Network Monitoring Best Practices for Small to Medium Networks

Why Network Monitoring Matters

If you can't see it, you can't manage it. Network monitoring is not optional for any organization that depends on its network infrastructure — which today means every organization.

The Basics: What to Monitor

At a minimum, you should be tracking:

1. Device Availability (Up/Down)

The most basic check — is the device responding?

• ICMP ping for basic reachability
• SNMP polling for deeper health checks
• Track uptime and reboot events

2. Interface Utilization

Know how much bandwidth your links are consuming:

Key SNMP OIDs for interface monitoring:
- ifInOctets  (1.3.6.1.2.1.2.2.1.10) — Bytes received
- ifOutOctets (1.3.6.1.2.1.2.2.1.16) — Bytes sent
- ifSpeed     (1.3.6.1.2.1.2.2.1.5)  — Interface speed

Calculate utilization: (delta_octets * 8 * 100) / (interval * ifSpeed)

3. Error Rates

Interface errors indicate hardware problems, cable issues, or congestion:

• ifInErrors — Input errors
• ifOutErrors — Output errors
• ifInDiscards — Dropped packets

4. Device Health

For managed switches and routers:

• CPU utilization
• Memory usage
• Temperature
• Fan status
• Power supply status

Setting Up Alerts

Monitoring without alerting is just collecting data nobody looks at. Set up alerts for:

• Critical: Device down, link down, high error rate
• Warning: CPU > 80%, memory > 85%, utilization > 70%
• Informational: Device rebooted, configuration changed

Alert Fatigue

The biggest mistake in monitoring is alerting on everything. This leads to alert fatigue where critical notifications get ignored. Instead:

1. Start with a small set of critical alerts
2. Tune thresholds based on your baseline
3. Use escalation policies (warn → critical → page)
4. Suppress flapping alerts (brief up/down cycles)

Polling Intervals

Choose intervals based on what you're monitoring:

| Data Type | Recommended Interval | |-----------|---------------------| | Device availability | 60 seconds | | Interface traffic | 300 seconds (5 min) | | CPU/Memory | 300 seconds | | Environmental | 600 seconds (10 min) | | Inventory/config | 3600 seconds (1 hour) |

Shorter intervals give more detail but increase load on both the monitoring server and managed devices.

Capacity Planning

Use historical monitoring data to:

• Identify bandwidth trends and predict when upgrades are needed
• Spot underutilized links that could be consolidated
• Plan maintenance windows during low-usage periods
• Justify infrastructure investments with data

Tools to Consider

• Custom NMS — Full control, tailored to your needs
• Zabbix — Open source, feature-rich, good SNMP support
• LibreNMS — Auto-discovery, community-driven
• Grafana + Prometheus — Great for visualization and alerting

Conclusion

Effective network monitoring is about collecting the right data, setting meaningful alerts, and using the information to make proactive decisions. Start simple, monitor the fundamentals, and expand as your network grows.

---

Check out our Services page for professional network monitoring setup assistance.