Proxmox VE guide (1): what it is and how to install it step by step
Part one of the Proxmox VE 9 guide: what a hypervisor is, what you need, how to prepare the USB stick and every installer screen explained.
Where this guide comes from: at work I administer a Proxmox VE cluster every day, with Ceph storage and backups to Proxmox Backup Server. This series is the version for starting from scratch at home, with a single server, and follows the official Proxmox VE 9.2 documentation. At the end of each part I share what I’ve learned doing it for real on the cluster at work.
What Proxmox VE is
Proxmox Virtual Environment is an operating system built for one thing: running many machines inside one computer. It’s what’s called a type 1 hypervisor: it installs directly on the hardware, with no Windows or other system underneath.
Under the hood it’s Debian (version 9 is based on Debian 13), with two virtualisation technologies:
- Virtual machines (KVM/QEMU): fully simulated computers, each with its own operating system. You can run Windows, Ubuntu and a lab server at the same time.
- LXC containers: lightweight Linux systems that share the host’s kernel. They start in seconds and use very little memory.
Everything is managed from a very complete web interface. It’s free and open source; the paid subscription gives access to a more tested update repository and support, but isn’t required to use it.
Why would you want it at home?
- Build a lab for studying (Windows and Linux servers, networking) without filling the house with computers.
- Keep several services separate on one machine: if one breaks, the others keep running.
- Take snapshots before changing something and roll back in seconds if it goes wrong.
What you need
| Component | Minimum | Recommended for home |
|---|---|---|
| CPU | 64-bit with Intel VT-x or AMD-V | 4 cores or more |
| RAM | 2 GB for Proxmox + whatever your machines need | 16 GB or more |
| Disk | Anything | SSD; two if you want a mirror (RAID 1) |
| Network | Ethernet | Wired, not Wi-Fi |
Three important details:
- Enable virtualisation in the BIOS/UEFI (usually called Intel VT-x, Intel Virtualization Technology, SVM or AMD-V). Without it, no accelerated VMs.
- Use a network cable. Proxmox connects VMs to the network through a “bridge”, and that doesn’t work well over Wi-Fi.
- Proxmox wipes the disk you install it on. Use a dedicated disk.
If you plan to use ZFS (a very robust filesystem the installer offers), budget more RAM: the usual rule is 1 GB per TB of storage.
A second-hand mini PC or an old desktop with 16 GB of RAM is a great starting point.
Step 1: download the ISO and prepare the USB stick
- Download the Proxmox VE ISO from the official website (proxmox.com, Downloads section).
- Write it to a USB stick of at least 2 GB:
- On Windows: Rufus (choose DD mode when asked) or balenaEtcher.
- On Linux:
ddor balenaEtcher.
With Rufus, if you pick ISO mode instead of DD, the stick may not boot. If that happens, write it again in DD mode.
Step 2: boot the installer
Boot the computer from the USB stick (usually F11, F12 or Esc at power-on) and choose Install Proxmox VE (Graphical).
Step 3: the installer screens
Target disk
Choose the disk to install on. Under Options you can pick the filesystem:
- ext4 (default): simple, with LVM-thin for VM disks. Perfect for starting with a single disk.
- ZFS (RAID1): if you have two identical disks, it mirrors them. If one dies, everything keeps working.
Country, time zone and keyboard
Set your country, time zone and keyboard layout. The keyboard matters: you’ll use it on the console.
Password and email
- The root password, the all-powerful admin. Make it long and keep it in your password manager.
- An email: Proxmox sends alerts there (failed backups, disk problems…).
Management network
This is the most important screen:
- Interface: the wired network card.
- Hostname (FQDN): a full name, for example
pve.home.lan. - IP address: a fixed IP for the server, for example
192.168.1.70/24. Keep it out of the range your router hands out, or use a DHCP reservation. - Gateway: your router’s IP (usually
192.168.1.1). - DNS: your router or a public DNS.
Summary
Check everything and press Install. After a few minutes it finishes, reboots, and you remove the USB stick.
Step 4: first login
On boot, the server’s screen shows the admin address:
https://192.168.1.70:8006/
Open it from another computer on the network:
- The browser will warn that the certificate isn’t trusted. That’s normal: Proxmox uses a self-signed certificate. Accept the risk to continue.
- User root, the password you set, and realm Linux PAM standard authentication.
- You’ll get a “No valid subscription” notice. Also fine: I explain it in the next part.
At work: why a cluster means three nodes
At home one server is plenty. At work (public sector) we run a three-node cluster with Ceph storage and backups on a separate Proxmox Backup Server. Two things about the install change when you go from one node to several:
- Static IPs everywhere. There’s no DHCP on that network: each node has its service IP plus another one on a management network. If a node changed IP, the cluster would lose it.
- Quorum. A cluster only makes decisions when more than half the nodes agree. With two nodes, if one goes down the other can’t tell whether it’s the one that’s cut off (so it locks up rather than do something stupid). With three, two are still a majority.
We checked this in a continuity test: we powered off an entire node and the other two kept quorum, both Proxmox’s and the Ceph monitors’. Services living on the other nodes didn’t even notice.
If you want a two-machine cluster at home, you need a third vote: a QDevice (a small service on another machine, a Raspberry Pi for example) that acts as tie-breaker.
Another rule we follow: all three nodes on the same version. They’re updated together, in the same window; part 2 explains how.
And Proxmox is up and running. In part 2 we set it up properly: update repositories, users, two-factor authentication and networking.
Did it stick?
Three quick questions. Each right answer is worth 10 XP.