Proxmox VE
virtualization as an operations tool

What Proxmox VE is

Proxmox Virtual Environment is a Debian distribution that combines two hypervisor technologies in a single interface: KVM for full virtual machines and LXC for system containers. Both are managed from the same web console, with the same storage, network and backup management.

It is free software, with no per-core or per-machine license, which matters when you operate a few servers for clients and do not want a recurring cost indexed to the number of processors.

The five advantages that decide, for us

• One role per virtual machine the web server, the database and batch processing share neither libraries, nor versions, nor the same attack surface. A failed update only takes down one component.
• Snapshots before changes before an upgrade, a snapshot: if the service does not restart, you roll back in thirty seconds instead of restoring last night's backup.
• Built-in backups the native tool backs up a complete machine, hot, without any agent to install. With Proxmox Backup Server, backups become deduplicated and verifiable.
• Controlled networking one bridge per VLAN, an admin interface on a separate network: segmentation happens at the hypervisor, not in application configuration.
• Automation a complete API and command-line interface. Creating a machine, backing it up, restoring it or migrating it can be scripted. That is what lets us use the same procedure for a client server and our own.

What you need to know before starting

• A snapshot is not a backup it lives on the same storage as the machine. A disk failure takes out both.
• ZFS needs memory the filesystem cache draws from RAM. Under-dimensioning the machine to "save" a few gigabytes is paid in latency.
• High availability requires shared storage without Ceph or external storage, a cluster cannot do automatic failover. It is not a checkbox, it is an architecture.
• The hypervisor does not replace the firewall segmenting internally does not protect from the Internet. Filtering remains the job of dedicated equipment, upstream.

How we use it

A recent physical machine, mirrored storage, and as many virtual machines as roles. Services that must stay up do not share the same virtual machine as those that can restart. Backups go to a separate, dedicated server, with retention and tested restoration, because a backup that has never been restored is not a backup.

Nothing exotic: it is operations. The hypervisor does not make you better; it makes mistakes reversible, which is already the essential part.