Building a multi-node Proxmox cluster
What a cluster brings, what it costs, and why three nodes beat two.
Starting at from €13.99/mo
Introduction
A Proxmox cluster gathers several servers under one interface: you see all your virtual machines in one place and move them from node to node, often without stopping them. That is the first benefit, and it already justifies itself for maintenance alone.
The second — high availability, a VM restarting elsewhere when its node dies — asks for more. It needs a quorum, meaning a majority of nodes that agree, which is why three nodes beat two: with two, losing one loses the majority and the cluster freezes as a safety measure.
You also need shared or replicated storage, otherwise the VM arrives without its disks. Ceph or ZFS replication answer that, at the price of some network and a lot of memory. Plan for the L plan or above per node: at this scale, 16 GB no longer leaves headroom.
What the server brings, across the whole range
Proxmox VE already installed
The server arrives with <strong>Proxmox VE</strong> in place and its web interface reachable. You create your virtual machines and LXC containers from the browser, without a support ticket or a wait. <strong>Reinstalls are unlimited</strong>: an experiment that goes wrong is redone in minutes.
RAM, because that is what runs out first
On a hypervisor it is almost never the processor that caps out first: it is <strong>memory</strong>, because every virtual machine reserves its own. The range starts at <strong>16 GB</strong> and goes up to <strong>96 GB</strong>, which leaves room to stack several VMs without starving them.
Unlimited snapshots and backups
A snapshot before an update, another before a configuration change: that is what separates a test from a gamble. Neither the count nor the frequency is billed, so nothing discourages taking one more.
Up to 16 additional IPv4 addresses
Each VM can hold its own public address, which avoids routing everything through port forwards on the host. Useful as soon as a virtual firewall, a web server and a third-party service have to coexist cleanly.
What people actually want from it
Maintenance without downtime
Migrate the VMs off a node, reboot the empty node, bring them back. It is the most common use, and it does not even require high availability.
One interface for everything
Three servers, one window. Backups, templates and permissions are handled at cluster level rather than three times over.
Automatic restart after failure
With HA enabled and shared storage, a VM whose node disappears restarts on another. This is not continuity: there is an outage, a short one.
Spread the load
A hungry VM on one node, the small ones elsewhere. The move happens live, with no maintenance window to announce.
Learn clustering
Three VDS cost less than three physical machines and reinstall without limit. It is the cheapest way to practise quorum and fencing for real.
Frequently asked questions
Two are enough to share an interface and migrate VMs. For high availability, count three: with two, losing one loses quorum and the survivor stands down as a safety measure.
No. The VM restarts on another node, which takes as long as a boot — usually tens of seconds. This is not service continuity, it is an automatic restart.
It is strongly recommended for cluster and storage traffic. Corosync is latency-sensitive, and pushing replication over the public network mixes two things with very different profiles.
Yes, a Proxmox cluster grows without starting over. Ordering an extra VDS and joining it to the cluster is routine; so is the reverse.