[obnox/wireshark/wip.git] / doc / tethereal.pod.template

=head1 NAME

tethereal - Dump and analyze network traffic

=head1 SYNOPSYS

B<tethereal>
S<[ B<-a> capture autostop condition ] ...>
S<[ B<-b> number of ring buffer files [:duration] ]>
S<[ B<-c> count ]>
S<[ B<-d> <layer type>==<selector>,<decode-as protocol> ]>
S<[ B<-D> ]>
S<[ B<-f> capture filter expression ]>
S<[ B<-F> file format ]>
S<[ B<-h> ]>
S<[ B<-i> interface ]> 
S<[ B<-l> ]>
S<[ B<-n> ]>
S<[ B<-N> resolving flags ]>
S<[ B<-o> preference setting ] ...>
S<[ B<-p> ]>
S<[ B<-q> ]>
S<[ B<-r> infile ]>
S<[ B<-R> display filter expression ]>
S<[ B<-s> snaplen ]>
S<[ B<-S> ]>
S<[ B<-t> time stamp format ]>
S<[ B<-v> ]>
S<[ B<-V> ]>
S<[ B<-w> savefile ]>
S<[ B<-x> ]>
S<[ B<-z> statistics-string ]>
S<[ filter expression ]>

=head1 DESCRIPTION

B<Tethereal> is a network protocol analyzer.  It lets you capture packet
data from a live network, or read packets from a previously saved
capture file, either printing a decoded form of those packets to the
standard output or writing the packets to a file.  B<Tethereal>'s native
capture file format is B<libpcap> format, which is also the format used
by B<tcpdump> and various other tools.  In addition, B<Tethereal> can
read capture files from B<snoop> and B<atmsnoop>, Shomiti/Finisar
B<Surveyor>, Novell B<LANalyzer>, Network General/Network Associates
DOS-based B<Sniffer> (compressed or uncompressed), Microsoft B<Network
Monitor>, AIX's B<iptrace>, Cinco Networks B<NetXRay>, Network
Associates Windows-based B<Sniffer>, AG Group/WildPackets
B<EtherPeek>/B<TokenPeek>/B<AiroPeek>, B<RADCOM>'s WAN/LAN analyzer,
B<Lucent/Ascend> router debug output, HP-UX's B<nettl>, the dump output
from B<Toshiba's> ISDN routers, the output from B<i4btrace> from the
ISDN4BSD project, the output in B<IPLog> format from the Cisco Secure
Intrusion Detection System, B<pppd logs> (pppdump format), the output
from VMS's B<TCPIPtrace>/B<TCPtrace>/B<UCX$TRACE> utilities, the text
output from the B<DBS Etherwatch> VMS utility, traffic capture files
from Visual Networks' Visual UpTime, the output from B<CoSine> L2 debug,
and the output from Accellent's 5Views LAN agents.  There is no need to
tell B<Tethereal> what type of file you are reading; it will determine
the file type by itself.  B<Tethereal> is also capable of reading any of
these file formats if they are compressed using gzip.  B<Tethereal>
recognizes this directly from the file; the '.gz' extension is not
required for this purpose.

If the B<-w> flag is not specified, B<Tethereal> prints a decoded form
of the packets it captures or reads; otherwise, it writes those packets
to the file specified by that flag.

When printing a decoded form of packets, B<Tethereal> prints, by
default, a summary line containing the fields specified by the
preferences file (which are also the fields displayed in the packet list
pane in B<Ethereal>), although if it's printing packets as it captures
them, rather than printing packets from a saved capture file, it won't
print the "frame number" field.  If the B<-V> flag is specified, it
prints intead a protocol tree, showing all the fields of all protocols
in the packet.

When writing packets to a file, B<Tethereal>, by default, writes the
file in B<libpcap> format, and writes all of the packets it sees to the
output file.  The B<-F> flag can be used to specify the format in which
to write the file.  The following formats are supported:

=over 8

=item B<libpcap> - libpcap (tcpdump, Ethereal, etc.)

=item B<rh6_1libpcap> - Red Hat Linux 6.1 libpcap (tcpdump)

=item B<suse6_3libpcap> - SuSE Linux 6.3 libpcap (tcpdump)

=item B<modlibpcap> - modified libpcap (tcpdump)

=item B<nokialibpcap> - Nokia libpcap (tcpdump)

=item B<lanalyzer> - Novell LANalyzer

=item B<ngsniffer> - Network Associates Sniffer (DOS-based)

=item B<snoop> - Sun snoop

=item B<netmon1> - Microsoft Network Monitor 1.x

=item B<netmon2> - Microsoft Network Monitor 2.x

=item B<ngwsniffer_1_1> - Network Associates Sniffer (Windows-based) 1.1

=item B<ngwsniffer_2_0> - Network Associates Sniffer (Windows-based) 2.00x

=item B<visual> - Visual Networks traffic capture

=back

This list is also displayed by the B<-h> flag.

Read filters in B<Tethereal>, which allow you to select which packets
are to be decoded or written to a file, are very powerful; more fields
are filterable in B<Tethereal> than in other protocol analyzers, and the
syntax you can use to create your filters is richer.  As B<Tethereal>
progresses, expect more and more protocol fields to be allowed in read
filters.

Packet capturing is performed with the pcap library.  The capture filter
syntax follows the rules of the pcap library.  This syntax is different
from the read filter syntax.  A read filter can also be specified when
capturing, and only packets that pass the read filter will be displayed
or saved to the output file; note, however, that capture filters are much
more efficient than read filters, and it may be more difficult for
B<Tethereal> to keep up with a busy network if a read filter is
specified for a live capture.

Compressed file support uses (and therefore requires) the zlib library. 
If the zlib library is not present, B<Tethereal> will compile, but will
be unable to read compressed files.

A capture or read filter can either be specified with the B<-f> or B<-R>
option, respectively, in which case the entire filter expression must be
specified as a single argument (which means that if it contains spaces,
it must be quoted), or can be specified with command-line arguments
after the option arguments, in which case all the arguments after the
filter arguments are treated as a filter expression.  If the filter is
specified with command-line arguments after the option arguments, it's a
capture filter if a capture is being done (i.e., if no B<-r> flag was
specified) and a read filter if a capture file is being read (i.e., if a
B<-r> flag was specified).

=head1 OPTIONS

=over 4

=item -a

Specify a criterion that specifies when B<Tethereal> is to stop writing
to a capture file.  The criterion is of the form I<test>B<:>I<value>,
where I<test> is one of:

=for man .RS

=for html <P><DL>

=item duration

Stop writing to a capture file after I<value> seconds have elapsed.

=item filesize

Stop writing to a capture file after it reaches a size of I<value>
kilobytes (where a kilobyte is 1000 bytes, not 1024 bytes).

=for man .RE

=for html </DL>

=item -b

If a maximum capture file size was specified, cause B<Tethereal> to run
in "ring buffer" mode, with the specified number of files.  In "ring
buffer" mode, B<Tethereal> will write to several capture files. 
Their name is based on the number of the file and on the creation date 
and time.

When the first capture file fills up, B<Tethereal> will switch to writing
to the next file, until it fills up the last file, at which point it'll
discard the data in the first file (unless 0 is specified, in which case,
the number of files is unlimited) and start writing to that file and so on.

If the optional duration is specified, B<Tethereal> will switch also 
to the next file when the specified number of seconds has elapsed even
if the current file is not completely fills up.

You can only save files in B<libpcap> format when using a ring buffer.

=item -c

Set the default number of packets to read when capturing live
data.

=item -d

Specify that if the layer type in question (for example, B<tcp.port> or
B<udp.port> for a TCP or UDP port number) has the specified selector
value, packets should be dissected as the specified protocol.

=item -D

Print a list of the interfaces on which B<Tethereal> can capture, and
exit.  Note that "can capture" means that B<Tethereal> was able to open
that device to do a live capture; if, on your system, a program doing a
network capture must be run from an account with special privileges (for
example, as root), then, if B<Tethereal> is run with the B<-D> flag and
is not run from such an account, it will not list any interfaces.

=item -f

Set the capture filter expression.

=item -F

Set the file format of the output capture file.

=item -h

Print the version and options and exits.

=item -i

Set the name of the network interface or pipe to use for live packet
capture. 

Network interface names should match one of the names listed in
"B<tethereal -D>" (described above).  If you're using Unix, "B<netstat
-i>" or "B<ifconfig -a>" might also work to list interface names,
although not all versions of Unix support the B<-a> flag to B<ifconfig>.
If no interface is specified, B<Tethereal> searches the list of
interfaces, choosing the first non-loopback interface if there are any
non-loopback interfaces, and choosing the first loopback interface if
there are no non-loopback interfaces; if there are no interfaces,
B<Tethereal> reports an error and doesn't start the capture.

Pipe names should be either the name of a FIFO (named pipe) or ``-'' to
read data from the standard input.  Data read from pipes must be in
standard libpcap format.

=item -l

Flush the standard output after the information for each packet is
printed.  (This is not, strictly speaking, line-buffered if B<-V>
was specified; however, it is the same as line-buffered if B<-V> wasn't
specified, as only one line is printed for each packet, and, as B<-l> is
normally used when piping a live capture to a program or script, so that
output for a packet shows up as soon as the packet is seen and
dissected, it should work just as well as true line-buffering.  We do
this as a workaround for a deficiency in the Microsoft Visual C++ C
library.)

This may be useful when piping the output of B<Tethereal> to another
program, as it means that the program to which the output is piped will
see the dissected data for a packet as soon as B<Tethereal> sees the
packet and generates that output, rather than seeing it only when the
standard output buffer containing that data fills up.

=item -n

Disable network object name resolution (such as hostname, TCP and UDP port
names).

=item -N

Turn on name resolving for particular types of addresses and port
numbers, with name resolving for other types of addresses and port
numbers turned off; the argument is a string that may contain the
letters B<m> to enable MAC address resolution, B<n> to enable network
address resolution, and B<t> to enable transport-layer port number
resolution.  This overrides B<-n> if both B<-N> and B<-n> are present.
The letter B<C> enables concurrent (asynchronous) DNS lookups.

=item -o

Set a preference value, overriding the default value and any value read
from a preference file.  The argument to the flag is a string of the
form I<prefname>B<:>I<value>, where I<prefname> is the name of the
preference (which is the same name that would appear in the preference
file), and I<value> is the value to which it should be set.

=item -p

I<Don't> put the interface into promiscuous mode.  Note that the
interface might be in promiscuous mode for some other reason; hence,
B<-p> cannot be used to ensure that the only traffic that is captured is
traffic sent to or from the machine on which B<Tethereal> is running,
broadcast traffic, and multicast traffic to addresses received by that
machine.

=item -q

Don't display the continuous count of packets captured that is normally
shown when saving a capture to a file; instead, just display, at the end
of the capture, a count of packets captured.  On systems that support
the SIGINFO signal, such as various BSDs, typing your "status" character
(typically control-T) will cause the current count to be displayed.

=item -r

Read packet data from I<infile>.

=item -R

Cause the specified filter (which uses the syntax of read filters,
rather than that of capture filters) to be applied before printing a
decoded form of packets or writing packets to a file; packets not
matching the filter are discarded rather than being printed or written.

=item -s

Set the default snapshot length to use when capturing live data. 
No more than I<snaplen> bytes of each network packet will be read into
memory, or saved to disk.

=item -S

Decode and display packets even while writing to file.

=item -t

Set the format of the packet timestamp printed in summary lines.  The
format can be one of 'r' (relative), 'a' (absolute), 'ad' (absolute with
date), or 'd' (delta).  The relative time is the time elapsed between
the first packet and the current packet.  The absolute time is the
actual time the packet was captured, with no date displayed; the
absolute date and time is the actual time and date the packet was
captured.  The delta time is the time since the previous packet was
captured.  The default is relative.

=item -v

Print the version and exit.

=item -V

Cause B<Tethereal> to print a protocol tree for each packet rather than
a one-line summary of the packet.

=item -w

Write packet data to I<savefile> or to the standard output if
I<savefile> is "-".

=item -x

Cause B<Tethereal> to print a hex and ASCII dump of the packet data
after printing the summary or protocol tree.

=item -z

Get B<Tethereal> to collect various types of statistics and display the result
after finishing reading the capture file.
Currently implemented statistics are:

B<-z> dcerpc,rtt,I<uuid>,I<major>.I<minor>[,I<filter>]

Collect call/reply RTT data for DCERPC interface I<uuid>, 
version I<major>.I<minor>.
Data collected is number of calls for each procedure, MinRTT, MaxRTT 
and AvgRTT. 
Example: use B<-z dcerpc,rtt,12345778-1234-abcd-ef00-0123456789ac,1.0> to collect data for CIFS SAMR Interface.  
This option can be used multiple times on the command line. 

If the optional filterstring is provided, the stats will only be calculated
on those calls that match that filter.
Example: use B<-z dcerpc,rtt,12345778-1234-abcd-ef00-0123456789ac,1.0,ip.addr==1.2.3.4> to collect SAMR
RTT statistics for a specific host.


B<-z> io,phs[,I<filter>]

Create Protocol Hierarchy Statistics listing both number of frames and bytes.
If no I<filter> is specified the statistics will be calculated for all frames.
If a I<filters> is specified statistics will be only calculated for those
packets that match the filter.

This option can be used multiple times on the command line. 


B<-z> io,stat,I<interval>[,I<filter>][,I<filter>][,I<filter>]...

Collect frame/bytes statistics for the capture in intervals of I<interval> 
seconds. I<Intervals> can be specified either as whole or fractional seconds.
Interval can be specified in ms resolution.

If no I<filter> is specified the statistics will be calculated for all frames.
If one or more I<filters> are specified statistics will be calculated for
all filters and presented with one column of statistics for each filter.

This option can be used multiple times on the command line. 


Example: B<-z io,stat,1,ip.addr==1.2.3.4> to generate 1 second
statistics for all traffic to/from host 1.2.3.4.

Example: B<-z "io,stat,0.001,smb&&ip.addr==1.2.3.4"> to generate 1ms
statistics for all SMB frames to/from host 1.2.3.4.

The examples above all use the standard syntax for generating statistics
which only calculates the number of frames and bytes in each interval.


io,stat can also do much more statistics and calculate COUNT() SUM() MIN() 
MAX() and AVG() using a slightly filter syntax:
  [COUNT|SUM|MIN|MAX|AVG](<field>)<filter>
One important thing to note here is that the field that the calculation is 
based on MUST also be part of the filter string or else the calculation will
fail.

So: B<-z io,stat,0.010,AVG(smb.time)> does not work.  Use B<-z
io,stat,0.010,AVG(smb.time)smb.time> instead.  Also be aware that a field
can exist multiple times inside the same packet and will then be counted
multiple times in those packets. 


COUNT(<field>) can be used on any type which has a display filter name. 
It will count how many times this particular field is encountered in the
filtered packet list.

Example: B<-z io,stat,0.010,COUNT(smb.sid)smb.sid>
This will count the total number of SIDs seen in each 10ms interval.

SUM(<field>) can only be used on named fields of integer type.
This will sum together every occurence of this fields value for each interval.

Example: B<-z io,stat,0.010,SUM(frame.pkt_len)frame.pkt_len>
This will report the total number of bytes seen in all the frames within
an interval.

MIN/MAX/AVG(<field>) can only be used on named fields that are either
integers or relative time fields.  This will calculate maximum/minimum
or average seen in each interval.  If the field is a relative time field
the output will be presented in seconds and three digits after the
decimal point.  The resolution for time calculations is 1ms and anything
smaller will be truncated.

Example:  B<-z "io,stat,0.010,smb.time&&ip.addr==1.1.1.1,MIN(smb.time)smb.time&&ip.addr==1.1.1.1,MAX(smb.time)smb.time&&ip.addr==1.1.1.1,MAX(smb.time)smb.time&&ip.addr==1.1.1.1">

This will calculate statistics for all smb response times we see to/from
host 1.1.1.1 in 10ms intervals.  The output will be displayed in 4
columns; number of frames/bytes, minimum response time, maximum response
time and average response time.



B<-z> talkers,I<type>[,I<filter>]

Create a table that lists all conversations that could be seen in the capture.
I<type> specifies which type of conversation we want to generate the 
statistics for, currently the supported ones are
  "eth"   Ethernet
  "fc"    Fibre Channel
  "ip"    IP addresses
  "ipx"   IPX addresses
  "tcp"   TCP/IP socketpairs   Both IPv4 and IPv6 are supported
  "tr"    TokenRing
  "udp"   UDP/IP socketpairs   Both IPv4 and IPv6 are supported

If the optional filter string is specified, only those packets that match the
filter will be used in the calculations.

The table is presented with one line for each conversation and displays
number of frames/bytes in each direction as well as total number of 
frames/bytes.
The table is sorted according to total number of bytes.


B<-z> proto,colinfo,I<filter>,I<field>

Append all I<field> values for the packet to the COL_INFO information line.
This feature can be used to append arbitrary fields to the COL_INFO line
in addition to the normal content of the COL_INFO line.
I<field> is the display-filter name of a field which value should be placed
on the COL_INFO line.
I<filter> is a filter string that controls for which packets the field value
will be presented on COL_INFO line. I<field> will only be presented on the
COL_INFO line for the packets which match I<filter>.

NOTE: In order for B<Tethereal> to be able to extract the I<field> value
from the packet, I<field> MUST be part of the I<filter> string.  If not,
B<Tethereal> will not be able to extract its value.

For a simple example to add the "nfs.fh.hash" field to COL_INFO for all
packets containing the "nfs.fh.hash" field, use

B<-z proto,colinfo,nfs.fh.hash,nfs.fh.hash>


To put "nfs.fh.hash" on COL_INFO but only for packets coming from host 1.2.3.4
use :

B<-z "proto,colinfo,nfs.fh.hash && ip.src==1.2.3.4,nfs.fh.hash">

This option can be used multiple times on the command line. 


B<-z> rpc,rtt,I<program>,I<version>[,I<filter>]

Collect call/reply RTT data for I<program>/I<version>.  Data collected
is number of calls for each procedure, MinRTT, MaxRTT and AvgRTT. 
Example: use B<-z rpc,rtt,100003,3> to collect data for NFS v3.  This
option can be used multiple times on the command line. 

If the optional filterstring is provided, the stats will only be calculated
on those calls that match that filter.
Example: use B<-z rpc,rtt,100003,3,nfs.fh.hash==0x12345678> to collect NFS v3
RTT statistics for a specific file.


B<-z> rpc,programs

Collect call/reply RTT data for all known ONC-RPC programs/versions.  
Data collected is number of calls for each protocol/version, MinRTT, 
MaxRTT and AvgRTT. 
This option can only be used once on the command line.

B<-z> smb,rtt[,I<filter>]

Collect call/reply RTT data for SMB.  Data collected
is number of calls for each SMB command, MinRTT, MaxRTT and AvgRTT. 
Example: use B<-z smb,rtt>.
The data will be presented as separate tables for all normal SMB commands,
all Transaction2 commands and all NT Transaction commands.
Only those commands that are seen in the capture will have its stats
displayed.
Only the first command in a xAndX command chain will be used in the
calculation.  So for common SessionSetupAndX + TreeConnectAndX chains,
only the SessionSetupAndX call will be used in the statistics.
This is a flaw that might be fixed in the future.

This option can be used multiple times on the command line. 

If the optional filterstring is provided, the stats will only be calculated
on those calls that match that filter.
Example: use B<-z "smb,rtt,ip.addr==1.2.3.4"> to only collect stats for
SMB packets echanged by the host at IP address 1.2.3.4 .

B<-z> smb,sids

When this feature is used B<Tethereal> will print a report with all the
discovered SID and account name mappings.  Only those SIDs where the
account name is known will be presented in the table.

For this feature to work you will need to either to enable
"Edit/Preferences/Protocols/SMB/Snoop SID to name mappings" in the
preferences or you can override the preferences by specifying
B<-o "smb.sid_name_snooping:TRUE"> on the B<Tethereal> command line.

The current methods used by B<Tethereal> to find the SID->name mapping
is relatively restricted but is hoped to be expanded in the future.

B<-z> mgcp,rtd[I<,filter>]

Collect requests/response RTD (Response Time Delay) data for MGCP. 
This is similar to B<-z smb,rtt>). Data collected is number of calls
for each known MGCP Type, MinRTD, MaxRTD and AvgRTD.
Additionally you get the number of duplicate requests/responses, 
unresponded requests, responses ,which don't match with
any request. 
Example: use B<-z mgcp,rtd>.

This option can be used multiple times on the command line. 

If the optional filterstring is provided, the stats will only be calculated
on those calls that match that filter.
Example: use B<-z "mgcp,rtd,ip.addr==1.2.3.4"> to only collect stats for
MGCP packets exchanged by the host at IP address 1.2.3.4 .

=back

=head1 CAPTURE FILTER SYNTAX

See manual page of tcpdump(8).

=head1 READ FILTER SYNTAX

Read filters help you remove the noise from a packet trace and let you
see only the packets that interest you.  If a packet meets the
requirements expressed in your read filter, then it is printed.  Read
filters let you compare the fields within a protocol against a specific
value, compare fields against fields, and to check the existence of
specified fields or protocols.

The simplest read filter allows you to check for the existence of a
protocol or field.  If you want to see all packets which contain the IPX
protocol, the filter would be "ipx".  (Without the quotation marks) To
see all packets that contain a Token-Ring RIF field, use "tr.rif".

Fields can also be compared against values.  The comparison operators
can be expressed either through C-like symbols, or through English-like
abbreviations:

    eq, ==    Equal
    ne, !=    Not equal
    gt, >     Greater than
    lt, <     Less Than
    ge, >=    Greater than or Equal to
    le, <=    Less than or Equal to

Furthermore, each protocol field is typed. The types are:

    Unsigned integer (either 8-bit, 16-bit, 24-bit, or 32-bit)
    Signed integer (either 8-bit, 16-bit, 24-bit, or 32-bit)
    Boolean
    Ethernet address (6 bytes)
    Byte string (n-number of bytes)
    IPv4 address
    IPv6 address
    IPX network number
    String (text)
    Double-precision floating point number

An integer may be expressed in decimal, octal, or hexadecimal notation. 
The following three read filters are equivalent:

    frame.pkt_len > 10
    frame.pkt_len > 012
    frame.pkt_len > 0xa

Boolean values are either true or false.  In a read filter expression
testing the value of a Boolean field, "true" is expressed as 1 or any
other non-zero value, and "false" is expressed as zero.  For example, a
token-ring packet's source route field is boolean.  To find any
source-routed packets, a read filter would be:

    tr.sr == 1

Non source-routed packets can be found with:

    tr.sr == 0

Ethernet addresses, as well as a string of bytes, are represented in hex
digits.  The hex digits may be separated by colons, periods, or hyphens:

    fddi.dst eq ff:ff:ff:ff:ff:ff
    ipx.srcnode == 0.0.0.0.0.1
    eth.src == aa-aa-aa-aa-aa-aa

If a string of bytes contains only one byte, then it is represented as
an unsigned integer.  That is, if you are testing for hex value 'ff' in
a one-byte byte-string, you must compare it agains '0xff' and not 'ff'. 

IPv4 addresses can be represented in either dotted decimal notation, or
by using the hostname:

    ip.dst eq www.mit.edu
    ip.src == 192.168.1.1

IPv4 addresses can be compared with the same logical relations as numbers:
eq, ne, gt, ge, lt, and le.  The IPv4 address is stored in host order,
so you do not have to worry about how the endianness of an IPv4 address
when using it in a read filter.

Classless InterDomain Routing (CIDR) notation can be used to test if an
IPv4 address is in a certain subnet.  For example, this display filter
will find all packets in the 129.111 Class-B network:

    ip.addr == 129.111.0.0/16

Remember, the number after the slash represents the number of bits used
to represent the network.  CIDR notation can also be used with
hostnames, in this example of finding IP addresses on the same Class C
network as 'sneezy':

    ip.addr eq sneezy/24

The CIDR notation can only be used on IP addresses or hostnames, not in
variable names.  So, a display filter like "ip.src/24 == ip.dst/24" is
not valid.  (yet)

IPX networks are represented by unsigned 32-bit integers.  Most likely
you will be using hexadecimal when testing for IPX network values:

    ipx.srcnet == 0xc0a82c00

Strings are enclosed in double-quotes:

    http.request.method == "POST"

Inside double quotes, you may use the backslash to embed a double-quote,
or an arbitrary byte represented in either octal or hexadecimal.

    browser.comment = "An embedded \" double-quote"

Use of hexadecimal to look for "HEAD":

    http.request.method == "\x48EAD"

Use of octal to look for "HEAD":

    http.request.method == "\x110EAD"

This means that you must escape backslashes with backslashes inside
double quotes:

    smb.path == "\\\\SERVER\\SHARE"

to look for \\SERVER\SHARE in "smb.path".

A slice operator also exists.  You can check the substring
(byte-string) of any protocol or field.  For example, you can filter on
the vendor portion of an ethernet address (the first three bytes) like
this:

    eth.src[0:3] == 00:00:83

If the length of your byte-slice is only one byte, then it is still
represented in hex, but without the preceding "0x": 

    llc[3] == aa

You can use the slice operator on a protocol name, too.  And
remember, the "frame" protocol encompasses the entire packet, allowing
you to look at the nth byte of a packet regardless of its frame type
(Ethernet, token-ring, etc.).

    token[0:5] ne 0.0.0.1.1
    ipx[0:2] == ff:ff
    llc[3:1] eq 0xaa

The following syntax governs slices:

        [i:j]   i = start_offset, j = length
        [i-j]   i = start_offset, j = end_offset, inclusive.
        [i]     i = start_offset, length = 1
        [:j]    start_offset = 0, length = j
        [i:]    start_offset = i, end_offset = end_of_field

Offsets and lengths can be negative, in which case they indicate the
offset from the B<end> of the field.  Here's how to check the last 4
bytes of a frame:

    frame[-4:4] == 0.1.2.3

or

    frame[-4:] == 0.1.2.3

You can create complex concatenations of slices using the comma operator:

        field[1,3-5,9:] == 01:03:04:05:09:0a:0b

All the above tests can be combined together with logical expressions. 
These too are expressable in C-like syntax or with English-like
abbreviations:

    and, &&   Logical AND
    or, ||    Logical OR
    not, !    Logical NOT

Expressions can be grouped by parentheses as well.  The following are
all valid read filter expression:

    tcp.port == 80 and ip.src == 192.168.2.1
    not llc
    (ipx.srcnet == 0xbad && ipx.srnode == 0.0.0.0.0.1) || ip
    tr.dst[0:3] == 0.6.29 xor tr.src[0:3] == 0.6.29

A special caveat must be given regarding fields that occur more than
once per packet.  "ip.addr" occurs twice per IP packet, once for the
source address, and once for the destination address.  Likewise,
tr.rif.ring fields can occur more than once per packet.  The following
two expressions are not equivalent:

        ip.addr ne 192.168.4.1
    not ip.addr eq 192.168.4.1

The first filter says "show me IP packets where an ip.addr exists that
does not equal 192.168.4.1".  That is, as long as one ip.addr in the
packet does not equal 192.168.44.1, the packet passes the read
filter.  The second filter "don't show me any packets that have at least
one ip.addr field equal to 192.168.4.1".  If one ip.addr is 192.168.4.1,
the packet does not pass.  If B<neither> ip.addr fields is 192.168.4.1,
then the packet passes.

It is easy to think of the 'ne' and 'eq' operators as having an implict
"exists" modifier when dealing with multiply-recurring fields.  "ip.addr
ne 192.168.4.1" can be thought of as "there exists an ip.addr that does
not equal 192.168.4.1".

Be careful with multiply-recurring fields; they can be confusing.

Care must also be taken when using the read filter to remove noise
from the packet trace. If you want to e.g. filter out all IP multicast
packets to address 224.1.2.3, then using:

    ip.dst ne 224.1.2.3

may be too restrictive. Filtering with "ip.dst" selects only those
B<IP> packets that satisfy the rule. Any other packets, including all
non-IP packets, will not be printed. For printing also the non-IP
packets, you can use one of the following two expressions:

    not ip or ip.dst ne 224.1.2.3
    not ip.addr eq 224.1.2.3

The first filter uses "not ip" to include all non-IP packets and then
lets "ip.dst ne 224.1.2.3" to filter out the unwanted IP packets. The
second filter has already been explained above where filtering with
multiply occuring fields was discussed.

The following is a table of protocol and protocol fields that are
filterable in B<Tethereal>.  The abbreviation of the protocol or field is
given.  This abbreviation is what you use in the read filter.  The
type of the field is also given.

=insert_dfilter_table

=head1 FILES

The F<ethereal.conf> file, which is installed in the F<etc> directory
under the main installation directory (for example, F</usr/local/etc>)
on UNIX-compatible systems, and in the main installation directory (for
example, F<C:\Program Files\Ethereal>) on Windows systems, and the
personal preferences file, which is F<$HOME/.ethereal/preferences> on
UNIX-compatible systems and F<%APPDATA%\Ethereal\preferences> (or, if
%APPDATA% isn't defined,
F<%USERPROFILE%\Application Data\Ethereal\preferences>) on
Windows systems, contain system-wide and personal preference settings,
respectively.  The file contains preference settings of the form
I<prefname>B<:>I<value>, one per line, where I<prefname> is the name of
the preference (which is the same name that would appear in the
preference file), and I<value> is the value to which it should be set;
white space is allowed between B<:> and I<value>.  A preference setting
can be continued on subsequent lines by indenting the continuation lines
with white space.  A B<#> character starts a comment that runs to the
end of the line.

The system-wide preference file is read first, if it exists, overriding
B<Tethereal>'s default values; the personal preferences file is then
read, if it exists, overriding default values and values read from the
system-wide preference file.

The disabled protocols file, which is F<$HOME/.ethereal/disabled_protos>
on UNIX-compatible systems and F<%APPDATA%\Ethereal\disabled_protos>
(or, if %APPDATA% isn't defined, F<%USERPROFILE%\Application
Data\Ethereal\disabled_protos>) on Windows systems, contain a list of
protocols that have been disabled, so that their dissectors are never
called.  The file contains protocol names, one per line, where the
protocol name is the same name that would be used in a display filter
for the protocol.  A B<#> character starts a comment that runs to the
end of the line.

The F<ethers> file, which is found in the F</etc> directory on
UNIX-compatible systems, and in the main installation directory (for
example, F<C:\Program Files\Ethereal>) on Windows systems, is consulted
to correlate 6-byte hardware addresses to names.  If an address is not
found in the F<ethers> file, the F<$HOME/.ethereal/ethers> file on
UNIX-compatible systems, and the F<%APPDATA%\Ethereal\ethers> file (or, if
%APPDATA% isn't defined, the
F<%USERPROFILE%\Application Data\Ethereal\ethers> file) on Windows
systems is consulted next.  Each line contains one hardware
address and name, separated by whitespace.  The digits of the hardware
address are separated by either a colon (:), a dash (-), or a period
(.).  The following three lines are valid lines of an ethers file:

  ff:ff:ff:ff:ff:ff          Broadcast
  c0-00-ff-ff-ff-ff          TR_broadcast
  00.00.00.00.00.00          Zero_broadcast

The F<manuf> file, which is installed in the F<etc> directory under the
main installation directory (for example, F</usr/local/etc>) on
UNIX-compatible systems, and in the main installation directory (for
example, F<C:\Program Files\Ethereal>) on Windows systems, matches the
3-byte vendor portion of a 6-byte hardware address with the
manufacturer's name; it can also contain well-known MAC addresses and
address ranges specified with a netmask.  The format of the file is the
same as the F<ethers> file, except that entries of the form

  00:00:0C      Cisco

can be provided, with the 3-byte OUI and the name for a vendor, and
entries of the form

  00-00-0C-07-AC/40     All-HSRP-routers

can be specified, with a MAC address and a mask indicating how many bits
of the address must match.  Trailing zero bytes can be omitted from
address ranges.  That entry, for example, will match addresses from
00-00-0C-07-AC-00 through 00-00-0C-07-AC-FF.  The mask need not be a
multiple of 8.

The F<ipxnets> file, which is found in the F</etc> directory on
UNIX-compatible systems, and in the main installation directory (for
example, F<C:\Program Files\Ethereal>) on Windows systems, correlates
4-byte IPX network numbers to names.  If a network number is not found
in the F<ipxnets> file, the F<$HOME/.ethereal/ipxnets> file on
UNIX-compatible systems, and the F<%APPDATA%\Ethereal\ipxnets> file (or,
if %APPDATA% isn't defined, the
F<%USERPROFILE%\Application Data\Ethereal\ipxnets> file)
on Windows systems, is consulted next.  The format is the same as the
F<ethers> file, except that each address if four bytes instead of six. 
Additionally, the address can be represented a single hexadecimal
number, as is more common in the IPX world, rather than four hex octets. 
For example, these four lines are valid lines of an ipxnets file.

  C0.A8.2C.00              HR
  c0-a8-1c-00              CEO
  00:00:BE:EF              IT_Server1
  110f                     FileServer3

=head1 SEE ALSO

I<ethereal(1)>, I<editcap(1)>, I<tcpdump(8)>, I<pcap(3)>

=head1 NOTES

B<Tethereal> is part of the B<Ethereal> distribution.  The latest version
of B<Ethereal> can be found at B<http://www.ethereal.com>.

=head1 AUTHORS

B<Tethereal> uses the same packet dissection code that B<Ethereal> does,
as well as using many other modules from B<Ethereal>; see the list of
authors in the B<Ethereal> man page for a list of authors of that code.