Beta This project is experimental. Use at your own risk →

Traffic

Previous Next

Monitor real-time bandwidth usage across network interfaces, track cumulative data transfer, and identify bandwidth patterns. The traffic tool reads system network counters and displays throughput in human-readable units.

What it does

The traffic tool provides:

  • Live bandwidth dashboard — Real-time TX (upload) and RX (download) rates for a single interface
  • Interface stats snapshot — Cumulative bytes and packets sent/received since last system boot
  • Per-interface breakdown — See traffic stats for all interfaces or filter to a specific one
  • Automatic unit scaling — Bandwidth displayed in B/s, KB/s, MB/s, or GB/s

The tool uses system network I/O counters (via psutil), so it works on all platforms without requiring packet capture or root privileges (except for advanced packet-level features not yet exposed in the CLI).

Quick start

Show cumulative traffic on all interfaces:

netglance traffic stats

Filter to a specific interface (macOS example):

netglance traffic stats --interface en0

Live bandwidth monitor for 10 seconds (updates twice per second):

netglance traffic live en0

Monitor with custom sampling interval (2-second samples):

netglance traffic live eth0 --interval 2.0

Commands

traffic stats

Display cumulative network interface traffic counters since last boot.

Options:

Option Alias Description
--interface NAME -i Filter to a specific interface (e.g. en0, eth0). If not provided, shows all interfaces.

Output columns:

  • Interface — Network interface name
  • Bytes Sent — Cumulative bytes transmitted (auto-scaled)
  • Bytes Recv — Cumulative bytes received (auto-scaled)
  • Pkts Sent — Total packets transmitted
  • Pkts Recv — Total packets received

Example:

netglance traffic stats
Interface Traffic Stats
┌───────────┬────────────┬────────────┬──────────┬──────────┐
│ Interface │ Bytes Sent │ Bytes Recv │ Pkts Sent│ Pkts Recv│
├───────────┼────────────┼────────────┼──────────┼──────────┤
│ en0       │ 2.34 GB    │ 8.12 GB    │ 1,245,678│ 987,654  │
│ lo0       │ 156.78 MB  │ 156.78 MB  │ 45,123   │ 45,123   │
└───────────┴────────────┴────────────┴──────────┴──────────┘

traffic live

Live bandwidth dashboard for a network interface. Updates in real-time (2x per second).

Arguments:

  • INTERFACE (required) — Network interface name to monitor (e.g. en0, eth0, wlan0)

Options:

Option Alias Description
--interval SECONDS -n Sampling interval in seconds. Default: 1.0. Lower values = more responsive, higher CPU.

Output:

Live table showing:

  • TX (upload) — Bytes per second transmitted
  • RX (download) — Bytes per second received

Press Ctrl+C to stop.

Example:

netglance traffic live en0 --interval 0.5

Updates live:

Live Bandwidth: en0
┌──────────────┬──────────┐
│ Direction    │ Rate     │
├──────────────┼──────────┤
│ TX (upload)  │ 1.23 MB/s│
│ RX (download)│ 45.67 MB/s
└──────────────┴──────────┘

Understanding the output

Bandwidth units

Bandwidth is displayed in bytes per second with automatic scaling:

  • B/s — Bytes per second (< 1 KB/s)
  • KB/s — Kilobytes per second (≥ 1 KB/s, < 1 MB/s)
  • MB/s — Megabytes per second (≥ 1 MB/s, < 1 GB/s)
  • GB/s — Gigabytes per second (≥ 1 GB/s)

Note: 1 KB = 1024 bytes (binary, not decimal).

TX vs RX

  • TX — Transmit (upload). Bytes your device sent out.
  • RX — Receive (download). Bytes your device received.

Throughput vs cumulative transfer

The live command shows instantaneous bandwidth (bytes per second right now). The stats command shows cumulative transfer (total since boot or last reset), useful for checking daily/monthly limits.

Interface names

Common interface names by OS:

  • macOS — en0 (Wi-Fi), en1 (Ethernet), lo0 (loopback)
  • Linux — eth0, wlan0, lo (loopback)
  • Windows — Interface numbers or descriptive names

List all interfaces on your system:

netglance traffic stats

The interface names shown are the ones available on your system.

  • Speed test (netglance speed) — Measure maximum download/upload capacity. Compare this to current bandwidth from traffic live to see utilization.
  • Performance analysis (netglance perf) — Measure latency, jitter, and packet loss alongside bandwidth.
  • Daemon & monitoring (netglance daemon) — Collect continuous traffic samples in the background for historical trending.

Troubleshooting

Interface not found

Error: Interface 'en0' not found. Available: ...

Fix: Check your system's actual interface names:

netglance traffic stats

The first column lists all available interfaces. Use that name with --interface.

Live monitor shows zero or very low bandwidth

Cause: Interface is idle or sampling interval is too short to capture traffic.

Fix: - Ensure traffic is actually flowing on that interface (e.g., open a browser tab and stream a video) - Increase the sampling interval: --interval 2.0 instead of --interval 1.0

Counters reset unexpectedly

Cause: System reboot, or the interface was disabled and re-enabled.

Behavior: netglance traffic stats shows cumulative counters from the OS kernel. These reset on reboot or device restart.

Fix: For historical tracking, use the daemon (netglance daemon) to log snapshots to the database before reboots.

VPN or tunnel traffic appears twice

Cause: Some VPN software creates virtual interfaces that duplicate traffic from physical interfaces.

Behavior: Total traffic from traffic stats exceeds actual WAN usage.

Fix: Monitor only your physical interface (e.g., en0 for Wi-Fi, not utun0 for VPN), or subtract tunnel interface counts manually.

Uptime Dns