When running a Raspberry Pi cluster, sometimes there’s just not enough power to build native armhf Docker images in a reasonable amount of time.

Running the Azure Container Instances Connector for Kubernetes on a Raspberry Pi Cluster.

Notes from setting up a three node Raspberry Pi 3 Model B Kubernetes cluster using HypriotOS 64-bit.

You have your services containerized, you’ve just deployed them to your fleet of 1000 container hosts and everything is running smoothly. Then you realize that you missed a crucial command in the Dockerfile, and well, re-building the container image and pushing to the repository is easy, but what about the 100’s of containers out there running the old version?

Almost every IT department has one; a Perl script written over a decade ago by a long-gone employee. It does it’s job well, but no one wants to touch or take responsibility for it. You just want to upgrade your infrastructure and bring it along, but there are so many CPAN modules and inconsitencies you want never want to look at it again. What to do?

Over the last few years I’ve championed the ultra-low-power and high-density of a 64-bit ARM platform in the datacenter. The promise of thousands of cores all sipping power and taking up less room than the equivalent Intel architecture to me is both exciting and pragmatic as a large systems IT engineer.

A couple of weeks ago a friend of mine pointed me to the web-based interactive fiction piece called Aisle. As I played it, I immediately started wondering how I could get a text blob of all the possible paths instead of trying to figure out all the actions required to input into the interpreter.

After a long time of www.ecliptik.com being offline I finally brought it back online with Github Pages. I really like the concept of having a static site using Markdown files and a template to quickly bring online a modern looking blog without the hassle of configuring a content management system, web server, or database.