possible. For this reason, I chose a preconfigured VM image with the blogging software Ghost that was offered by my cloud provider. In the meantime, writing has become a hobby to me, even though I keep most of what I write private.