How to Monitor Proxmox VE Performance

Why Monitoring Proxmox VE Matters

Running a Proxmox VE environment without proper monitoring is like driving without a dashboard. You might get away with it for a while, but when something goes wrong — a disk fills up, a VM consumes all available RAM, or CPU usage spikes — you will not know until users start complaining or services go down. Effective monitoring gives you visibility into your infrastructure, helps you plan capacity, and lets you catch problems before they become outages.

This guide covers every layer of Proxmox monitoring, from the built-in tools that ship with every installation to enterprise-grade solutions with Grafana and Prometheus, and even mobile monitoring with ProxmoxR so you can keep an eye on your cluster from anywhere.

Built-In Proxmox VE Monitoring

RRD Graphs in the Web UI

Proxmox VE includes built-in RRD (Round Robin Database) graphs accessible directly from the web interface. Every node, VM, and container has a Summary tab that displays time-series graphs for CPU usage, memory consumption, network traffic, and disk I/O. These graphs cover the last hour, day, week, month, and year.

To access them, select any node or guest in the left sidebar and click the Summary tab. While these graphs are useful for quick spot-checks, they have limitations: low resolution over longer time periods, no alerting capabilities, and no way to create custom dashboards.

Resource Overview

The Datacenter view provides a high-level overview of resource usage across your entire cluster. You can see total and allocated CPU cores, memory, and storage at a glance. The Notes panel on each node also lets you document important details about the hardware.

For a quick CLI-based status check, SSH into any node and run:

# Check overall node resource usage
pvesh get /nodes/$(hostname)/status

# List all VMs with their status and resource usage
qm list

# List all containers
pct list

# Check storage usage across all configured storage
pvesm status

Grafana + InfluxDB: Building a Monitoring Dashboard

For serious monitoring, most administrators set up Grafana with InfluxDB as the time-series database. Proxmox VE has native support for sending metrics to external databases, making this integration straightforward.

Step 1: Install InfluxDB

Deploy InfluxDB on a separate VM or container to avoid impacting your hypervisor's performance:

# Install InfluxDB on a Debian/Ubuntu system
apt-get update
apt-get install -y influxdb
systemctl enable --now influxdb

# Create the Proxmox database
influx -execute "CREATE DATABASE proxmox"
influx -execute "CREATE RETENTION POLICY autogen ON proxmox DURATION 90d REPLICATION 1 DEFAULT"

Step 2: Configure Proxmox to Send Metrics

On your Proxmox node, edit the metric server configuration:

# Add InfluxDB as a metric server
cat >> /etc/pve/status.cfg <<EOF
influxdb: influx1
    server 192.168.1.50
    port 8089
EOF

# Restart the status daemon
systemctl restart pvedaemon

Step 3: Install and Configure Grafana

# Install Grafana
apt-get install -y grafana
systemctl enable --now grafana-server

Access Grafana at http://your-server:3000, add InfluxDB as a data source pointing to the proxmox database, and import a community dashboard. Dashboard ID 10048 is one of the most popular Proxmox dashboards on Grafana's marketplace and gives you comprehensive visibility into node and guest metrics.

Prometheus and Exporters

If you prefer the Prometheus ecosystem, Proxmox VE also supports Prometheus-compatible metric export. The pve-exporter project exposes Proxmox metrics in a format Prometheus can scrape.

# Install the Proxmox VE exporter (on a monitoring host)
pip install prometheus-pve-exporter

# Create a configuration file
cat > /etc/pve-exporter.yml <<EOF
default:
    user: monitoring@pve
    password: your-secure-password
    verify_ssl: false
EOF

# Run the exporter
pve_exporter /etc/pve-exporter.yml --listen-address 0.0.0.0 --listen-port 9221

Then add the scrape target to your Prometheus configuration:

# prometheus.yml
scrape_configs:
  - job_name: 'proxmox'
    metrics_path: /pve
    params:
      target: ['192.168.1.10']  # Proxmox node IP
    static_configs:
      - targets: ['localhost:9221']

Setting Up Alerts

Monitoring without alerting is only half the picture. You need to be notified when something goes wrong. Here are the most critical alerts every Proxmox administrator should configure:

• CPU usage above 90% sustained for more than 5 minutes
• Memory usage above 85% on any node
• Disk usage above 80% on any storage pool
• Node offline — any cluster node becomes unreachable
• Backup failures — scheduled backups did not complete successfully
• CEPH health warnings — degraded pools, slow OSDs

In Grafana, you can create alert rules directly on any panel. For Prometheus, use Alertmanager with rules like:

# alert.rules.yml
groups:
  - name: proxmox
    rules:
      - alert: NodeHighCPU
        expr: pve_cpu_usage_ratio > 0.9
        for: 5m
        labels:
          severity: warning
        annotations:
          summary: "High CPU usage on {{ $labels.id }}"

Mobile Monitoring with ProxmoxR

All the tools above are powerful, but they assume you are sitting at a desk with a browser open. In practice, you need to check on your infrastructure from anywhere — while commuting, during meetings, or when an alert wakes you up at 3 AM.

ProxmoxR brings full Proxmox monitoring to your phone. The app connects directly to your Proxmox VE API and provides real-time performance graphs for every node, VM, and container in your cluster. You can see CPU, memory, network, and disk metrics at a glance without needing to open a laptop or VPN into your network first.

Key monitoring features in ProxmoxR include:

• Real-time resource graphs — View live CPU, memory, network, and disk I/O charts for any node or guest, rendered natively on your device for a smooth experience.
• Multi-cluster overview — If you manage multiple Proxmox clusters (production, staging, homelab), ProxmoxR lets you monitor all of them from a single app, switching between clusters instantly.
• Quick status checks — See at a glance which VMs are running, stopped, or paused across your entire infrastructure without navigating through multiple web UI pages.
• Instant power actions — If monitoring reveals a problem, you can immediately take action: restart a hung VM, shut down a runaway container, or start a backup — all from the same app.

ProxmoxR pairs especially well with traditional monitoring stacks. Use Grafana and Prometheus for deep historical analysis and complex alerting rules, and use ProxmoxR for on-the-go visibility and quick interventions when alerts fire.

Best Practices for Proxmox Monitoring

1. Monitor the hypervisor separately from guests — Node-level and VM-level metrics serve different purposes. Track both.
2. Set retention policies — Keep high-resolution data for a week and aggregated data for a year. This balances detail with storage costs.
3. Use dedicated monitoring infrastructure — Run your monitoring stack on a separate system so it stays available even when your Proxmox cluster has issues.
4. Test your alerts — Simulate failure scenarios to make sure alerts actually fire and reach you.
5. Keep a mobile option ready — Tools like ProxmoxR ensure you are never blind to your infrastructure, even when you are away from your desk.

Pro tip: Start with Proxmox's built-in monitoring and ProxmoxR for mobile access. Add Grafana and InfluxDB when you need historical trending and custom dashboards. Layer in Prometheus and Alertmanager when you need sophisticated alerting rules.