High-resolution network monitoring with ping

The Pinger Program

To use these observations in practice for long-term monitoring of network connections, I wrote the Pinger [5] command-line program, which calls the ping executable installed on a system only once per second in the current version.

On the gigabit LAN, Pinger thus normally achieves adequate resolution for a daily average RTT of (as low as) 153ns, which corresponds to around 15 meters of cable length. The 1,000-second average provides a resolution of (as low as) 1.5µs, which is equivalent to 1/20th of the additional latency from the gigabit switch. To monitor 1,000 connections with a 28-byte ping, the C program only needs 2*28KBps of bandwidth. For a higher resolution, users can modify the Pinger source code to ping faster – for example, every millisecond.

The program consists of a main thread that starts 10 threads, each of which sends and evaluates a ping every other second. A couple of other functions perform analysis and output. To avoid manipulation of the pings, the program fills each with a 128-bit random number. They are also flagged don't fragment so that ping sends them in one piece.


Pinger only outputs the exponential moving averages in the current version. Compared with the arithmetic mean values (i.e., the values for the last n measurements), you reach a much shorter delay with the same degree of smoothing, or the same delay with considerably better smoothing. The disadvantage is that exponential smoothing keeps old values for a long time (exponential decay), whereas the arithmetic mean values no longer consider the old values after n measurements.

To visualize short-term, medium-term, and long-term changes in the RTT and the return values from ping, the averaging uses smoothing factors 0.1, 0.001, and 1/86,400, for (exponential) 10-second, 1,000-second, and daily mean values, respectively. If you want, you can compute other parameters from this – for example, the standard deviation and other moments of the distribution function – also with smoothing if needed.

Program Start

Pinger source code [5] can be compiled using the command

gcc -D_REENTRANT -Wall -O3 -lm -pthread -o pinger \
  pinger.c && strip pinger

to create the pinger executable. You need to launch the resulting program as root for high priority. It expects three parameters: the IP of the target computer, the cut-off RTT in nanoseconds, and the net ping size in bytes:

./pinger 400000 16

For troubleshooting purposes, the program writes run times to rtt* files and the ping return values to retval* files in the working directory. The plotting3.sh [5] GNU plot script is used for evaluation in the form of a 2D plot:

./plotting3.sh rtt_10s_16B_192.168.1.1.txt | gnuplot

It refreshes the display every second, so you can also follow a live session.

Buy this article as PDF

Express-Checkout as PDF
Price $2.95
(incl. VAT)

Buy Linux Magazine

Get it on Google Play

US / Canada

Get it on Google Play

UK / Australia

Related content

  • Security Lessons: Bufferbloat

    An abundance of buffers hides the Internet’s dirty little secret.

  • Programming Snapshot – Go Network Diagnostics

    Why is the WiFi not working? Instead of always typing the same steps to diagnose the problem, Mike Schilli writes a tool in Go that puts the wireless network through its paces and helps isolate the cause.

  • Zing

    Zing is a lightweight, zero-packet network utility similar to ping that provides ping functionality without the payload.

  • Command Line: Network Diagnostic Tools

    Linux has the right tools to track down network errors and open the way for data packets.

  • Charly's Column

    If you do not receive a response to a ping, or if the response is seriously delayed, you might like to take this as a warning. But who wants to ping all day? You need a ping-based monitoring utility like Smokeping.

comments powered by Disqus
Subscribe to our Linux Newsletters
Find Linux and Open Source Jobs
Subscribe to our ADMIN Newsletters

Support Our Work

Linux Magazine content is made possible with support from readers like you. Please consider contributing when you’ve found an article to be beneficial.

Learn More