Logo Alexandre Nuttinck
  • Home
  • About
  • Experiences
  • Skills
  • Projects
  • More
    Education Publications
  • Posts
  • Dark Theme
    Light Theme Dark Theme System Theme
Logo Inverted Logo
  • Posts
  • Adventures in Uptime
    • 1 - The Reproducibility Crisis
    • 2 - Cloud Loops
    • 3 - The Blue Screen of Container Death
    • 4 - The Million-Dollar Mistake
    • 5 - The Day Sysadmins Quit
    • 6 - Untouchable Server
    • 7 - AI Title Generator
    • 8 - The Accidental Production
    • 9 - The Coffee Crisis
    • 10 - Load Test Reality
    • OS1 - Christmas Sweater Outage
    • 11 - Just Use Nix
Hero Image
Just Use Nix

It all started with a classic developer frustration: “Why doesn’t this app build the same way on my machine?!” A colleague leaned over, smiling with the confidence of someone who has already been converted: “You should just use Nix.” How complicated could it be? Well… Nix isn’t just a package manager. It’s a whole paradigm shift: reproducibility, determinism, portability, system stability: everything defined declaratively. It promises environments that never drift and setups that behave exactly the same everywhere. And then there’s NixOS. Not just a tool, but an entire operating system built on the same ideas.

Monday, January 12, 2026 Read
Hero Image
Christmas Sweater Outage

We had a Christmas Sweater Day at the office with colleagues. Let’s just say… that day was particularly inspiring. 😄 This also feels like a good moment to slow things down a bit. There won’t be any new Adventures in Uptime stories next week. But don’t worry, I still enjoy creating these little comic strips just as much, and there will be more to come in 2026!

Monday, December 22, 2025 Read
Hero Image
Load Test Reality

It was a new mission: run performance tests on an application built by another consultant. The goal was simple: either confirm everything was fine, or prove there were real performance issues. The original consultant was very confident. From his point of view, there was no problem at all. “We tested with JMeter. It scales perfectly. No problem.” Let’s just say… our results didn’t exactly match that statement.

Monday, December 15, 2025 Read
Hero Image
The Coffee Crisis

It’s a classic scene in any office… but even more so in IT. Caffeine consumption is high, dependency is real, and when the coffee machine goes down, everything suddenly becomes a critical incident. What always makes me laugh is how seriously teams treat it, almost like a real outage (including myself 😅). Emergency discussions, improvised failover plans, desperate workarounds… All for a working espresso.

Tuesday, December 9, 2025 Read
Hero Image
The Accidental Production

It started as a simple mission: build a Kubernetes POC to explore the benefits and rethink the current architecture. The demo worked, the checklist was green, everyone was happy. We made it clear to the client: “This is only a demo environment, nothing critical.” They nodded, satisfied. We moved on. A few weeks later, the phone rang. It was the same client. Except this time, they didn’t sound calm at all. 😨

Monday, December 1, 2025 Read
Hero Image
AI Title Generator

I was browsing LinkedIn and a few other platforms, looking for sysadmin or system engineer positions. Nothing. Or almost nothing. It turns out that “System Administrator” is apparently too old-school now. The title seems to have quietly disappeared, replaced by an explosion of new ones: DevOps Engineer, Cloud Engineer, Platform Engineer, Site Reliability Engineer, Cloud Architect… and even Operations Ninja. All these roles sound different, yet when you read the job descriptions, they all seem to mean the same thing. Run servers, automate stuff, fix things, you know, the usual.

Monday, November 24, 2025 Read
Hero Image
Untouchable Server

Every IT team has that one server. The one sitting in the corner of the rack with a dusty case, a yellowed label that says “DO NOT TOUCH”, and an aura of mystery. This one ran on RHEL 3. When it was first installed, I was around 10 or 12 years old. It had lived through hardware migrations, OS updates, and entire generations of sysadmins: or rather, survived them.

Monday, November 17, 2025 Read
Hero Image
The Day Sysadmins Quit

This story is inspired by something very real: the invisible work of system administrators. Their job isn’t to ship new features or flashy updates, it’s to make sure that everything just works. Servers stay up, backups run, emails flow, and systems stay secure. Most of it happens quietly, behind the scenes, where no one really notices… until something breaks. Unfortunately, some managers don’t see that. When they don’t understand the technical side, they think “nothing’s happening.” Micromanagement takes over, trust disappears, and appreciation fades. And eventually, the sysadmins decide to leave. Because, let’s be honest, there’s no shortage of work for skilled people elsewhere.

Monday, November 10, 2025 Read
Hero Image
The Million-Dollar Mistake

A major project had just been approved: a collaboration between several partners, with a generous budget to set up new infrastructure. One partner was in charge of purchasing the servers and hosting them in their datacenter. The plan was simple: coordinate with everyone, agree on the specifications, and move forward together. Well… that last part didn’t happen. Without consulting anyone, the partner decided to order high-end IBM proprietary Big Data servers, proudly announcing the purchase once it was too late to change anything. 💸

Monday, November 3, 2025 Read
Hero Image
The Blue Screen of Container Death

It started as a new and exciting mission. The client needed help setting up proper CI/CD pipelines and introducing DevOps best practices. Everything sounded great. Then came the small detail that hadn’t been mentioned during the initial discussions. “Oh, by the way, our application runs on Windows… and we use Windows containers.” I froze for a second. Windows… containers? I didn’t even know those existed before that project.

Wednesday, October 29, 2025 Read
Hero Image
Cloud Loops

It happened during a mission for a client. They had decided to move everything to AWS, hoping to scale faster and more easily than with their old bare-metal self-hosted infrastructure. At first, everything looked perfect. Deployments were smoother, scaling was instant, and everyone was proud of how modern it all felt. The promise of “infinite scalability” seemed within reach. But then, the invoices started growing. Each month, the costs climbed higher, and every time the cluster expanded, the bill followed.

Wednesday, October 22, 2025 Read
Hero Image
The Reproducibility Crisis

This one goes back to one of my very first missions. I arrived at the client’s office, full of energy, ready to dive into their development environment. So I asked the classic first question: “Can I get access to the dev environment?” The answer? A USB key. Inside it, a virtual machine image, several gigabytes of it, along with a Word document as the only piece of documentation for the setup. No Git repository, no configuration management, …

Wednesday, October 15, 2025 Read
  • ««
  • «
  • 1
  • 2
  • »
  • »»
Navigation
  • About
  • Experiences
  • Skills
  • Projects
  • Education
  • Publications
Contact me:
  • alexandre@nuttinck.be
  • alexnuttinck
  • Alexandre Nuttinck
  • alexnuttinck@k8s.social

Toha Theme Logo Toha
Alexandre Nuttinck © 2025 Copyright.
Powered by Hugo Logo