Hands-on tutorial

Tutorials · Monitoring

How to Monitor WAN Latency, Jitter and Packet Loss

Measure the quality of internet and VPN links continuously, so you can prove when the ISP is at fault and catch problems before video calls and VoIP suffer.

Intermediate9 min6 steps

Before you start

  • A gateway or small Linux host on the site
  • Two or three stable targets: the ISP gateway, a public DNS resolver and a server at head office
  • A place to store and graph results

Step by step

  1. Pick the targets

    Probe the first ISP hop (tests the access line), a stable public resolver (tests the internet path), and the head office over VPN (tests the business path).

  2. Measure every minute

    Send a small burst of pings every minute and record average latency, jitter (variation) and loss. A single ping is not enough.

    ping -c 20 -i 0.2 -q 9.9.9.9
    
  3. Know the thresholds

    For voice and video: latency under 150 ms one way, jitter under 30 ms, and loss under 1%. Above that, users notice.

  4. Graph and keep history

    Store results in your monitoring tool. A month of history shows patterns such as evening congestion or a line that drops every night.

  5. Alert on sustained problems

    Alert when loss is above 2% or latency doubles for 5 minutes or more, not on single spikes.

  6. Use the data with the ISP

    Send the ISP graphs with timestamps, the target addresses and the hop where loss starts. A traceroute or mtr at the moment of the problem helps.

    mtr -rwc 100 9.9.9.9
    

Commands are generic, industry-standard examples. Exact syntax and menu names vary by model and firmware; check your device's manual on the Documentation page.

Pro tips

  • Some routers deprioritise ping replies; judge loss at the final target, not at middle hops.
  • On dual-WAN sites, monitor each link separately so failover decisions use real quality.
  • Compare against the contracted SLA.