1 .\" $NetBSD: tcpdump.8,v 1.9 2003/03/31 00:18:17 perry Exp $
3 .\" Copyright (c) 1987, 1988, 1989, 1990, 1991, 1992, 1994, 1995, 1996, 1997
4 .\" The Regents of the University of California. All rights reserved.
5 .\" All rights reserved.
7 .\" Redistribution and use in source and binary forms, with or without
8 .\" modification, are permitted provided that: (1) source code distributions
9 .\" retain the above copyright notice and this paragraph in its entirety, (2)
10 .\" distributions including binary code include the above copyright notice and
11 .\" this paragraph in its entirety in the documentation or other materials
12 .\" provided with the distribution, and (3) all advertising materials mentioning
13 .\" features or use of this software display the following acknowledgement:
14 .\" ``This product includes software developed by the University of California,
15 .\" Lawrence Berkeley Laboratory and its contributors.'' Neither the name of
16 .\" the University nor the names of its contributors may be used to endorse
17 .\" or promote products derived from this software without specific prior
18 .\" written permission.
19 .\" THIS SOFTWARE IS PROVIDED ``AS IS'' AND WITHOUT ANY EXPRESS OR IMPLIED
20 .\" WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF
21 .\" MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
23 .TH TCPDUMP 1 "1 July 2003"
25 tcpdump \- dump traffic on a network
30 .B \-AdDeflLnNOpqRStuUvxX
76 .I spi@ipaddr algo:secret,...
92 \fITcpdump\fP prints out the headers of packets on a network interface
93 that match the boolean \fIexpression\fP. It can also be run with the
95 flag, which causes it to save the packet data to a file for later
96 analysis, and/or with the
98 flag, which causes it to read from a saved packet file rather than to
99 read packets from a network interface. In all cases, only packets that
106 will, if not run with the
108 flag, continue capturing packets until it is interrupted by a SIGINT
109 signal (generated, for example, by typing your interrupt character,
110 typically control-C) or a SIGTERM signal (typically generated with the
112 command); if run with the
114 flag, it will capture packets until it is interrupted by a SIGINT or
115 SIGTERM signal or the specified number of packets have been processed.
119 finishes capturing packets, it will report counts of:
121 packets ``received by filter'' (the meaning of this depends on the OS on
124 and possibly on the way the OS was configured - if a filter was
125 specified on the command line, on some OSes it counts packets regardless
126 of whether they were matched by the filter expression, and on other OSes
127 it counts only packets that were matched by the filter expression and
131 packets ``dropped by kernel'' (this is the number of packets that were
132 dropped, due to a lack of buffer space, by the packet capture mechanism
135 is running, if the OS reports that information to applications; if not,
136 it will be reported as 0).
138 On platforms that support the SIGINFO signal, such as most BSDs, it will
139 report those counts when it receives a SIGINFO signal (generated, for
140 example, by typing your ``status'' character, typically control-T) and
141 will continue capturing packets.
143 Reading packets from a network interface may require that you have
146 .B Under SunOS 3.x or 4.x with NIT or BPF:
147 You must have read access to
152 .B Under Solaris with DLPI:
153 You must have read/write access to the network pseudo device, e.g.
155 On at least some versions of Solaris, however, this is not sufficient to
158 to capture in promiscuous mode; on those versions of Solaris, you must
161 must be installed setuid to root, in order to capture in promiscuous
162 mode. Note that, on many (perhaps all) interfaces, if you don't capture
163 in promiscuous mode, you will not see any outgoing packets, so a capture
164 not done in promiscuous mode may not be very useful.
166 .B Under HP-UX with DLPI:
169 must be installed setuid to root.
171 .B Under IRIX with snoop:
174 must be installed setuid to root.
179 must be installed setuid to root (unless your distribution has a kernel
180 that supports capability bits such as CAP_NET_RAW and code to allow
181 those capability bits to be given to particular accounts and to cause
182 those bits to be set on a user's initial processes when they log in, in
183 which case you must have CAP_NET_RAW in order to capture and
184 CAP_NET_ADMIN to enumerate network devices with, for example, the
188 .B Under Ultrix and Digital UNIX/Tru64 UNIX:
189 Any user may capture network traffic with
191 However, no user (not even the super-user) can capture in promiscuous
192 mode on an interface unless the super-user has enabled promiscuous-mode
193 operation on that interface using
195 and no user (not even the super-user) can capture unicast traffic
196 received by or sent by the machine on an interface unless the super-user
197 has enabled copy-all-mode operation on that interface using
201 packet capture on an interface probably requires that either
202 promiscuous-mode or copy-all-mode operation, or both modes of
203 operation, be enabled on that interface.
205 .B Under BSD (this includes Mac OS X):
206 You must have read access to
208 On BSDs with a devfs (this includes Mac OS X), this might involve more
209 than just having somebody with super-user access setting the ownership
210 or permissions on the BPF devices - it might involve configuring devfs
211 to set the ownership or permissions every time the system is booted,
212 if the system even supports that; if it doesn't support that, you might
213 have to find some other way to make that happen at boot time.
215 Reading a saved packet file doesn't require special privileges.
219 Print each packet (minus its link level header) in ASCII. Handy for
223 Exit after receiving \fIcount\fP packets.
226 Before writing a raw packet to a savefile, check whether the file is
227 currently larger than \fIfile_size\fP and, if so, close the current
228 savefile and open a new one. Savefiles after the first savefile will
229 have the name specified with the
231 flag, with a number after it, starting at 2 and continuing upward.
232 The units of \fIfile_size\fP are millions of bytes (1,000,000 bytes,
233 not 1,048,576 bytes).
236 Dump the compiled packet-matching code in a human readable form to
237 standard output and stop.
240 Dump packet-matching code as a
245 Dump packet-matching code as decimal numbers (preceded with a count).
248 Print the list of the network interfaces available on the system and on
251 can capture packets. For each network interface, a number and an
252 interface name, possibly followed by a text description of the
253 interface, is printed. The interface name or the number can be supplied
256 flag to specify an interface on which to capture.
258 This can be useful on systems that don't have a command to list them
259 (e.g., Windows systems, or UNIX systems lacking
260 .BR "ifconfig \-a" );
261 the number can be useful on Windows 2000 and later systems, where the
262 interface name is a somewhat complex string.
266 flag will not be supported if
268 was built with an older version of
271 .B pcap_findalldevs()
275 Print the link-level header on each dump line.
278 Use \fIspi@ipaddr algo:secret\fP for decrypting IPsec ESP packets that
279 are addressed to \fIaddr\fP and contain Security Parameter Index value
280 \fIspi\fP. This combination may be repeated with comma or newline seperation.
282 Note that setting the secret for IPv4 ESP packets is supported at this time.
289 \fBcast128-cbc\fP, or
291 The default is \fBdes-cbc\fP.
292 The ability to decrypt packets is only present if \fItcpdump\fP was compiled
293 with cryptography enabled.
295 \fIsecret\fP is the ASCII text for ESP secret key.
296 If preceeded by 0x, then a hex value will be read.
298 The option assumes RFC2406 ESP, not RFC1827 ESP.
299 The option is only for debugging purposes, and
300 the use of this option with a true `secret' key is discouraged.
301 By presenting IPsec secret key onto command line
302 you make it visible to others, via
306 In addition to the above syntax, the syntax \fIfile name\fP may be used
307 to have tcpdump read the provided file in. The file is opened upon
308 receiving the first ESP packet, so any special permissions that tcpdump
309 may have been given should already have been given up.
312 Print `foreign' IPv4 addresses numerically rather than symbolically
313 (this option is intended to get around serious brain damage in
314 Sun's NIS server \(em usually it hangs forever translating non-local
317 The test for `foreign' IPv4 addresses is done using the IPv4 address and
318 netmask of the interface on which capture is being done. If that
319 address or netmask are not available, available, either because the
320 interface on which capture is being done has no address or netmask or
321 because the capture is being done on the Linux "any" interface, which
322 can capture on more than one interface, this option will not work
326 Use \fIfile\fP as input for the filter expression.
327 An additional expression given on the command line is ignored.
330 Listen on \fIinterface\fP.
331 If unspecified, \fItcpdump\fP searches the system interface list for the
332 lowest numbered, configured up interface (excluding loopback).
333 Ties are broken by choosing the earliest match.
335 On Linux systems with 2.2 or later kernels, an
337 argument of ``any'' can be used to capture packets from all interfaces.
338 Note that captures on the ``any'' device will not be done in promiscuous
343 flag is supported, an interface number as printed by that flag can be
349 Make stdout line buffered.
350 Useful if you want to see the data
354 ``tcpdump\ \ \-l\ \ |\ \ tee dat'' or
355 ``tcpdump\ \ \-l \ \ > dat\ \ &\ \ tail\ \ \-f\ \ dat''.
358 List the known data link types for the interface and exit.
361 Load SMI MIB module definitions from file \fImodule\fR.
363 can be used several times to load several MIB modules into \fItcpdump\fP.
366 Don't convert addresses (i.e., host addresses, port numbers, etc.) to names.
369 Don't print domain name qualification of host names.
371 if you give this flag then \fItcpdump\fP will print ``nic''
372 instead of ``nic.ddn.mil''.
375 Do not run the packet-matching code optimizer.
377 if you suspect a bug in the optimizer.
380 \fIDon't\fP put the interface
381 into promiscuous mode.
382 Note that the interface might be in promiscuous
383 mode for some other reason; hence, `-p' cannot be used as an abbreviation for
384 `ether host {local-hw-addr} or ether broadcast'.
387 Quick (quiet?) output.
388 Print less protocol information so output
392 Assume ESP/AH packets to be based on old specification (RFC1825 to RFC1829).
393 If specified, \fItcpdump\fP will not print replay prevention field.
394 Since there is no protocol version field in ESP/AH specification,
395 \fItcpdump\fP cannot deduce the version of ESP/AH protocol.
398 Read packets from \fIfile\fR (which was created with the
401 Standard input is used if \fIfile\fR is ``-''.
404 Print absolute, rather than relative, TCP sequence numbers.
407 Snarf \fIsnaplen\fP bytes of data from each packet rather than the
408 default of 68 (with SunOS's NIT, the minimum is actually 96).
409 68 bytes is adequate for IP, ICMP, TCP
410 and UDP but may truncate protocol information from name server and NFS
412 Packets truncated because of a limited snapshot
413 are indicated in the output with ``[|\fIproto\fP]'', where \fIproto\fP
414 is the name of the protocol level at which the truncation has occurred.
415 Note that taking larger snapshots both increases
416 the amount of time it takes to process packets and, effectively,
417 decreases the amount of packet buffering.
418 This may cause packets to be
420 You should limit \fIsnaplen\fP to the smallest number that will
421 capture the protocol information you're interested in.
423 \fIsnaplen\fP to 0 means use the required length to catch whole packets.
426 Force packets selected by "\fIexpression\fP" to be interpreted the
427 specified \fItype\fR.
428 Currently known types are
429 \fBaodv\fR (Ad-hoc On-demand Distance Vector protocol),
430 \fBcnfp\fR (Cisco NetFlow protocol),
431 \fBrpc\fR (Remote Procedure Call),
432 \fBrtp\fR (Real-Time Applications protocol),
433 \fBrtcp\fR (Real-Time Applications control protocol),
434 \fBsnmp\fR (Simple Network Management Protocol),
435 \fBtftp\fR (Trivial File Transfer Protocol),
436 \fBvat\fR (Visual Audio Tool),
438 \fBwb\fR (distributed White Board).
441 \fIDon't\fP print a timestamp on each dump line.
444 Print an unformatted timestamp on each dump line.
447 Print a delta (in micro-seconds) between current and previous line
451 Print a timestamp in default format proceeded by date on each dump line.
454 Print undecoded NFS handles.
457 Make output saved via the
459 option ``packet-buffered''; i.e., as each packet is saved, it will be
460 written to the output file, rather than being written only when the
465 flag will not be supported if
467 was built with an older version of
474 (Slightly more) verbose output.
475 For example, the time to live,
476 identification, total length and options in an IP packet are printed.
477 Also enables additional packet integrity checks such as verifying the
478 IP and ICMP header checksum.
481 Even more verbose output.
482 For example, additional fields are
483 printed from NFS reply packets, and SMB packets are fully decoded.
486 Even more verbose output.
488 telnet \fBSB\fP ... \fBSE\fP options
492 Telnet options are printed in hex as well.
495 Write the raw packets to \fIfile\fR rather than parsing and printing
497 They can later be printed with the \-r option.
498 Standard output is used if \fIfile\fR is ``-''.
501 Print each packet (minus its link level header) in hex.
502 The smaller of the entire packet or
504 bytes will be printed. Note that this is the entire link-layer
505 packet, so for link layers that pad (e.g. Ethernet), the padding bytes
506 will also be printed when the higher layer packet is shorter than the
512 its link level header, in hex.
515 Print each packet (minus its link level header) in hex and ASCII.
516 This is very handy for analysing new protocols.
521 its link level header, in hex and ASCII.
524 Set the data link type to use while capturing packets to \fIdatalinktype\fP.
525 .IP "\fI expression\fP"
527 selects which packets will be dumped.
528 If no \fIexpression\fP
529 is given, all packets on the net will be dumped.
531 only packets for which \fIexpression\fP is `true' will be dumped.
533 The \fIexpression\fP consists of one or more
535 Primitives usually consist of an
537 (name or number) preceded by one or more qualifiers.
539 different kinds of qualifier:
541 qualifiers say what kind of thing the id name or number refers to.
547 E.g., `host foo', `net 128.3', `port 20'.
553 qualifiers specify a particular transfer direction to and/or from
555 Possible directions are
562 E.g., `src foo', `dst net 128.3', `src or dst port ftp-data'.
564 there is no dir qualifier,
567 For some link layers, such as SLIP and the ``cooked'' Linux capture mode
568 used for the ``any'' device and for some other device types, the
572 qualifiers can be used to specify a desired direction.
574 qualifiers restrict the match to a particular protocol.
589 E.g., `ether src foo', `arp net 128.3', `tcp port 21'.
591 no proto qualifier, all protocols consistent with the type are
593 E.g., `src foo' means `(ip or arp or rarp) src foo'
594 (except the latter is not legal syntax), `net bar' means `(ip or
595 arp or rarp) net bar' and `port 53' means `(tcp or udp) port 53'.
597 [`fddi' is actually an alias for `ether'; the parser treats them
598 identically as meaning ``the data link level used on the specified
599 network interface.'' FDDI headers contain Ethernet-like source
600 and destination addresses, and often contain Ethernet-like packet
601 types, so you can filter on these FDDI fields just as with the
602 analogous Ethernet fields.
603 FDDI headers also contain other fields,
604 but you cannot name them explicitly in a filter expression.
606 Similarly, `tr' and `wlan' are aliases for `ether'; the previous
607 paragraph's statements about FDDI headers also apply to Token Ring
608 and 802.11 wireless LAN headers. For 802.11 headers, the destination
609 address is the DA field and the source address is the SA field; the
610 BSSID, RA, and TA fields aren't tested.]
612 In addition to the above, there are some special `primitive' keywords
613 that don't follow the pattern:
618 and arithmetic expressions.
619 All of these are described below.
621 More complex filter expressions are built up by using the words
626 to combine primitives.
627 E.g., `host foo and not port ftp and not port ftp-data'.
628 To save typing, identical qualifier lists can be omitted.
630 `tcp dst port ftp or ftp-data or domain' is exactly the same as
631 `tcp dst port ftp or tcp dst port ftp-data or tcp dst port domain'.
633 Allowable primitives are:
634 .IP "\fBdst host \fIhost\fR"
635 True if the IPv4/v6 destination field of the packet is \fIhost\fP,
636 which may be either an address or a name.
637 .IP "\fBsrc host \fIhost\fR"
638 True if the IPv4/v6 source field of the packet is \fIhost\fP.
639 .IP "\fBhost \fIhost\fP
640 True if either the IPv4/v6 source or destination of the packet is \fIhost\fP.
641 Any of the above host expressions can be prepended with the keywords,
642 \fBip\fP, \fBarp\fP, \fBrarp\fP, or \fBip6\fP as in:
645 \fBip host \fIhost\fR
648 which is equivalent to:
651 \fBether proto \fI\\ip\fB and host \fIhost\fR
654 If \fIhost\fR is a name with multiple IP addresses, each address will
655 be checked for a match.
656 .IP "\fBether dst \fIehost\fP
657 True if the ethernet destination address is \fIehost\fP.
659 may be either a name from /etc/ethers or a number (see
662 .IP "\fBether src \fIehost\fP
663 True if the ethernet source address is \fIehost\fP.
664 .IP "\fBether host \fIehost\fP
665 True if either the ethernet source or destination address is \fIehost\fP.
666 .IP "\fBgateway\fP \fIhost\fP
667 True if the packet used \fIhost\fP as a gateway.
669 source or destination address was \fIhost\fP but neither the IP source
670 nor the IP destination was \fIhost\fP.
671 \fIHost\fP must be a name and
672 must be found both by the machine's host-name-to-IP-address resolution
673 mechanisms (host name file, DNS, NIS, etc.) and by the machine's
674 host-name-to-Ethernet-address resolution mechanism (/etc/ethers, etc.).
675 (An equivalent expression is
678 \fBether host \fIehost \fBand not host \fIhost\fR
681 which can be used with either names or numbers for \fIhost / ehost\fP.)
682 This syntax does not work in IPv6-enabled configuration at this moment.
683 .IP "\fBdst net \fInet\fR"
684 True if the IPv4/v6 destination address of the packet has a network
686 \fINet\fP may be either a name from /etc/networks
687 or a network number (see \fInetworks(4)\fP for details).
688 .IP "\fBsrc net \fInet\fR"
689 True if the IPv4/v6 source address of the packet has a network
691 .IP "\fBnet \fInet\fR"
692 True if either the IPv4/v6 source or destination address of the packet has a network
694 .IP "\fBnet \fInet\fR \fBmask \fInetmask\fR"
695 True if the IP address matches \fInet\fR with the specific \fInetmask\fR.
696 May be qualified with \fBsrc\fR or \fBdst\fR.
697 Note that this syntax is not valid for IPv6 \fInet\fR.
698 .IP "\fBnet \fInet\fR/\fIlen\fR"
699 True if the IPv4/v6 address matches \fInet\fR with a netmask \fIlen\fR
701 May be qualified with \fBsrc\fR or \fBdst\fR.
702 .IP "\fBdst port \fIport\fR"
703 True if the packet is ip/tcp, ip/udp, ip6/tcp or ip6/udp and has a
704 destination port value of \fIport\fP.
705 The \fIport\fP can be a number or a name used in /etc/services (see
709 If a name is used, both the port
710 number and protocol are checked.
711 If a number or ambiguous name is used,
712 only the port number is checked (e.g., \fBdst port 513\fR will print both
713 tcp/login traffic and udp/who traffic, and \fBport domain\fR will print
714 both tcp/domain and udp/domain traffic).
715 .IP "\fBsrc port \fIport\fR"
716 True if the packet has a source port value of \fIport\fP.
717 .IP "\fBport \fIport\fR"
718 True if either the source or destination port of the packet is \fIport\fP.
719 Any of the above port expressions can be prepended with the keywords,
720 \fBtcp\fP or \fBudp\fP, as in:
723 \fBtcp src port \fIport\fR
726 which matches only tcp packets whose source port is \fIport\fP.
727 .IP "\fBless \fIlength\fR"
728 True if the packet has a length less than or equal to \fIlength\fP.
729 This is equivalent to:
732 \fBlen <= \fIlength\fP.
735 .IP "\fBgreater \fIlength\fR"
736 True if the packet has a length greater than or equal to \fIlength\fP.
737 This is equivalent to:
740 \fBlen >= \fIlength\fP.
743 .IP "\fBip proto \fIprotocol\fR"
744 True if the packet is an IP packet (see
746 of protocol type \fIprotocol\fP.
747 \fIProtocol\fP can be a number or one of the names
748 \fIicmp\fP, \fIicmp6\fP, \fIigmp\fP, \fIigrp\fP, \fIpim\fP, \fIah\fP,
749 \fIesp\fP, \fIvrrp\fP, \fIudp\fP, or \fItcp\fP.
750 Note that the identifiers \fItcp\fP, \fIudp\fP, and \fIicmp\fP are also
751 keywords and must be escaped via backslash (\\), which is \\\\ in the C-shell.
752 Note that this primitive does not chase the protocol header chain.
753 .IP "\fBip6 proto \fIprotocol\fR"
754 True if the packet is an IPv6 packet of protocol type \fIprotocol\fP.
755 Note that this primitive does not chase the protocol header chain.
756 .IP "\fBip6 protochain \fIprotocol\fR"
757 True if the packet is IPv6 packet,
758 and contains protocol header with type \fIprotocol\fR
759 in its protocol header chain.
763 \fBip6 protochain 6\fR
766 matches any IPv6 packet with TCP protocol header in the protocol header chain.
767 The packet may contain, for example,
768 authentication header, routing header, or hop-by-hop option header,
769 between IPv6 header and TCP header.
770 The BPF code emitted by this primitive is complex and
771 cannot be optimized by BPF optimizer code in \fItcpdump\fP,
772 so this can be somewhat slow.
773 .IP "\fBip protochain \fIprotocol\fR"
774 Equivalent to \fBip6 protochain \fIprotocol\fR, but this is for IPv4.
775 .IP "\fBether broadcast\fR"
776 True if the packet is an ethernet broadcast packet.
779 .IP "\fBip broadcast\fR"
780 True if the packet is an IPv4 broadcast packet.
781 It checks for both the all-zeroes and all-ones broadcast conventions,
782 and looks up the subnet mask on the interface on which the capture is
785 If the subnet mask of the interface on which the capture is being done
786 is not available, either because the interface on which capture is being
787 done has no netmask or because the capture is being done on the Linux
788 "any" interface, which can capture on more than one interface, this
789 check will not work correctly.
790 .IP "\fBether multicast\fR"
791 True if the packet is an ethernet multicast packet.
794 This is shorthand for `\fBether[0] & 1 != 0\fP'.
795 .IP "\fBip multicast\fR"
796 True if the packet is an IP multicast packet.
797 .IP "\fBip6 multicast\fR"
798 True if the packet is an IPv6 multicast packet.
799 .IP "\fBether proto \fIprotocol\fR"
800 True if the packet is of ether type \fIprotocol\fR.
801 \fIProtocol\fP can be a number or one of the names
802 \fIip\fP, \fIip6\fP, \fIarp\fP, \fIrarp\fP, \fIatalk\fP, \fIaarp\fP,
803 \fIdecnet\fP, \fIsca\fP, \fIlat\fP, \fImopdl\fP, \fImoprc\fP,
804 \fIiso\fP, \fIstp\fP, \fIipx\fP, or \fInetbeui\fP.
805 Note these identifiers are also keywords
806 and must be escaped via backslash (\\).
808 [In the case of FDDI (e.g., `\fBfddi protocol arp\fR'), Token Ring
809 (e.g., `\fBtr protocol arp\fR'), and IEEE 802.11 wireless LANS (e.g.,
810 `\fBwlan protocol arp\fR'), for most of those protocols, the
811 protocol identification comes from the 802.2 Logical Link Control (LLC)
812 header, which is usually layered on top of the FDDI, Token Ring, or
815 When filtering for most protocol identifiers on FDDI, Token Ring, or
816 802.11, \fItcpdump\fR checks only the protocol ID field of an LLC header
817 in so-called SNAP format with an Organizational Unit Identifier (OUI) of
818 0x000000, for encapsulated Ethernet; it doesn't check whether the packet
819 is in SNAP format with an OUI of 0x000000.
824 \fItcpdump\fR checks the DSAP (Destination Service Access Point) and
825 SSAP (Source Service Access Point) fields of the LLC header;
827 \fBstp\fP and \fInetbeui\fP
828 \fItcpdump\fR checks the DSAP of the LLC header;
831 \fItcpdump\fR checks for a SNAP-format packet with an OUI of 0x080007
832 and the AppleTalk etype.
835 In the case of Ethernet, \fItcpdump\fR checks the Ethernet type field
836 for most of those protocols. The exceptions are:
839 \fBiso\fP, \fBsap\fP, and \fBnetbeui\fP
840 \fItcpdump\fR checks for an 802.3 frame and then checks the LLC header as
841 it does for FDDI, Token Ring, and 802.11;
844 \fItcpdump\fR checks both for the AppleTalk etype in an Ethernet frame and
845 for a SNAP-format packet as it does for FDDI, Token Ring, and 802.11;
848 \fItcpdump\fR checks for the AppleTalk ARP etype in either an Ethernet
849 frame or an 802.2 SNAP frame with an OUI of 0x000000;
852 \fItcpdump\fR checks for the IPX etype in an Ethernet frame, the IPX
853 DSAP in the LLC header, the 802.3-with-no-LLC-header encapsulation of
854 IPX, and the IPX etype in a SNAP frame.
856 .IP "\fBdecnet src \fIhost\fR"
857 True if the DECNET source address is
859 which may be an address of the form ``10.123'', or a DECNET host
861 [DECNET host name support is only available on Ultrix systems
862 that are configured to run DECNET.]
863 .IP "\fBdecnet dst \fIhost\fR"
864 True if the DECNET destination address is
866 .IP "\fBdecnet host \fIhost\fR"
867 True if either the DECNET source or destination address is
869 .IP "\fBifname \fIinterface\fR"
870 True if the packet was logged as coming from the specified interface (applies
871 only to packets logged by OpenBSD's
873 .IP "\fBon \fIinterface\fR"
877 .IP "\fBrnr \fInum\fR"
878 True if the packet was logged as matching the specified PF rule number
879 (applies only to packets logged by OpenBSD's
881 .IP "\fBrulenum \fInum\fR"
885 .IP "\fBreason \fIcode\fR"
886 True if the packet was logged with the specified PF reason code. The known
895 (applies only to packets logged by OpenBSD's
897 .IP "\fBaction \fIact\fR"
898 True if PF took the specified action when the packet was logged. Known actions
903 (applies only to packets logged by OpenBSD's
905 .IP "\fBip\fR, \fBip6\fR, \fBarp\fR, \fBrarp\fR, \fBatalk\fR, \fBaarp\fR, \fBdecnet\fR, \fBiso\fR, \fBstp\fR, \fBipx\fR, \fInetbeui\fP"
909 \fBether proto \fIp\fR
912 where \fIp\fR is one of the above protocols.
913 .IP "\fBlat\fR, \fBmoprc\fR, \fBmopdl\fR"
917 \fBether proto \fIp\fR
920 where \fIp\fR is one of the above protocols.
922 \fItcpdump\fP does not currently know how to parse these protocols.
923 .IP "\fBvlan \fI[vlan_id]\fR"
924 True if the packet is an IEEE 802.1Q VLAN packet.
925 If \fI[vlan_id]\fR is specified, only true is the packet has the specified
927 Note that the first \fBvlan\fR keyword encountered in \fIexpression\fR
928 changes the decoding offsets for the remainder of \fIexpression\fR
929 on the assumption that the packet is a VLAN packet.
930 .IP "\fBtcp\fR, \fBudp\fR, \fBicmp\fR"
934 \fBip proto \fIp\fR\fB or ip6 proto \fIp\fR
937 where \fIp\fR is one of the above protocols.
938 .IP "\fBiso proto \fIprotocol\fR"
939 True if the packet is an OSI packet of protocol type \fIprotocol\fP.
940 \fIProtocol\fP can be a number or one of the names
941 \fIclnp\fP, \fIesis\fP, or \fIisis\fP.
942 .IP "\fBclnp\fR, \fBesis\fR, \fBisis\fR"
949 where \fIp\fR is one of the above protocols.
950 .IP "\fBl1\fR, \fBl2\fR, \fBiih\fR, \fBlsp\fR, \fBsnp\fR, \fBcsnp\fR, \fBpsnp\fR"
951 Abbreviations for IS-IS PDU types.
952 .IP "\fBvpi\fP \fIn\fR
953 True if the packet is an ATM packet, for SunATM on Solaris, with a
954 virtual path identifier of
956 .IP "\fBvci\fP \fIn\fR
957 True if the packet is an ATM packet, for SunATM on Solaris, with a
958 virtual channel identifier of
961 True if the packet is an ATM packet, for SunATM on Solaris, and is
963 Note that the first \fBlane\fR keyword encountered in \fIexpression\fR
964 changes the tests done in the remainder of \fIexpression\fR
965 on the assumption that the packet is either a LANE emulated Ethernet
966 packet or a LANE LE Control packet. If \fBlane\fR isn't specified, the
967 tests are done under the assumption that the packet is an
968 LLC-encapsulated packet.
970 True if the packet is an ATM packet, for SunATM on Solaris, and is
971 an LLC-encapsulated packet.
973 True if the packet is an ATM packet, for SunATM on Solaris, and is
974 a segment OAM F4 flow cell (VPI=0 & VCI=3).
976 True if the packet is an ATM packet, for SunATM on Solaris, and is
977 an end-to-end OAM F4 flow cell (VPI=0 & VCI=4).
979 True if the packet is an ATM packet, for SunATM on Solaris, and is
980 a segment or end-to-end OAM F4 flow cell (VPI=0 & (VCI=3 | VCI=4)).
982 True if the packet is an ATM packet, for SunATM on Solaris, and is
983 a segment or end-to-end OAM F4 flow cell (VPI=0 & (VCI=3 | VCI=4)).
985 True if the packet is an ATM packet, for SunATM on Solaris, and is
986 on a meta signaling circuit (VPI=0 & VCI=1).
988 True if the packet is an ATM packet, for SunATM on Solaris, and is
989 on a broadcast signaling circuit (VPI=0 & VCI=2).
991 True if the packet is an ATM packet, for SunATM on Solaris, and is
992 on a signaling circuit (VPI=0 & VCI=5).
994 True if the packet is an ATM packet, for SunATM on Solaris, and is
995 on an ILMI circuit (VPI=0 & VCI=16).
997 True if the packet is an ATM packet, for SunATM on Solaris, and is
998 on a signaling circuit and is a Q.2931 Setup, Call Proceeding, Connect,
999 Connect Ack, Release, or Release Done message.
1000 .IP \fBmetaconnect\fP
1001 True if the packet is an ATM packet, for SunATM on Solaris, and is
1002 on a meta signaling circuit and is a Q.2931 Setup, Call Proceeding, Connect,
1003 Release, or Release Done message.
1004 .IP "\fIexpr relop expr\fR"
1005 True if the relation holds, where \fIrelop\fR is one of >, <, >=, <=, =, !=,
1006 and \fIexpr\fR is an arithmetic expression composed of integer constants
1007 (expressed in standard C syntax), the normal binary operators
1008 [+, -, *, /, &, |], a length operator, and special packet data accessors.
1010 data inside the packet, use the following syntax:
1013 \fIproto\fB [ \fIexpr\fB : \fIsize\fB ]\fR
1016 \fIProto\fR is one of \fBether, fddi, tr, wlan, ppp, slip, link,
1017 ip, arp, rarp, tcp, udp, icmp\fR or \fBip6\fR, and
1018 indicates the protocol layer for the index operation.
1019 (\fBether, fddi, wlan, tr, ppp, slip\fR and \fBlink\fR all refer to the
1021 Note that \fItcp, udp\fR and other upper-layer protocol types only
1022 apply to IPv4, not IPv6 (this will be fixed in the future).
1023 The byte offset, relative to the indicated protocol layer, is
1024 given by \fIexpr\fR.
1025 \fISize\fR is optional and indicates the number of bytes in the
1026 field of interest; it can be either one, two, or four, and defaults to one.
1027 The length operator, indicated by the keyword \fBlen\fP, gives the
1028 length of the packet.
1030 For example, `\fBether[0] & 1 != 0\fP' catches all multicast traffic.
1031 The expression `\fBip[0] & 0xf != 5\fP'
1032 catches all IP packets with options.
1034 `\fBip[6:2] & 0x1fff = 0\fP'
1035 catches only unfragmented datagrams and frag zero of fragmented datagrams.
1036 This check is implicitly applied to the \fBtcp\fP and \fBudp\fP
1038 For instance, \fBtcp[0]\fP always means the first
1039 byte of the TCP \fIheader\fP, and never means the first byte of an
1040 intervening fragment.
1042 Some offsets and field values may be expressed as names rather than
1044 The following protocol header field offsets are
1045 available: \fBicmptype\fP (ICMP type field), \fBicmpcode\fP (ICMP
1046 code field), and \fBtcpflags\fP (TCP flags field).
1048 The following ICMP type field values are available: \fBicmp-echoreply\fP,
1049 \fBicmp-unreach\fP, \fBicmp-sourcequench\fP, \fBicmp-redirect\fP,
1050 \fBicmp-echo\fP, \fBicmp-routeradvert\fP, \fBicmp-routersolicit\fP,
1051 \fBicmp-timxceed\fP, \fBicmp-paramprob\fP, \fBicmp-tstamp\fP,
1052 \fBicmp-tstampreply\fP, \fBicmp-ireq\fP, \fBicmp-ireqreply\fP,
1053 \fBicmp-maskreq\fP, \fBicmp-maskreply\fP.
1055 The following TCP flags field values are available: \fBtcp-fin\fP,
1056 \fBtcp-syn\fP, \fBtcp-rst\fP, \fBtcp-push\fP,
1057 \fBtcp-ack\fP, \fBtcp-urg\fP.
1059 Primitives may be combined using:
1061 A parenthesized group of primitives and operators
1062 (parentheses are special to the Shell and must be escaped).
1064 Negation (`\fB!\fP' or `\fBnot\fP').
1066 Concatenation (`\fB&&\fP' or `\fBand\fP').
1068 Alternation (`\fB||\fP' or `\fBor\fP').
1070 Negation has highest precedence.
1071 Alternation and concatenation have equal precedence and associate
1073 Note that explicit \fBand\fR tokens, not juxtaposition,
1074 are now required for concatenation.
1076 If an identifier is given without a keyword, the most recent keyword
1081 \fBnot host vs and ace\fR
1087 \fBnot host vs and host ace\fR
1090 which should not be confused with
1093 \fBnot ( host vs or ace )\fR
1097 Expression arguments can be passed to \fItcpdump\fP as either a single
1098 argument or as multiple arguments, whichever is more convenient.
1099 Generally, if the expression contains Shell metacharacters, it is
1100 easier to pass it as a single, quoted argument.
1101 Multiple arguments are concatenated with spaces before being parsed.
1104 To print all packets arriving at or departing from \fIsundown\fP:
1107 \fBtcpdump host sundown\fP
1111 To print traffic between \fIhelios\fR and either \fIhot\fR or \fIace\fR:
1114 \fBtcpdump host helios and \\( hot or ace \\)\fP
1118 To print all IP packets between \fIace\fR and any host except \fIhelios\fR:
1121 \fBtcpdump ip host ace and not helios\fP
1125 To print all traffic between local hosts and hosts at Berkeley:
1129 tcpdump net ucb-ether
1133 To print all ftp traffic through internet gateway \fIsnup\fP:
1134 (note that the expression is quoted to prevent the shell from
1135 (mis-)interpreting the parentheses):
1139 tcpdump 'gateway snup and (port ftp or ftp-data)'
1143 To print traffic neither sourced from nor destined for local hosts
1144 (if you gateway to one other net, this stuff should never make it
1145 onto your local net).
1149 tcpdump ip and not net \fIlocalnet\fP
1153 To print the start and end packets (the SYN and FIN packets) of each
1154 TCP conversation that involves a non-local host.
1158 tcpdump 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0 and not src and dst net \fIlocalnet\fP'
1162 To print IP packets longer than 576 bytes sent through gateway \fIsnup\fP:
1166 tcpdump 'gateway snup and ip[2:2] > 576'
1170 To print IP broadcast or multicast packets that were
1172 sent via ethernet broadcast or multicast:
1176 tcpdump 'ether[0] & 1 = 0 and ip[16] >= 224'
1180 To print all ICMP packets that are not echo requests/replies (i.e., not
1185 tcpdump 'icmp[icmptype] != icmp-echo and icmp[icmptype] != icmp-echoreply'
1190 The output of \fItcpdump\fP is protocol dependent.
1192 gives a brief description and examples of most of the formats.
1200 If the '-e' option is given, the link level header is printed out.
1201 On ethernets, the source and destination addresses, protocol,
1202 and packet length are printed.
1204 On FDDI networks, the '-e' option causes \fItcpdump\fP to print
1205 the `frame control' field, the source and destination addresses,
1206 and the packet length.
1207 (The `frame control' field governs the
1208 interpretation of the rest of the packet.
1209 Normal packets (such
1210 as those containing IP datagrams) are `async' packets, with a priority
1211 value between 0 and 7; for example, `\fBasync4\fR'.
1213 are assumed to contain an 802.2 Logical Link Control (LLC) packet;
1214 the LLC header is printed if it is \fInot\fR an ISO datagram or a
1215 so-called SNAP packet.
1217 On Token Ring networks, the '-e' option causes \fItcpdump\fP to print
1218 the `access control' and `frame control' fields, the source and
1219 destination addresses, and the packet length.
1220 As on FDDI networks,
1221 packets are assumed to contain an LLC packet.
1222 Regardless of whether
1223 the '-e' option is specified or not, the source routing information is
1224 printed for source-routed packets.
1226 On 802.11 networks, the '-e' option causes \fItcpdump\fP to print
1227 the `frame control' fields, all of the addresses in the 802.11 header,
1228 and the packet length.
1229 As on FDDI networks,
1230 packets are assumed to contain an LLC packet.
1232 \fI(N.B.: The following description assumes familiarity with
1233 the SLIP compression algorithm described in RFC-1144.)\fP
1235 On SLIP links, a direction indicator (``I'' for inbound, ``O'' for outbound),
1236 packet type, and compression information are printed out.
1237 The packet type is printed first.
1238 The three types are \fIip\fP, \fIutcp\fP, and \fIctcp\fP.
1239 No further link information is printed for \fIip\fR packets.
1240 For TCP packets, the connection identifier is printed following the type.
1241 If the packet is compressed, its encoded header is printed out.
1242 The special cases are printed out as
1243 \fB*S+\fIn\fR and \fB*SA+\fIn\fR, where \fIn\fR is the amount by which
1244 the sequence number (or sequence number and ack) has changed.
1245 If it is not a special case,
1246 zero or more changes are printed.
1247 A change is indicated by U (urgent pointer), W (window), A (ack),
1248 S (sequence number), and I (packet ID), followed by a delta (+n or -n),
1249 or a new value (=n).
1250 Finally, the amount of data in the packet and compressed header length
1253 For example, the following line shows an outbound compressed TCP packet,
1254 with an implicit connection identifier; the ack has changed by 6,
1255 the sequence number by 49, and the packet ID by 6; there are 3 bytes of
1256 data and 6 bytes of compressed header:
1259 \fBO ctcp * A+6 S+49 I+6 3 (6)\fP
1265 Arp/rarp output shows the type of request and its arguments.
1267 format is intended to be self explanatory.
1268 Here is a short sample taken from the start of an `rlogin' from
1269 host \fIrtsg\fP to host \fIcsam\fP:
1273 \f(CWarp who-has csam tell rtsg
1274 arp reply csam is-at CSAM\fR
1278 The first line says that rtsg sent an arp packet asking
1279 for the ethernet address of internet host csam.
1281 replies with its ethernet address (in this example, ethernet addresses
1282 are in caps and internet addresses in lower case).
1284 This would look less redundant if we had done \fItcpdump \-n\fP:
1288 \f(CWarp who-has 128.3.254.6 tell 128.3.254.68
1289 arp reply 128.3.254.6 is-at 02:07:01:00:01:c4\fP
1293 If we had done \fItcpdump \-e\fP, the fact that the first packet is
1294 broadcast and the second is point-to-point would be visible:
1298 \f(CWRTSG Broadcast 0806 64: arp who-has csam tell rtsg
1299 CSAM RTSG 0806 64: arp reply csam is-at CSAM\fR
1303 For the first packet this says the ethernet source address is RTSG, the
1304 destination is the ethernet broadcast address, the type field
1305 contained hex 0806 (type ETHER_ARP) and the total length was 64 bytes.
1309 \fI(N.B.:The following description assumes familiarity with
1310 the TCP protocol described in RFC-793.
1311 If you are not familiar
1312 with the protocol, neither this description nor \fItcpdump\fP will
1313 be of much use to you.)\fP
1315 The general format of a tcp protocol line is:
1319 \fIsrc > dst: flags data-seqno ack window urgent options\fP
1323 \fISrc\fP and \fIdst\fP are the source and destination IP
1324 addresses and ports.
1325 \fIFlags\fP are some combination of S (SYN),
1326 F (FIN), P (PUSH), R (RST), W (ECN CWR) or E (ECN-Echo), or a single
1328 \fIData-seqno\fP describes the portion of sequence space covered
1329 by the data in this packet (see example below).
1330 \fIAck\fP is sequence number of the next data expected the other
1331 direction on this connection.
1332 \fIWindow\fP is the number of bytes of receive buffer space available
1333 the other direction on this connection.
1334 \fIUrg\fP indicates there is `urgent' data in the packet.
1335 \fIOptions\fP are tcp options enclosed in angle brackets (e.g., <mss 1024>).
1337 \fISrc, dst\fP and \fIflags\fP are always present.
1339 depend on the contents of the packet's tcp protocol header and
1340 are output only if appropriate.
1342 Here is the opening portion of an rlogin from host \fIrtsg\fP to
1347 \s-2\f(CWrtsg.1023 > csam.login: S 768512:768512(0) win 4096 <mss 1024>
1348 csam.login > rtsg.1023: S 947648:947648(0) ack 768513 win 4096 <mss 1024>
1349 rtsg.1023 > csam.login: . ack 1 win 4096
1350 rtsg.1023 > csam.login: P 1:2(1) ack 1 win 4096
1351 csam.login > rtsg.1023: . ack 2 win 4096
1352 rtsg.1023 > csam.login: P 2:21(19) ack 1 win 4096
1353 csam.login > rtsg.1023: P 1:2(1) ack 21 win 4077
1354 csam.login > rtsg.1023: P 2:3(1) ack 21 win 4077 urg 1
1355 csam.login > rtsg.1023: P 3:4(1) ack 21 win 4077 urg 1\fR\s+2
1359 The first line says that tcp port 1023 on rtsg sent a packet
1362 The \fBS\fP indicates that the \fISYN\fP flag was set.
1363 The packet sequence number was 768512 and it contained no data.
1364 (The notation is `first:last(nbytes)' which means `sequence
1366 up to but not including \fIlast\fP which is \fInbytes\fP bytes of user data'.)
1367 There was no piggy-backed ack, the available receive window was 4096
1368 bytes and there was a max-segment-size option requesting an mss of
1371 Csam replies with a similar packet except it includes a piggy-backed
1373 Rtsg then acks csam's SYN.
1376 The packet contained no data so there is no data sequence number.
1377 Note that the ack sequence
1378 number is a small integer (1).
1379 The first time \fItcpdump\fP sees a
1380 tcp `conversation', it prints the sequence number from the packet.
1381 On subsequent packets of the conversation, the difference between
1382 the current packet's sequence number and this initial sequence number
1384 This means that sequence numbers after the
1385 first can be interpreted
1386 as relative byte positions in the conversation's data stream (with the
1387 first data byte each direction being `1').
1388 `-S' will override this
1389 feature, causing the original sequence numbers to be output.
1391 On the 6th line, rtsg sends csam 19 bytes of data (bytes 2 through 20
1392 in the rtsg \(-> csam side of the conversation).
1393 The PUSH flag is set in the packet.
1394 On the 7th line, csam says it's received data sent by rtsg up to
1395 but not including byte 21.
1396 Most of this data is apparently sitting in the
1397 socket buffer since csam's receive window has gotten 19 bytes smaller.
1398 Csam also sends one byte of data to rtsg in this packet.
1399 On the 8th and 9th lines,
1400 csam sends two bytes of urgent, pushed data to rtsg.
1402 If the snapshot was small enough that \fItcpdump\fP didn't capture
1403 the full TCP header, it interprets as much of the header as it can
1404 and then reports ``[|\fItcp\fP]'' to indicate the remainder could not
1406 If the header contains a bogus option (one with a length
1407 that's either too small or beyond the end of the header), \fItcpdump\fP
1408 reports it as ``[\fIbad opt\fP]'' and does not interpret any further
1409 options (since it's impossible to tell where they start).
1411 length indicates options are present but the IP datagram length is not
1412 long enough for the options to actually be there, \fItcpdump\fP reports
1413 it as ``[\fIbad hdr length\fP]''.
1415 .B Capturing TCP packets with particular flag combinations (SYN-ACK, URG-ACK, etc.)
1417 There are 8 bits in the control bits section of the TCP header:
1419 .I CWR | ECE | URG | ACK | PSH | RST | SYN | FIN
1421 Let's assume that we want to watch packets used in establishing
1423 Recall that TCP uses a 3-way handshake protocol
1424 when it initializes a new connection; the connection sequence with
1425 regard to the TCP control bits is
1431 2) Recipient responds with SYN, ACK
1437 Now we're interested in capturing packets that have only the
1438 SYN bit set (Step 1).
1439 Note that we don't want packets from step 2
1440 (SYN-ACK), just a plain initial SYN.
1441 What we need is a correct filter
1442 expression for \fItcpdump\fP.
1444 Recall the structure of a TCP header without options:
1448 -----------------------------------------------------------------
1449 | source port | destination port |
1450 -----------------------------------------------------------------
1452 -----------------------------------------------------------------
1453 | acknowledgment number |
1454 -----------------------------------------------------------------
1455 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
1456 -----------------------------------------------------------------
1457 | TCP checksum | urgent pointer |
1458 -----------------------------------------------------------------
1461 A TCP header usually holds 20 octets of data, unless options are
1463 The first line of the graph contains octets 0 - 3, the
1464 second line shows octets 4 - 7 etc.
1466 Starting to count with 0, the relevant TCP control bits are contained
1471 ----------------|---------------|---------------|----------------
1472 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
1473 ----------------|---------------|---------------|----------------
1474 | | 13th octet | | |
1477 Let's have a closer look at octet no. 13:
1487 These are the TCP control bits we are interested
1489 We have numbered the bits in this octet from 0 to 7, right to
1490 left, so the PSH bit is bit number 3, while the URG bit is number 5.
1492 Recall that we want to capture packets with only SYN set.
1493 Let's see what happens to octet 13 if a TCP datagram arrives
1494 with the SYN bit set in its header:
1505 control bits section we see that only bit number 1 (SYN) is set.
1507 Assuming that octet number 13 is an 8-bit unsigned integer in
1508 network byte order, the binary value of this octet is
1512 and its decimal representation is
1516 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 1*2 + 0*2 = 2
1519 We're almost done, because now we know that if only SYN is set,
1520 the value of the 13th octet in the TCP header, when interpreted
1521 as a 8-bit unsigned integer in network byte order, must be exactly 2.
1523 This relationship can be expressed as
1529 We can use this expression as the filter for \fItcpdump\fP in order
1530 to watch packets which have only SYN set:
1533 tcpdump -i xl0 tcp[13] == 2
1536 The expression says "let the 13th octet of a TCP datagram have
1537 the decimal value 2", which is exactly what we want.
1539 Now, let's assume that we need to capture SYN packets, but we
1540 don't care if ACK or any other TCP control bit is set at the
1542 Let's see what happens to octet 13 when a TCP datagram
1543 with SYN-ACK set arrives:
1553 Now bits 1 and 4 are set in the 13th octet.
1559 which translates to decimal
1563 0*2 + 0*2 + 0*2 + 1*2 + 0*2 + 0*2 + 1*2 + 0*2 = 18
1566 Now we can't just use 'tcp[13] == 18' in the \fItcpdump\fP filter
1567 expression, because that would select only those packets that have
1568 SYN-ACK set, but not those with only SYN set.
1569 Remember that we don't care
1570 if ACK or any other control bit is set as long as SYN is set.
1572 In order to achieve our goal, we need to logically AND the
1573 binary value of octet 13 with some other value to preserve
1575 We know that we want SYN to be set in any case,
1576 so we'll logically AND the value in the 13th octet with
1577 the binary value of a SYN:
1581 00010010 SYN-ACK 00000010 SYN
1582 AND 00000010 (we want SYN) AND 00000010 (we want SYN)
1584 = 00000010 = 00000010
1587 We see that this AND operation delivers the same result
1588 regardless whether ACK or another TCP control bit is set.
1589 The decimal representation of the AND value as well as
1590 the result of this operation is 2 (binary 00000010),
1591 so we know that for packets with SYN set the following
1592 relation must hold true:
1594 ( ( value of octet 13 ) AND ( 2 ) ) == ( 2 )
1596 This points us to the \fItcpdump\fP filter expression
1599 tcpdump -i xl0 'tcp[13] & 2 == 2'
1602 Note that you should use single quotes or a backslash
1603 in the expression to hide the AND ('&') special character
1609 UDP format is illustrated by this rwho packet:
1613 \f(CWactinide.who > broadcast.who: udp 84\fP
1617 This says that port \fIwho\fP on host \fIactinide\fP sent a udp
1618 datagram to port \fIwho\fP on host \fIbroadcast\fP, the Internet
1620 The packet contained 84 bytes of user data.
1622 Some UDP services are recognized (from the source or destination
1623 port number) and the higher level protocol information printed.
1624 In particular, Domain Name service requests (RFC-1034/1035) and Sun
1625 RPC calls (RFC-1050) to NFS.
1627 UDP Name Server Requests
1629 \fI(N.B.:The following description assumes familiarity with
1630 the Domain Service protocol described in RFC-1035.
1631 If you are not familiar
1632 with the protocol, the following description will appear to be written
1635 Name server requests are formatted as
1639 \fIsrc > dst: id op? flags qtype qclass name (len)\fP
1641 \f(CWh2opolo.1538 > helios.domain: 3+ A? ucbvax.berkeley.edu. (37)\fR
1645 Host \fIh2opolo\fP asked the domain server on \fIhelios\fP for an
1646 address record (qtype=A) associated with the name \fIucbvax.berkeley.edu.\fP
1647 The query id was `3'.
1648 The `+' indicates the \fIrecursion desired\fP flag
1650 The query length was 37 bytes, not including the UDP and
1651 IP protocol headers.
1652 The query operation was the normal one, \fIQuery\fP,
1653 so the op field was omitted.
1654 If the op had been anything else, it would
1655 have been printed between the `3' and the `+'.
1656 Similarly, the qclass was the normal one,
1657 \fIC_IN\fP, and omitted.
1658 Any other qclass would have been printed
1659 immediately after the `A'.
1661 A few anomalies are checked and may result in extra fields enclosed in
1662 square brackets: If a query contains an answer, authority records or
1663 additional records section,
1668 are printed as `[\fIn\fPa]', `[\fIn\fPn]' or `[\fIn\fPau]' where \fIn\fP
1669 is the appropriate count.
1670 If any of the response bits are set (AA, RA or rcode) or any of the
1671 `must be zero' bits are set in bytes two and three, `[b2&3=\fIx\fP]'
1672 is printed, where \fIx\fP is the hex value of header bytes two and three.
1674 UDP Name Server Responses
1676 Name server responses are formatted as
1680 \fIsrc > dst: id op rcode flags a/n/au type class data (len)\fP
1682 \f(CWhelios.domain > h2opolo.1538: 3 3/3/7 A 128.32.137.3 (273)
1683 helios.domain > h2opolo.1537: 2 NXDomain* 0/1/0 (97)\fR
1687 In the first example, \fIhelios\fP responds to query id 3 from \fIh2opolo\fP
1688 with 3 answer records, 3 name server records and 7 additional records.
1689 The first answer record is type A (address) and its data is internet
1690 address 128.32.137.3.
1691 The total size of the response was 273 bytes,
1692 excluding UDP and IP headers.
1693 The op (Query) and response code
1694 (NoError) were omitted, as was the class (C_IN) of the A record.
1696 In the second example, \fIhelios\fP responds to query 2 with a
1697 response code of non-existent domain (NXDomain) with no answers,
1698 one name server and no authority records.
1699 The `*' indicates that
1700 the \fIauthoritative answer\fP bit was set.
1702 answers, no type, class or data were printed.
1704 Other flag characters that might appear are `\-' (recursion available,
1705 RA, \fInot\fP set) and `|' (truncated message, TC, set).
1707 `question' section doesn't contain exactly one entry, `[\fIn\fPq]'
1710 Note that name server requests and responses tend to be large and the
1711 default \fIsnaplen\fP of 68 bytes may not capture enough of the packet
1713 Use the \fB\-s\fP flag to increase the snaplen if you
1714 need to seriously investigate name server traffic.
1716 has worked well for me.
1721 \fItcpdump\fP now includes fairly extensive SMB/CIFS/NBT decoding for data
1722 on UDP/137, UDP/138 and TCP/139.
1723 Some primitive decoding of IPX and
1724 NetBEUI SMB data is also done.
1726 By default a fairly minimal decode is done, with a much more detailed
1727 decode done if -v is used.
1728 Be warned that with -v a single SMB packet
1729 may take up a page or more, so only use -v if you really want all the
1732 If you are decoding SMB sessions containing unicode strings then you
1733 may wish to set the environment variable USE_UNICODE to 1.
1735 auto-detect unicode strings would be welcome.
1737 For information on SMB packet formats and what all te fields mean see
1738 www.cifs.org or the pub/samba/specs/ directory on your favorite
1739 samba.org mirror site.
1740 The SMB patches were written by Andrew Tridgell
1744 NFS Requests and Replies
1746 Sun NFS (Network File System) requests and replies are printed as:
1750 \fIsrc.xid > dst.nfs: len op args\fP
1751 \fIsrc.nfs > dst.xid: reply stat len op results\fP
1754 sushi.6709 > wrl.nfs: 112 readlink fh 21,24/10.73165
1755 wrl.nfs > sushi.6709: reply ok 40 readlink "../var"
1756 sushi.201b > wrl.nfs:
1757 144 lookup fh 9,74/4096.6878 "xcolors"
1758 wrl.nfs > sushi.201b:
1759 reply ok 128 lookup fh 9,74/4134.3150
1764 In the first line, host \fIsushi\fP sends a transaction with id \fI6709\fP
1765 to \fIwrl\fP (note that the number following the src host is a
1766 transaction id, \fInot\fP the source port).
1767 The request was 112 bytes,
1768 excluding the UDP and IP headers.
1769 The operation was a \fIreadlink\fP
1770 (read symbolic link) on file handle (\fIfh\fP) 21,24/10.731657119.
1771 (If one is lucky, as in this case, the file handle can be interpreted
1772 as a major,minor device number pair, followed by the inode number and
1774 \fIWrl\fP replies `ok' with the contents of the link.
1776 In the third line, \fIsushi\fP asks \fIwrl\fP to lookup the name
1777 `\fIxcolors\fP' in directory file 9,74/4096.6878.
1778 Note that the data printed
1779 depends on the operation type.
1780 The format is intended to be self
1781 explanatory if read in conjunction with
1782 an NFS protocol spec.
1784 If the \-v (verbose) flag is given, additional information is printed.
1790 sushi.1372a > wrl.nfs:
1791 148 read fh 21,11/12.195 8192 bytes @ 24576
1792 wrl.nfs > sushi.1372a:
1793 reply ok 1472 read REG 100664 ids 417/0 sz 29388
1798 (\-v also prints the IP header TTL, ID, length, and fragmentation fields,
1799 which have been omitted from this example.) In the first line,
1800 \fIsushi\fP asks \fIwrl\fP to read 8192 bytes from file 21,11/12.195,
1801 at byte offset 24576.
1802 \fIWrl\fP replies `ok'; the packet shown on the
1803 second line is the first fragment of the reply, and hence is only 1472
1804 bytes long (the other bytes will follow in subsequent fragments, but
1805 these fragments do not have NFS or even UDP headers and so might not be
1806 printed, depending on the filter expression used).
1807 Because the \-v flag
1808 is given, some of the file attributes (which are returned in addition
1809 to the file data) are printed: the file type (``REG'', for regular file),
1810 the file mode (in octal), the uid and gid, and the file size.
1812 If the \-v flag is given more than once, even more details are printed.
1814 Note that NFS requests are very large and much of the detail won't be printed
1815 unless \fIsnaplen\fP is increased.
1816 Try using `\fB\-s 192\fP' to watch
1819 NFS reply packets do not explicitly identify the RPC operation.
1821 \fItcpdump\fP keeps track of ``recent'' requests, and matches them to the
1822 replies using the transaction ID.
1823 If a reply does not closely follow the
1824 corresponding request, it might not be parsable.
1826 AFS Requests and Replies
1828 Transarc AFS (Andrew File System) requests and replies are printed
1834 \fIsrc.sport > dst.dport: rx packet-type\fP
1835 \fIsrc.sport > dst.dport: rx packet-type service call call-name args\fP
1836 \fIsrc.sport > dst.dport: rx packet-type service reply call-name args\fP
1839 elvis.7001 > pike.afsfs:
1840 rx data fs call rename old fid 536876964/1/1 ".newsrc.new"
1841 new fid 536876964/1/1 ".newsrc"
1842 pike.afsfs > elvis.7001: rx data fs reply rename
1847 In the first line, host elvis sends a RX packet to pike.
1849 a RX data packet to the fs (fileserver) service, and is the start of
1851 The RPC call was a rename, with the old directory file id
1852 of 536876964/1/1 and an old filename of `.newsrc.new', and a new directory
1853 file id of 536876964/1/1 and a new filename of `.newsrc'.
1855 responds with a RPC reply to the rename call (which was successful, because
1856 it was a data packet and not an abort packet).
1858 In general, all AFS RPCs are decoded at least by RPC call name.
1860 AFS RPCs have at least some of the arguments decoded (generally only
1861 the `interesting' arguments, for some definition of interesting).
1863 The format is intended to be self-describing, but it will probably
1864 not be useful to people who are not familiar with the workings of
1867 If the -v (verbose) flag is given twice, acknowledgement packets and
1868 additional header information is printed, such as the the RX call ID,
1869 call number, sequence number, serial number, and the RX packet flags.
1871 If the -v flag is given twice, additional information is printed,
1872 such as the the RX call ID, serial number, and the RX packet flags.
1873 The MTU negotiation information is also printed from RX ack packets.
1875 If the -v flag is given three times, the security index and service id
1878 Error codes are printed for abort packets, with the exception of Ubik
1879 beacon packets (because abort packets are used to signify a yes vote
1880 for the Ubik protocol).
1882 Note that AFS requests are very large and many of the arguments won't
1883 be printed unless \fIsnaplen\fP is increased.
1884 Try using `\fB-s 256\fP'
1885 to watch AFS traffic.
1887 AFS reply packets do not explicitly identify the RPC operation.
1889 \fItcpdump\fP keeps track of ``recent'' requests, and matches them to the
1890 replies using the call number and service ID.
1891 If a reply does not closely
1893 corresponding request, it might not be parsable.
1896 KIP AppleTalk (DDP in UDP)
1898 AppleTalk DDP packets encapsulated in UDP datagrams are de-encapsulated
1899 and dumped as DDP packets (i.e., all the UDP header information is
1903 is used to translate appletalk net and node numbers to names.
1904 Lines in this file have the form
1916 The first two lines give the names of appletalk networks.
1918 line gives the name of a particular host (a host is distinguished
1919 from a net by the 3rd octet in the number \-
1920 a net number \fImust\fP have two octets and a host number \fImust\fP
1921 have three octets.) The number and name should be separated by
1922 whitespace (blanks or tabs).
1925 file may contain blank lines or comment lines (lines starting with
1928 AppleTalk addresses are printed in the form
1934 \f(CW144.1.209.2 > icsd-net.112.220
1935 office.2 > icsd-net.112.220
1936 jssmag.149.235 > icsd-net.2\fR
1942 doesn't exist or doesn't contain an entry for some appletalk
1943 host/net number, addresses are printed in numeric form.)
1944 In the first example, NBP (DDP port 2) on net 144.1 node 209
1945 is sending to whatever is listening on port 220 of net icsd node 112.
1946 The second line is the same except the full name of the source node
1947 is known (`office').
1948 The third line is a send from port 235 on
1949 net jssmag node 149 to broadcast on the icsd-net NBP port (note that
1950 the broadcast address (255) is indicated by a net name with no host
1951 number \- for this reason it's a good idea to keep node names and
1952 net names distinct in /etc/atalk.names).
1954 NBP (name binding protocol) and ATP (AppleTalk transaction protocol)
1955 packets have their contents interpreted.
1956 Other protocols just dump
1957 the protocol name (or number if no name is registered for the
1958 protocol) and packet size.
1960 \fBNBP packets\fP are formatted like the following examples:
1964 \s-2\f(CWicsd-net.112.220 > jssmag.2: nbp-lkup 190: "=:LaserWriter@*"
1965 jssmag.209.2 > icsd-net.112.220: nbp-reply 190: "RM1140:LaserWriter@*" 250
1966 techpit.2 > icsd-net.112.220: nbp-reply 190: "techpit:LaserWriter@*" 186\fR\s+2
1970 The first line is a name lookup request for laserwriters sent by net icsd host
1971 112 and broadcast on net jssmag.
1972 The nbp id for the lookup is 190.
1973 The second line shows a reply for this request (note that it has the
1974 same id) from host jssmag.209 saying that it has a laserwriter
1975 resource named "RM1140" registered on port 250.
1977 another reply to the same request saying host techpit has laserwriter
1978 "techpit" registered on port 186.
1980 \fBATP packet\fP formatting is demonstrated by the following example:
1984 \s-2\f(CWjssmag.209.165 > helios.132: atp-req 12266<0-7> 0xae030001
1985 helios.132 > jssmag.209.165: atp-resp 12266:0 (512) 0xae040000
1986 helios.132 > jssmag.209.165: atp-resp 12266:1 (512) 0xae040000
1987 helios.132 > jssmag.209.165: atp-resp 12266:2 (512) 0xae040000
1988 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1989 helios.132 > jssmag.209.165: atp-resp 12266:4 (512) 0xae040000
1990 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1991 helios.132 > jssmag.209.165: atp-resp 12266:6 (512) 0xae040000
1992 helios.132 > jssmag.209.165: atp-resp*12266:7 (512) 0xae040000
1993 jssmag.209.165 > helios.132: atp-req 12266<3,5> 0xae030001
1994 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1995 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1996 jssmag.209.165 > helios.132: atp-rel 12266<0-7> 0xae030001
1997 jssmag.209.133 > helios.132: atp-req* 12267<0-7> 0xae030002\fR\s+2
2001 Jssmag.209 initiates transaction id 12266 with host helios by requesting
2002 up to 8 packets (the `<0-7>').
2003 The hex number at the end of the line
2004 is the value of the `userdata' field in the request.
2006 Helios responds with 8 512-byte packets.
2007 The `:digit' following the
2008 transaction id gives the packet sequence number in the transaction
2009 and the number in parens is the amount of data in the packet,
2010 excluding the atp header.
2011 The `*' on packet 7 indicates that the
2014 Jssmag.209 then requests that packets 3 & 5 be retransmitted.
2016 resends them then jssmag.209 releases the transaction.
2018 jssmag.209 initiates the next request.
2019 The `*' on the request
2020 indicates that XO (`exactly once') was \fInot\fP set.
2025 Fragmented Internet datagrams are printed as
2029 \fB(frag \fIid\fB:\fIsize\fB@\fIoffset\fB+)\fR
2030 \fB(frag \fIid\fB:\fIsize\fB@\fIoffset\fB)\fR
2034 (The first form indicates there are more fragments.
2036 indicates this is the last fragment.)
2038 \fIId\fP is the fragment id.
2039 \fISize\fP is the fragment
2040 size (in bytes) excluding the IP header.
2041 \fIOffset\fP is this
2042 fragment's offset (in bytes) in the original datagram.
2044 The fragment information is output for each fragment.
2046 fragment contains the higher level protocol header and the frag
2047 info is printed after the protocol info.
2049 after the first contain no higher level protocol header and the
2050 frag info is printed after the source and destination addresses.
2051 For example, here is part of an ftp from arizona.edu to lbl-rtsg.arpa
2052 over a CSNET connection that doesn't appear to handle 576 byte datagrams:
2056 \s-2\f(CWarizona.ftp-data > rtsg.1170: . 1024:1332(308) ack 1 win 4096 (frag 595a:328@0+)
2057 arizona > rtsg: (frag 595a:204@328)
2058 rtsg.1170 > arizona.ftp-data: . ack 1536 win 2560\fP\s+2
2062 There are a couple of things to note here: First, addresses in the
2063 2nd line don't include port numbers.
2064 This is because the TCP
2065 protocol information is all in the first fragment and we have no idea
2066 what the port or sequence numbers are when we print the later fragments.
2067 Second, the tcp sequence information in the first line is printed as if there
2068 were 308 bytes of user data when, in fact, there are 512 bytes (308 in
2069 the first frag and 204 in the second).
2070 If you are looking for holes
2071 in the sequence space or trying to match up acks
2072 with packets, this can fool you.
2074 A packet with the IP \fIdon't fragment\fP flag is marked with a
2075 trailing \fB(DF)\fP.
2079 By default, all output lines are preceded by a timestamp.
2081 is the current clock time in the form
2087 and is as accurate as the kernel's clock.
2088 The timestamp reflects the time the kernel first saw the packet.
2090 is made to account for the time lag between when the
2091 ethernet interface removed the packet from the wire and when the kernel
2092 serviced the `new packet' interrupt.
2094 traffic(1C), nit(4P), bpf(4), pcap(3)
2096 The original authors are:
2100 Steven McCanne, all of the
2101 Lawrence Berkeley National Laboratory, University of California, Berkeley, CA.
2103 It is currently being maintained by tcpdump.org.
2105 The current version is available via http:
2108 .I http://www.tcpdump.org/
2111 The original distribution is available via anonymous ftp:
2114 .I ftp://ftp.ee.lbl.gov/tcpdump.tar.Z
2117 IPv6/IPsec support is added by WIDE/KAME project.
2118 This program uses Eric Young's SSLeay library, under specific configuration.
2120 Please send problems, bugs, questions, desirable enhancements, etc. to:
2123 tcpdump-workers@tcpdump.org
2126 Please send source code contributions, etc. to:
2132 NIT doesn't let you watch your own outbound traffic, BPF will.
2133 We recommend that you use the latter.
2135 On Linux systems with 2.0[.x] kernels:
2137 packets on the loopback device will be seen twice;
2139 packet filtering cannot be done in the kernel, so that all packets must
2140 be copied from the kernel in order to be filtered in user mode;
2142 all of a packet, not just the part that's within the snapshot length,
2143 will be copied from the kernel (the 2.0[.x] packet capture mechanism, if
2144 asked to copy only part of a packet to userland, will not report the
2145 true length of the packet; this would cause most IP packets to get an
2149 capturing on some PPP devices won't work correctly.
2151 We recommend that you upgrade to a 2.2 or later kernel.
2153 Some attempt should be made to reassemble IP fragments or, at least
2154 to compute the right length for the higher level protocol.
2156 Name server inverse queries are not dumped correctly: the (empty)
2157 question section is printed rather than real query in the answer
2159 Some believe that inverse queries are themselves a bug and
2160 prefer to fix the program generating them rather than \fItcpdump\fP.
2162 A packet trace that crosses a daylight savings time change will give
2163 skewed time stamps (the time change is ignored).
2165 Filter expressions on fields other than those in Token Ring headers will
2166 not correctly handle source-routed Token Ring packets.
2168 Filter expressions on fields other than those in 802.11 headers will not
2169 correctly handle 802.11 data packets with both To DS and From DS set.
2172 should chase header chain, but at this moment it does not.
2173 .BR "ip6 protochain"
2174 is supplied for this behavior.
2176 Arithmetic expression against transport layer headers, like \fBtcp[0]\fP,
2177 does not work against IPv6 packets.
2178 It only looks at IPv4 packets.
projects / tcpdump / blob