Over the past several years, I worked in marketing and spent a lot of time trying to make technical demos more compelling. The goal was always to demonstrate a capability or product feature, but there’s no reason a demo can’t also be fun, memorable, and grounded in reality. “Hello World” gets the job done, but it doesn’t feel real. It doesn’t resemble how applications are actually built, deployed, or operated. And because of that, it’s easy to forget. ...
Today marks the beginning of the 2024 Formula 1 season, a notable day for my mental well-being as it begins the end of a long, cold winter. The winter months often challenge me, limiting my time outdoors and casting shadows over my daily routines. The start of the Formula 1 season provides a needed distraction and spark of excitement as I look forward to warmer weather. My interest in Formula 1 was ignited by the popular Netflix series, " Drive to Survive." Although new to the sport, I quickly became engrossed, but I initially found myself without a team to support. Over time, the Mercedes Formula 1 team stood out, embodying values I’ve always respected. ...
Maybe it’s bougie, but one of the reasons I purchased E200-8d Supermicro Servers for my home lab was because they had IPMI built into them. Being able to remote into my lab at night when I’m messing around with different configurations was a nice to have. When I was consulting and traveling a lot, being able to remotely start my servers up was really important. I tested out things many times in my lab so that customers wouldn’t have to spend time figuring it out themselves. Anyway, the point is I bought servers with IPMI in them because I thought it was important. ...
Virtual infrastructure has a lot of advantages, but one of the biggest time savers is being able to provision new virtual machines quickly from a template. As part of this series we’ve covered, how to deploy harvester, how to setup a new virtual network , how to create a new virtual machine, and how to backup and restore VMs. In this post we’ll take a look at how we can create templates and deploy harvester VMs from templates. ...
In our previous posts we setup a Harvester cluster in our lab, configured the network, and deployed our first VM. Here we will explore the backup and restore options available to us from the Harvester console. If you’re new to running virtual machines on Kubernetes, you’re probably questioning a few things, like will your current VM backup tool work? The answer is maybe, but only if it is able to backup a container since our VMs are now wrapped inside of a Kubernetes Pod thanks to Kubevirt. There are several options out there (I work for one of those companies myself) but this post only focuses on whats built into the Harvester solution. ...
I have a very soft spot in my heart for VMware. Pretty early on in my career when I was a System Administrator I got the opportunity to convert my companies physical infrastructure to a virtual infrastructure on VMware vSphere. The version of VMware we moved to, is not relevant to this discussion other than to show how truly old I am, and in order to save the parties involved some embarrassment we’ll ignore this implementation detail. ...
During the previous posts in this series, we deployed a Harvester cluster, setup virtual networks to segment traffic, and deployed our first virtual machine. So far this has been a pretty good experience, but most of my day job requires me to do a lot of work on containers so I’d like to have Kubernetes clusters at home. Of course I could deploy a bunch of VMs for a new Kubernetes cluster, but Harvester is built on top of Kubernetes already. So this post will show us how we can connect Rancher to our Harvester cluster so we can use the underlying Kubernetes cluster that Harvester runs on to run our own containers. ...
In the previous posts, we setup the Harvester cluster and optionally added a virtual machine network and some cluster configs. In this post we’ll deploy our first Harvester virtual machine on our cluster. Before we do this, I wanted to point out that we’ve gotten to this point really without ever needing Kubernetes. If you’re paying close attention to the screens that have been showing up in the GUI, you might notice things like a namespace or labels or annotations that are clearly Kubernetes references. Even still, there’s really nothing in the steps up to this point where it would’ve required any Kubernetes experience at all. We’ll continue that trend as we build our first virtual machine. ...
In the first post in this series I deployed a Harvester cluster to my home lab servers. Before I get into deploying virtual machines though, I want to make some common configuration updates to my cluster. For instance, I want my virtual machine network traffic to be placed on a different network. Placing the small amount of network traffic that my virtual machines might be using on their own NIC is probably not necessary in my lab. But I wanted to build the lab in a way that somewhat mimics the way virtual machines are deployed in production environments so I can see how it works. When deploying VMware ESXi hosts, I would commonly have several virtual machine networks on their own NIC, each on their own VLAN. In this post we’ll deploy a Harvester virtual machine network. ...
The installation of Harvester starts with deploying your nodes. This process would be similar to deploying ESXi hosts in a VMware deployment. To start the process, download the latest stable harvester release from their github repository. You can then mount this ISO file from your bare metal server. Create the First Harvester Node in a Cluster After booting to the iso choose the version of Harvester to deploy from their GRUB list. In this example we’ll be using v1.2.1. Since this is the first node in my cluster, I will choose to create a new cluster, and then hit enter. ...