Proxmox cluster + PBS

Deployment of a Proxmox cluster with more than 7 nodes for production and development environments, plus hot backups with Proxmox Backup Server. Before: isolated servers with no high availability or automated backups.

Key metrics

RPO: no backups → 1 h
RTO: hours/days → 30-60 min
Retention: 30 days
Restore test: monthly

Key facts

Scope: Proxmox cluster with 7+ nodes across production and development.
Backups: 2 daily (every 12 h) for critical VMs.
Backups: 1 daily (every 24 h) for medium VMs.
Backups: 1 weekly (every 7 days) for low-priority VMs.
Goal: operational continuity, scalability, and fast recovery.

Stack

Proxmox VE
Proxmox Backup Server
Linux

Sanitized continuity diagram

Proxmox cluster
Critical VMs
PBS backups
Monthly restore test

Logical view without node names, datastores, IPs, or customer data.

Context

The ISP ran on physical servers and isolated VMs. There was no high availability, backups were manual, and there was no fast recovery from failures. Critical infrastructure needed to be consolidated with guaranteed operational continuity.

Actions

Designed and deployed the cluster with dedicated nodes for production and development.
Standardized VM templates and hot backup policies with tiered retention (12h / 24h / 7d).
Monitored cluster health and ran monthly restore tests.

Anonymized Proxmox Backup Server panel with backup tasks and datastore status — Anonymized PBS panel: backup tasks, retention, and overall repository status.

Results

Recovery from incidents in under 60 minutes for critical services (it used to take hours or days).
A stable, scalable foundation to run 50+ VMs in production with high availability.
Automated backups: human error eliminated and operational risk reduced.