Back up the configuration of your Proxmox VE nodes

Why

Proxmox VE backup jobs save your VMs and containers, not the node they run on. The node has its own configuration: bridges, bonds and VLANs in /etc/network/interfaces, /etc/hosts, storage definitions, the cluster configuration in /etc/pve, certificates, SSH keys, apt repositories, kernel parameters, the cron jobs and scripts you added.

When a node’s boot disk dies, or someone breaks a file by mistake, that configuration has to come back. Rebuilding it by hand from memory and screenshots is slow and easy to get wrong: a bridge with a different name, a missing VLAN, and guests do not start on the reinstalled node.

cv4pve-node-protect copies those files from every node into a dated tar.gz, on a schedule, so you can compare what changed and put back exactly what was there.

Where it fits

The cv4pve suite follows the Unix philosophy: each tool does one thing and does it well. cv4pve-node-protect protects the node, the other tools the guests and the knowledge of the cluster.

Tool	Protects	How
cv4pve-node-protect	Node configuration files	SSH, tar.gz per node
Proxmox VE backup / Proxmox Backup Server	VMs and containers	Built into Proxmox VE
cv4pve-autosnap	VM and container snapshots on a schedule	Proxmox VE API
cv4pve-report	A readable inventory of the cluster, to document it	Proxmox VE API

Prefer a web interface with scheduled backups? cv4pve-admin runs the same engine as its Node Protect module.

Outside the nodes, over SSH

cv4pve-node-protect runs outside the cluster (on your workstation, a management VM or the machine that keeps your backups) and connects to each node over SSH, not through the Proxmox VE API. On every node it runs a single tar command and streams the archive back; nothing is installed and nothing is written on the node.

Reading the whole configuration needs root, and the archives hold secrets such as /etc/shadow, SSH keys and storage passwords. SSH access and security explains the account, the host key and how to keep the archives safe: read it before the first run.

What an archive contains

One run with the recommended paths on a node of our test cluster, grouped (about 1 MB compressed):

Part	Examples
Network	/etc/network/interfaces, /etc/hosts, /etc/hostname, /etc/resolv.conf
System	/etc/fstab, /etc/kernel/cmdline, /etc/modprobe.d/, /etc/apt/sources.list.d/, /etc/vzdump.conf, /etc/zfs/, /etc/lvm/
Cluster	/etc/corosync/corosync.conf, and /var/lib/pve-cluster/config.db: the database behind /etc/pve
/etc/pve as files	storage.cfg, datacenter.cfg, user.cfg, jobs.cfg, firewall rules, guest configurations, node certificates pve-ssl.pem / pve-ssl.key, priv/
Accounts and keys	/etc/shadow, SSH host keys, /root/.ssh/, /var/spool/cron/crontabs/

What it does for you

01

The whole cluster in one run

Pass every node in --host: one archive per node, all in the same dated folder.

02

You choose what goes in

/etc, the readable /etc/pve files, the cluster database, crontabs, root SSH keys, your own scripts.

03

Nothing left on the nodes

tar writes to the SSH channel and the archive lands straight in your local file: no temporary files, no agent.

04

Retention built in

--keep removes the oldest dated folders after each run, and never touches other folders.

05

A restore you control

Plain tar.gz files: extract to a staging folder, compare, then put back only what you need.

06

Scheduled and scriptable

cron or Task Scheduler, an options file for the connection, exit code 1 when a run fails.

Part of the cv4pve suite →

Take your first node backup

Download the binary, point it at your nodes over SSH, open the archive.

Made by Corsinvest,
official Proxmox partner

Get startedProfessional support