.\" Copyright (c) 2000 Paul Herman
.\" All rights reserved.
.\"
.\" Redistribution and use in source and binary forms, with or without
.\" modification, are permitted provided that the following conditions
.\" are met:
.\" 1. Redistributions of source code must retain the above copyright
.\"    notice, this list of conditions and the following disclaimer.
.\" 2. Redistributions in binary form must reproduce the above copyright
.\"    notice, this list of conditions and the following disclaimer in the
.\"    documentation and/or other materials provided with the distribution.
.\"
.\" THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
.\" ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
.\" IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
.\" ARE DISCLAIMED.  IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
.\" FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
.\" DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
.\" OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
.\" HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
.\" LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
.\" OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
.\" SUCH DAMAGE.
.\"
.\" $Id: tcpstat.1,v 1.33 2003/01/24 19:52:44 pherman Exp $
.\" ---
.\" Paul says: man-o-man I just went through 2 whole bags of doritos(tm)
.\" trying to write this manpage by hand.  There's gotta be a better
.\" way to do this . . .
.\"
.Dd April 29, 2000
.Dt TCPSTAT 1
.Os
.Sh NAME
.Nm tcpstat
.Nd report network interface statistics
.Sh SYNOPSIS
.Nm tcpstat
.Op Fl ?haeFlp
.Op Fl B Ar bps
.Op Fl b Ar bps
.Op Fl f Ar filter expr
.Op Fl i Ar interface
.Op Fl o Ar output
.Op Fl R Ar seconds
.Op Fl r Ar filename
.Op Fl s Ar seconds
.Op Ar interval
.Sh DESCRIPTION
.Nm
reports certain network interface statistics much like
.Xr vmstat 8
does for system statistics.  Statistics include bandwidth being used,
number of packets, average packet size, and much more.
.Pp
Network information is collected either by reading data from
.Ar filename ,
or by directly monitoring the network interface
.Ar interface .
The default action for
.Nm
is to automatically search for an appropriate
interface, and to show current statistics on it.
.Pp
.Ar interval 
is the sample interval, in seconds, in which the statistics are based upon
and when in default mode, how often the display is updated. If \-1 is
given, then the interval is taken to be the entire length of the sample.  
Default is 5 seconds.
.Pp
When reading data from
.Ar filename ,
.Nm
will exit immediately after the entire file has been processed.  When
collecting data from
.Ar interface ,
.Nm
will keep running unless the
.Fl s
option had been specified.
.Pp
.Sh OPTIONS
The options are as follows:
.Bl -tag -width Fl
.It Fl a
Accounting mode.  Displays the estimated number of bytes per second,
minute, hour, day, and month.
.It Fl b Ar bps
Bandwidth mode.
Displays the total number of seconds the
data-throughput exceeded
.Ar bps ,
and the percentage of total time this was,
as if the interface were limited to
.Ar bps
bits per second.  See the
.Sx NOTES
section below to see how the
.Ar interval
affects bandwidth calculation.
.It Fl B Ar bps
"Dumb" bandwidth mode.  Displays the total number of seconds the
data-throughput exceeded
.Ar bps ,
and the percentage of total time this was.  See the
.Sx NOTES
section below to see difference between "dumb" and normal
bandwidth modes.
.It Fl e
Suppresses the display of empty intervals.
.It Fl F
Flush all output streams after printing each interval.  Sometimes useful
when redirecting output into a file, or piping tcpstat into another
program like
.Xr grep 1 .
.It Fl f Ar filter expr
Filter the packets according the rules given by
.Ar filter expr .
For the syntax of these rules, see
.Xr tcpdump 1 .
The argument must be quoted if it contains spaces in order to separate it
from other options.
.It Fl h , Fl ?
Display version and a brief help message.
.It Fl i Ar interface
Do a live capture (rather than read from a file) on the interface
.Ar interface
given on the command line.  If
.Ar interface
is "auto" then
.Nm
tries to find an appropriate one by itself.
.It Fl l
Include the size of the link-layer header when calculating statistics.
(Ethernet only, right now.  Usually 14 bytes per packet.)
.It Fl p
Set the interface into non-promiscuous mode (promiscuous
is the default) when doing live captures.
.It Fl o Ar format
Set the output format when displaying statistics.  See the
.Sx OUTPUT FORMAT
section below for a description of the syntax.
.It Fl R Ar seconds
Show the timestamp relative to
.Ar seconds .
Avoid this option, because it will most likely go away in future versions.
.It Fl r Ar filename
Read all data from
.Ar filename ,
which may be a regular file, a named pipe or "-" to read it's data from
standard input. Acceptable file formats include pcap
.Xr (tcpdump 1
files) and "snoop" format files.
.Ar filename
is usually a file created by the
.Xr tcpdump 1
command using the "\-w" option.
.It Fl s Ar seconds
When monitoring an interface, 
.Nm
runs for only 
.Ar seconds 
seconds, and then quits.  When reading from a data file,
.Nm
prints statistics for
.Ar seconds
seconds relative to the first packet seen.
.El
.Sh OUTPUT FORMAT
The
.Ar output
string is any quoted string, and
.Nm
will write this string to the stdout.  In addition,
.Nm
will substitute certain values for substrings which begin with a "%", as
well as most standard
.Xr printf 3
"\\" escape characters. Here is a list of all substitution strings:
.Bl -tag -width %%%%
.It % Ns A
the number of ARP packets
.It % Ns a
the average packet size in bytes
.It % Ns B
the number of bytes per second
.It % Ns b
the number of bits per second
.It % Ns C
the number of ICMP and ICMPv6 packets
.It % Ns d
the standard deviation of the size of each packet in bytes
.It % Ns I
the number of IPv4 packets
.It % Ns l
the network "load" over the last minute, similar to
.Xr uptime 1
.It % Ns M
the maximum packet size in bytes
.It % Ns m
the minimum packet size in bytes
.It % Ns N
the number of bytes
.It % Ns n
the number of packets
.It % Ns p
the number of packets per second
.It % Ns R
same as %S, but relative to the first packet seen
.It % Ns r
same as %s, but relative to the first packet seen
.It % Ns S
the timestamp for the interval in seconds after the "UNIX epoch"
.It % Ns s
the timestamp for the interval in seconds.microseconds after the "UNIX epoch"
.It % Ns T
the number of TCP packets
.It % Ns U
the number of UDP packets
.It % Ns V
the number of IPv6 packets
.It % Ns Ar number
switch the output to the file descriptor
.Ar number
at this point in the string.  All output for each interval before this
parameter is by default the standard output (file descriptor 1).  Useful
when redirecting the output into more than one file (or fifo) for separate
statistics.  Be sure you know where they are going.  Writing to "dangling"
file descriptors (without directing them to a specific destination) may
produce unexpected results.
.It % Ns %
the "%" character
.El
.Pp
The default
.Ar format
string for
.Nm
is:
.Pp
"Time:%S\\tn=%n\\tavg=%a\\tstddev=%d\\tbps=%b\en"
.Pp
which will produce an output which would look similar to:
.Bd -literal -offset left
Time:940948785	n=107	avg=251.81	stddev=422.45	bps=43110.40
Time:940948790	n=99	avg=400.21	stddev=539.39	bps=63393.60
Time:940948795	n=43	avg=257.16	stddev=352.83	bps=17692.80
.Ed
.Pp
It is worth noting for example, that many of the protocol filters (%T, %U,
etc.) may be seen as being redundant because protocols can be filtered
using
.Fl f
(see
.Sx OPTIONS
above)
.Sh SIGNALS
Upon receiving a SIGINT,
.Nm
will print any remaining statistics, and then exit.
Upon receiving a SIGUSR1 when printing intervals,
.Nm
will print the current statistics immediately.  This can be useful
when using an interval length of "\-1" to print statistics on demand.
.Sh FILES
.Bl -tag -width /dev/bpfXXX -compact
.It Pa /dev/bpf Ns Sy n
the packet filter device
.El
.Sh EXAMPLES
.Dl tcpstat -i fxp0
.Pp
Displays the default statistics every 5 seconds of all traffic currently
passing through the fxp0 network interface.
.Pp
.Dl tcpstat -r file.dump
.Pp
Displays the default statistics every 5 seconds from the
.Xr tcpdump 1
generated file "file.dump".
.Pp
.Dl "tcpstat -f 'port (smtp or http)' -o '%S %b\en' -r file.dump" 2.3
.Pp
Displays every 2.3 seconds the timestamp together with smtp and http
traffic throughput of the data from "file.dump", in a format which would
be suitable for
.Xr gnuplot 1 .
.Pp
.Dl tcpstat -b 28800 -r file.dump 0.5
.Pp
Displays what percentage of the traffic in file.dump exceeded the speed of
my modem (28800 bits per second.)
.Pp
.Sh SEE ALSO
.Xr tcpdump 1 ,
.Xr pcap 3 ,
.Xr bpf 4 ,
.Xr printf 3
.Sh NOTES
.Ss Interval size affects bandwidth
Due to the nature of how bandwidth is actually measured (from discrete
samples of data), the bandwidth numbers displayed will vary according to the
.Ar interval
variable.  Generally speaking, if you often have rapid bursts of packet data,
the bandwidth reported will not reflect this when
.Ar interval
is sufficiently large.  This results in an "averaging" effect, which may or
may not be desired.  On the other hand, if
.Ar interval
is too small (say < 0.01), this results in unrealistically large
bandwidths for very short amounts of time.
.Pp
The reason for the latter is that most network interfaces do not hand over
packets bit by bit, but rather packet by packet.  Thus, each packet is
reported as being transferred "instantaneously", resulting in "infinite" (or
rather indeterminable) bandwidth.  Thus, when counting single bits on the
wire, there really is no such thing as "bandwidth" because they aren't
really moving from the network stack's point of view (cf. Zeno's Paradox.)
.Pp
A possible solution is to internally spline the packet sizes together and
report the bandwidth as the scalar integral over the given interval, but
this has yet to be implemented, and to be honest, would be the proverbial
cruise missile to destroy an ant hill.
.Pp
That being said (whew!), a "good value" for
.Ar interval
is usually somewhere between 0.5 and 2.
.Ss Difference between normal and 'dumb' bandwidth modes.
In normal bandwidth mode, when an interval exceeds the given bandwidth,
the extra bytes are "moved" into the next interval.  This has the effect
of trying to imagine how overloaded an interface would be if the interface
had a smaller bandwidth, yet same amount of data tried to get through.
.Pp
In "dumb" bandwidth mode, each interval which exceeds the given bandwidth
is simply counted.  Nothin' else.
.Sh HISTORY
.Nm
was first written in Winter 1998 using
.Fx 3.0 ,
and then finally retrofitted for Linux in Spring 2000.
.Sh AUTHORS
.An Paul Herman Aq pherman@frenchfries.net
.br
Cologne, Germany.
.Pp
Please send all bug reports to this address.
.Sh BUGS
Due to a bug in libpcap, tcpstat will hang indefinitely under Linux 
when no packets arrive.  This is because the timeout in pcap_open_live()
is ignored under Linux when the interface is idle, which causes pcap_dispatch()
to never return.
.Pp
Not tested with link types other than Ethernet, PPP, and "None" types.  
.Pp
There may be problems reading non-IPv4 packets across platforms when
reading null type link layers.  This is due to a lack of a standardized
packet type descriptor in libpcap for this link type.
.Pp
Snoop file formats cannot be read from stdin or named pipes.