I have about that many. Looks good to me! I have two Windows VMs. One for work and presentations. One for games and Adobe. A bunch of random Linux VMs trying to get a FireWire card to work and a Windows 7 VM for the same reason. I’ve also for several Linux VMs trying out new versions of Fedora, Ubuntu, or Debian. A couple servers. Almost none of them are ever turned on because my real virtualized workloads run in docker or LXC! I never could get Mac VM to work but I have an AMD CPU and a MacBook so not too high priority.
Linux
From Wikipedia, the free encyclopedia
Linux is a family of open source Unix-like operating systems based on the Linux kernel, an operating system kernel first released on September 17, 1991 by Linus Torvalds. Linux is typically packaged in a Linux distribution (or distro for short).
Distributions include the Linux kernel and supporting system software and libraries, many of which are provided by the GNU Project. Many Linux distributions use the word "Linux" in their name, but the Free Software Foundation uses the name GNU/Linux to emphasize the importance of GNU software, causing some controversy.
Rules
- Posts must be relevant to operating systems running the Linux kernel. GNU/Linux or otherwise.
- No misinformation
- No NSFW content
- No hate speech, bigotry, etc
Related Communities
Community icon by Alpár-Etele Méder, licensed under CC BY 3.0
Have you automated creation?
With that many Windows (gasp) ones, no... I'm afraid you are not
It's only insane if you have them all running at once.
GPU passthrough has always been one of those exciting ideas I’d love to dive into one day. My current GPU being a little older, has only 4GB of RAM. Oh the joy's of being a budget PC user. Thankfully it's more of a "would be nice rather" than an "actually need"....
Very few people need it but it’s awesome and a lot of fun and lets you spend more time in Linux than dealing with Windows. The VFIO Reddit and Arch wiki are great resources. I have GPU, USB, and Ethernet pass through on my Ubuntu machine and it works great, but I needed the Arch wiki to really figure out what I was doing wrong when I first set it up. Level1Techs is also a good resource on YouTube and forums because they are big into VFIO and SR-IOV. Next time you get a PC, make sure to look for more PCI lanes and bifurcation support on your motherboard. Gen 4 is a great option because it generally has enough lanes and the ram and ssd are much cheaper than Gen 5. GPU choice doesnt matter much but if you’ve got AMD watch out for the reset bug. Basically you can start a VM but once you quit it the cards state is unavailable for further use (eg a second VM session or reopening your DE if you’re using a single GPU setup) unless you restart your host. There are some workarounds but personally I’d avoid it if possible. Onboard graphics (iris or amd APU) are recommended. Older hardware can get cheap so good luck saving up if this is something you want to do!
Well I do but I have a machine with 3/4 of a terabyte of memory on it.
Work scraps are great sometimes.
How are you running the MacOS VMs. The machine I have is a cheese grater so that makes it easier.
Are you running macOS or Linux as your host? My MacBook is M1 and I found the performance running ARM windows and ARM Fedora via UTM (qemu) to be pretty good.
On the cheesegrater(2019 MacPro) it’s a little convoluted. During covid times it was my single box lab since it had so much memory (768TB). So I was running nested ESXI hosts and then VMs under that. I also have a M1 MacBook Pro that I had parallels run ARM VMs (mostly MacOS, Windows, and a couple of Debian installs I think).
I have been looking at VMWare alternatives at work so for the hypervisors I’ve been playing around.
I do this stuff for a living but I also do it home for fun and profit. Ok not so much profit. Ok no profit but definitely for the fun. And because I love large electric bills.
I found a prebuilt OpenCore for KVM. https://github.com/thenickdude/KVM-Opencore
I then changed the config.plist to make it think it was a 2019 Mac Pro.
Ok I’ll have to try this. The weird thing is my little test proxmox server is a 2013 trashcan. So this would be like a hackintosh running on Mac hardware. Would that technically be a hackintosh? I’m not really sure. According to the Apple license you can virtualize MacOS if it’s running on Mac hardware. I’m not sure if that requires MacOS as the hypervisor. Regardless this is not something I knew about. Very cool. Thanks for the info.
There are many many many insane people who are running no virtual machines at all.
I have about twice this many VMs and about this many running at any given time.
I use Qubes btw
I run a different LXC on Proxmox for every service, so it's a bunch. Probably a better way to do it since most of those just run a docker container inside them.
Why mix docker and VMs? Isn't docker sort of like a VM, an application-level VM maybe? (I obviously do not understand Docker well)
I have a real use case! I have a commercial server software that can run on Ubuntu or RHEL compatible distributions. My entire environment is Ubuntu. They also allow the server software to run in a docker container but the container must be running RHEL. Furthermore, their license terms require me to build the docker container myself to accept the EULA and the docker image must be built on RHEL! So I have an LXC container running Rocky Linux that gets docker installed for the purpose of building RHEL (Core is 8) imaged docker containers. It’s a total mess but it works! You must configure nested security because this doesn’t work by default.
Instructions here: https://ubuntu.com/tutorials/how-to-run-docker-inside-lxd-containers#1-overview
Serious answer, I'm not sure why someone would run a VM to run just a container inside the VM, aside from the VM providing volumes (directories) to the VM. That said, VMs are perfectly capable of running containers, and can run multiple containers without issue. For work, our Gitlab instance has runners that are VMs that just run containers.
Fun answer, have you heard of Docker in Docker?
I like to run a hypervisor host as just that, a hypervisor host. The host being stable is important, and also reduce attack surface by only having it as that.
An LXC per service is somewhat overkill. A docker host running on LXC could likely run all the docker containers.
I mentioned above, and not to spam, but there might be a use case that requires a different host distribution. Networking isolation might be another reason why. For 90% of use cases, you’re correct.
Not VMs but I have way more docker containers. I run most things as containers which keeps the base OS nice and clean and free from dependency hell.
I mean, people collect all sorts of weird shit
Bahah i have like 7 but im concerned by the fact i probably forgot the password to half of em xD
Yes, but usually they'd have a more robust VM management system to stay sane for long.
On the joke, define "sane". 😬
On a serious note, I think there are valid reasons to have several VMs other than "I was bored". In my case, for example, I have a total of 7 VMs, where 2 are miscellaneous systems to test things out, 2 are for stuff that I can't normally run on Linux, 2 are offline VMs for language dictionaries, and 1 is a BlissOS VM with Google programs in case I can't/don't want to use my phone.
Hell to update them regularly 👀
Nah, most of the windows ones don’t get updates any more and the Linux ones can get a script that updates on boot. Takes longer to start up but handles the job itself.
Yeah.
My home server runs that many, but it's a monster dual xeon.
The freebsd instances have a ton of jails, the Linux vms have a ton of lxc and docker containers.
It's how you run many services without losing your mind.
I've had physical esx servers running this many VMS simultaneously, and I can totally see why a hobbiest or dev would have a need for this many VMs on standby. You are sane, yes
I do have as many too at work.
I use one VM for each iteration of my automation software. Our factory has machines ranging from the 90s to present day, and they use different software environments to be programmed. In order to minimize the risk of data loss, we have one virtual machine with every software environment, that way if one gets corrupted, the damage is contained. It also makes them easier to export to new computers when we need to replace ours.
Looks normal for testing stuff. I have 5ish in my desktop hypervisor.
The biggest reason why I don't want maintain so many Vms is, because all the maintenance and updates that involve doing so.
And that's why there's a "-2" on the end of that arch vm - there was one before that I borked while trying to update it because I hadn't used it in so long.
I think you have a problem, there needs to be more to be normal.
insert MORE, MORE!-Kylo Ren meme here
I always remove any virtual machines every time I'm done with it and reinstall if I need to use it again
How much disk space have you got??
It's a terabyte SSD. I've currently got 136 GB left on it. I think part of it might be they're auto-expanding qcow2 images, so they don't actually take up the full space provisioned for them.
I have probably a couple of more Linux/BSD VMs than here (with some with GPU passthrough and one or two for ARM crossbuilding and so on) but only 2 Windows VMs - the only 2 I have legitimate licenses for.
But am I normal? Most would disagree. 😅
10, plain 11, 7, and funny enough, Server 2022 are all legit licenses (I can get a key for server through my university). Actually, I'm pretty sure the 11 one, I upgraded a Windows 7 VM to 10, then to 11.
Every other Windows version that needs it (11 LTSC, 8.1, and Vista), I just temporarily host a phony KMS server whenever it needs to be reactivated.
I apologize for talking so much about Windows on a Linux sub. May Stallman break into my house and give me 10 lashes as I slumber.
The Windows XP and Windows 7 I have are also from my university, from a long long time ago.😃