I'm wicked excited, it's finally here! Legacy Labs is officially starting its first ever event! For the next two months I'm going to be sysadmining like it's 2009, well sort of, I'll break down my full plan here in a second and it's a little bit anachronistic but there's a reasoning behind it. But I'm getting ahead of myself, you probably already have a couple of questions, like what is legacy labs exactly? And "I thought summer camps were for kids, what?".
What is a Legacy Lab?
For the last 5 years I've participated every summer in a group event called the OCC or Old Computer Challenge. The premise on the surface was always very simple, for a week we'd all get together and constrain ourselves to a low end low resourced computer and see how much we could do with it. I had an absolute blast doing this, and it turned into this thing I looked forward to every single summer where I got an excuse to use some old long forgotten hardware or try an esoteric operating system in earnest. I wasn't super heavily involved in the community necessarily, but despite that I was very invested in the idea surrounding it.
For me, the OCC signaled a week in which I got to explore something for the sake of exploring it and nothing further. The point was to learn by trying something strange.
The last year I participated the "challenge" was simply that there was no challenge, and that honestly resonates a bit. I daily drove a Motorola Droid 4 with Alpine Linux as my primary computer for 7-8 years simply because I could. The idea that the challenges weren't challenging was something that resonated with me deeply. So I tried to go over the top with my participation. But I faced time constraints and set backs because of my off the wall ideas, and the uphill struggle it can be working with old unreliable hardware and long abandoned systems. Because of that struggle my last OCC also devolved in to essentially "setup and fix Windows vista" which wasn't really what I wanted to do with my time.
So Legacy Labs is my attempt to provide a space more tailored towards what I want to do. Instead of picking a single week where we get together and try to force ourselves to use dial up speeds, 1/2gb of ram, or a specific operating system; we provide 2 months to pick any retro-computing/permacomputing topic/s that interest you and encourage you to go deep with them. Really dig in and figure out how things work, why they were designed the way they were, the history behind the systems that exist. Intentionally creating a space for my own curiosity and desire to exist.
How you personally go about doing that exploration isn't the important part. For example, this year I'll be doing a deep dive on Windows Server 2008 core, but I'm not going to be running everything under Hyper-V. I'll setup a Hyper-V server sure, but it's not so much about shoving myself into this anachronistic world and pretending like I'm back in 2009 as a solo sysadmin. I lived that through, purposefully chose to hone Linux skills professionally, I just want to learn more about the core variant of Windows frankly. So my hypervisor in my lab will be Incus, something inherently modern and a tool I use professionally. I think, personally, this invites greater flexibility and freedom of creativity for people to explore whatever they want to explore.
Maybe as LLSC evolves we'll do more thematic exploration as a group, but for our very first summer camp the only guidance I have is be curious! Whatever your criteria of "retro" or "perma" or "computing" might be is good enough to satisfy me and I look forward to any and all contributions!
Which brings us round nicely to what I'm doing for the next couple of months.
Vista, Server 2008, and other crazy ideas
A couple of my close friends are already aware of this, but I have a bit of a thing for Windows Vista. It's an objectively terribly operating system. I remember HATING it vehemently. But I kind of owe Vista my career in a sense. You see, one of the first times I really applied my Linux foo and solved real world problems was replacing Vista with Ubuntu to save a friends laptop. I got to learn so much about debugging wireless driver issues on both Vista, and then Ubuntu, and later how to configure WINE to get things like CIV and the Sims running under Ubuntu once Vista had been purged with prejudice. I'm really kind of fond of it in retrospect.
And thanks to last year's OCC I have this super sweet well configured Vista netbook that oddly enough sips power and breezes along on a single gig of ram. It's so wildly lightweight that it makes the under powered x86 atom CPU seem powerful. And I'm unfortunately aware of how under powered it is, having used an OCC to debug SBCL build errors. Despite that fact it runs Emacs and I can compile most of my Nim programs on it, and between those two things I have the development environment I personally need and a plethora of tooling to enable whatever I happen to want. And enough spare Mikrotik gear in my home to cordon it FAR away from anything else in the house during these events.
This setup combined with the acquisition of a bunch of old 100mb Cisco gear and a stint I spent remediating a plethora of server 2008 infrastructure lead me to this idea right before last year's occ. Essentially that I have enough random equipment to build an entire small business infrastructure out of random parts and pieces I'm not currently using in my homelab. I called that Project Half Duplex though I don't know that I'll keep that name. I think I'm going to give my "small business" a name and add some role playing color to what I do for LLSC.
And what precisely is Project Half Duplex but not by that name? Well, a lot actually
I've got a spare Incus node that I'm working with, it'll be the core of the hosting infrastructure. On it I plan to run the following VMs:
- Deployment of a Server 2008 R2 core Active Directory server
- Deployment of a Server 2008 R2 core File server
- Deployment of a Server 2008 R2 core DHCP server (which I want to then serve DHCP on a physical network using the aforementioned hardware)
- Deployment of a Server 2008 R2 Hyper-V server
- Deployment of a Server 2008 R2 Syteline + Progress ERP system
- At least one Vista domain joined system, which will be a Syteline client.
Server 2008 R2 doesn't need a lot of resources to run, and nor does Inucs on Alpine. So while praa only has a 7th gen i3 CPU and 16GB of ram, it should be plenty for a small cluster of nodes. Dealing with nested virtualization on the Hyper-V node will be interesting. Using Alpine for the base of the lab host allows me to work in an environment where all of the setup work is familiar and easy. I can completely nuke my Incus node if I need to and redeploy it using salt. If I have to travel during the LLSC I can throw Nebula on praa and have instant single node access to my entire lab environment without compromising on security of the underlying hypervisor. There's a lot of work that I've done professionally on these sorts of systems and being able to showcase unique implementations is honestly really fun.
Just from getting that lab environment setup I'm hoping to be able to write about the following topics:
- Creating golden images in Incus for legacy operating systems
- Configuring random Mikrotik devices as wlan to lan bridges
- Showcasing some of the work I've done on saltext-nebula and saltext-alpine
And from the point where I have a lab setup in any capacity I can pivot wherever I find interest. I could dig into doing node monitoring for my VMs using the Prometheus endpoints Incus provides and how to tie that into Zabbix. Or I could decide to ditch Incus in lieu of RDP from my Vista netbook once the initial nodes are setup. I have so many wonderful ideas of what I could do with such an odd ball lab!
I'm obviously an infrastructure guy so getting to build out a bunch of little servers by hand and document the process is my idea of fun. If you're thinking of participating and are thinking "this sounds cool but I'm not sure about all that", feel free to take it from any other angle that does sound interesting to you! Note that my common thread across all of these weird ideas is that I want to learn more about them. I think a future LLSC will likely involve just focusing on software development, like finishing this Vista IRC client.
I've also got a Canon Powershot G5 at the ready with a new compact flash card and battery, so expect to see some delightfully dated 5MP photos as LLSC progresses. With two months to dedicate to this I'm hoping I can get some really good photos and actually have time to figure out an editing workflow and dive into the mechanics of this particular camera a bit.
Anyways, enough from me, I have a lab to build!