1 .\" @(#) $Header: /tcpdump/master/tcpdump/Attic/tcpdump.1,v 1.120 2002-05-17 09:57:50 guy Exp $ (LBL)
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 "3 January 2001"
25 tcpdump \- dump traffic on a network
30 .B \-aAdeflnNOpqRStuvxX
85 \fITcpdump\fP prints out the headers of packets on a network interface
86 that match the boolean \fIexpression\fP. It can also be run with the
88 flag, which causes it to save the packet data to a file for later
89 analysis, and/or with the
91 flag, which causes it to read from a saved packet file rather than to
92 read packets from a network interface. In all cases, only packets that
99 will, if not run with the
101 flag, continue capturing packets until it is interrupted by a SIGINT
102 signal (generated, for example, by typing your interrupt character,
103 typically control-C) or a SIGTERM signal (typically generated with the
105 command); if run with the
107 flag, it will capture packets until it is interrupted by a SIGINT or
108 SIGTERM signal or the specified number of packets have been processed.
112 finishes capturing packets, it will report counts of:
114 packets ``received by filter'' (the meaning of this depends on the OS on
117 and possibly on the way the OS was configured - if a filter was
118 specified on the command line, on some OSes it counts packets regardless
119 of whether they were matched by the filter expression, and on other OSes
120 it counts only packets that were matched by the filter expression and
124 packets ``dropped by kernel'' (this is the number of packets that were
125 dropped, due to a lack of buffer space, by the packet capture mechanism
128 is running, if the OS reports that information to applications; if not,
129 it will be reported as 0).
131 On platforms that support the SIGINFO signal, such as most BSDs, it will
132 report those counts when it receives a SIGINFO signal (generated, for
133 example, by typing your ``status'' character, typically control-T) and
134 will continue capturing packets.
136 Reading packets from a network interface may require that you have
139 .B Under SunOS 3.x or 4.x with NIT or BPF:
140 You must have read access to
145 .B Under Solaris with DLPI:
146 You must have read/write access to the network pseudo device, e.g.
148 On at least some versions of Solaris, however, this is not sufficient to
151 to capture in promiscuous mode; on those versions of Solaris, you must
154 must be installed setuid to root, in order to capture in promiscuous
155 mode. Note that, on many (perhaps all) interfaces, if you don't capture
156 in promiscuous mode, you will not see any outgoing packets, so a capture
157 not done in promiscuous mode may not be very useful.
159 .B Under HP-UX with DLPI:
162 must be installed setuid to root.
164 .B Under IRIX with snoop:
167 must be installed setuid to root.
172 must be installed setuid to root.
174 .B Under Ultrix and Digital UNIX/Tru64 UNIX:
175 Any user may capture network traffic with
177 However, no user (not even the super-user) can capture in promiscuous
178 mode on an interface unless the super-user has enabled promiscuous-mode
179 operation on that interface using
181 and no user (not even the super-user) can capture unicast traffic
182 received by or sent by the machine on an interface unless the super-user
183 has enabled copy-all-mode operation on that interface using
187 packet capture on an interface probably requires that either
188 promiscuous-mode or copy-all-mode operation, or both modes of
189 operation, be enabled on that interface.
192 You must have read access to
195 Reading a saved packet file doesn't require special privileges.
200 Print each packet (minus its link level header) in ASCII. Handy for
203 Attempt to convert network and broadcast addresses to names.
206 Exit after receiving \fIcount\fP packets.
209 Before writing a raw packet to a savefile, check whether the file is
210 currently larger than \fIfile_size\fP and, if so, close the current
211 savefile and open a new one. Savefiles after the first savefile will
212 have the name specified with the
214 flag, with a number after it, starting at 2 and continuing upward.
215 The units of \fIfile_size\fP are millions of bytes (1,000,000 bytes,
216 not 1,048,576 bytes).
219 Dump the compiled packet-matching code in a human readable form to
220 standard output and stop.
223 Dump packet-matching code as a
228 Dump packet-matching code as decimal numbers (preceded with a count).
231 Print the link-level header on each dump line.
234 Use \fIalgo:secret\fP for decrypting IPsec ESP packets.
240 \fBcast128-cbc\fP, or
242 The default is \fBdes-cbc\fP.
243 The ability to decrypt packets is only present if \fItcpdump\fP was compiled
244 with cryptography enabled.
245 \fIsecret\fP the ascii text for ESP secret key.
246 We cannot take arbitrary binary value at this moment.
247 The option assumes RFC2406 ESP, not RFC1827 ESP.
248 The option is only for debugging purposes, and
249 the use of this option with truly `secret' key is discouraged.
250 By presenting IPsec secret key onto command line
251 you make it visible to others, via
256 Print `foreign' internet addresses numerically rather than symbolically
257 (this option is intended to get around serious brain damage in
258 Sun's yp server \(em usually it hangs forever translating non-local
262 Use \fIfile\fP as input for the filter expression.
263 An additional expression given on the command line is ignored.
266 Listen on \fIinterface\fP.
267 If unspecified, \fItcpdump\fP searches the system interface list for the
268 lowest numbered, configured up interface (excluding loopback).
269 Ties are broken by choosing the earliest match.
271 On Linux systems with 2.2 or later kernels, an
273 argument of ``any'' can be used to capture packets from all interfaces.
274 Note that captures on the ``any'' device will not be done in promiscuous
278 Make stdout line buffered.
279 Useful if you want to see the data
283 ``tcpdump\ \ \-l\ \ |\ \ tee dat'' or
284 ``tcpdump\ \ \-l \ \ > dat\ \ &\ \ tail\ \ \-f\ \ dat''.
287 Load SMI MIB module definitions from file \fImodule\fR.
289 can be used several times to load several MIB modules into \fItcpdump\fP.
292 Don't convert addresses (i.e., host addresses, port numbers, etc.) to names.
295 Don't print domain name qualification of host names.
297 if you give this flag then \fItcpdump\fP will print ``nic''
298 instead of ``nic.ddn.mil''.
301 Do not run the packet-matching code optimizer.
303 if you suspect a bug in the optimizer.
306 \fIDon't\fP put the interface
307 into promiscuous mode.
308 Note that the interface might be in promiscuous
309 mode for some other reason; hence, `-p' cannot be used as an abbreviation for
310 `ether host {local-hw-addr} or ether broadcast'.
313 Quick (quiet?) output.
314 Print less protocol information so output
318 Assume ESP/AH packets to be based on old specification (RFC1825 to RFC1829).
319 If specified, \fItcpdump\fP will not print replay prevention field.
320 Since there is no protocol version field in ESP/AH specification,
321 \fItcpdump\fP cannot deduce the version of ESP/AH protocol.
324 Read packets from \fIfile\fR (which was created with the -w option).
325 Standard input is used if \fIfile\fR is ``-''.
328 Print absolute, rather than relative, TCP sequence numbers.
331 Snarf \fIsnaplen\fP bytes of data from each packet rather than the
332 default of 68 (with SunOS's NIT, the minimum is actually 96).
333 68 bytes is adequate for IP, ICMP, TCP
334 and UDP but may truncate protocol information from name server and NFS
336 Packets truncated because of a limited snapshot
337 are indicated in the output with ``[|\fIproto\fP]'', where \fIproto\fP
338 is the name of the protocol level at which the truncation has occurred.
339 Note that taking larger snapshots both increases
340 the amount of time it takes to process packets and, effectively,
341 decreases the amount of packet buffering.
342 This may cause packets to be
344 You should limit \fIsnaplen\fP to the smallest number that will
345 capture the protocol information you're interested in.
347 \fIsnaplen\fP to 0 means use the required length to catch whole packets.
350 Force packets selected by "\fIexpression\fP" to be interpreted the
351 specified \fItype\fR.
352 Currently known types are
353 \fBcnfp\fR (Cisco NetFlow protocol),
354 \fBrpc\fR (Remote Procedure Call),
355 \fBrtp\fR (Real-Time Applications protocol),
356 \fBrtcp\fR (Real-Time Applications control protocol),
357 \fBsnmp\fR (Simple Network Management Protocol),
358 \fBvat\fR (Visual Audio Tool),
360 \fBwb\fR (distributed White Board).
363 \fIDon't\fP print a timestamp on each dump line.
366 Print an unformatted timestamp on each dump line.
369 Print a delta (in micro-seconds) between current and previous line
373 Print a timestamp in default format proceeded by date on each dump line.
376 Print undecoded NFS handles.
379 (Slightly more) verbose output.
380 For example, the time to live,
381 identification, total length and options in an IP packet are printed.
382 Also enables additional packet integrity checks such as verifying the
383 IP and ICMP header checksum.
386 Even more verbose output.
387 For example, additional fields are
388 printed from NFS reply packets, and SMB packets are fully decoded.
391 Even more verbose output.
393 telnet \fBSB\fP ... \fBSE\fP options
397 telnet options are printed in hex as well.
400 Write the raw packets to \fIfile\fR rather than parsing and printing
402 They can later be printed with the \-r option.
403 Standard output is used if \fIfile\fR is ``-''.
406 Print each packet (minus its link level header) in hex.
407 The smaller of the entire packet or
409 bytes will be printed.
412 When printing hex, print ascii too.
415 is also set, the packet is printed in hex/ascii.
416 This is very handy for analysing new protocols.
419 is not also set, some parts of some packets may be printed
421 .IP "\fI expression\fP"
423 selects which packets will be dumped.
424 If no \fIexpression\fP
425 is given, all packets on the net will be dumped.
427 only packets for which \fIexpression\fP is `true' will be dumped.
429 The \fIexpression\fP consists of one or more
431 Primitives usually consist of an
433 (name or number) preceded by one or more qualifiers.
435 different kinds of qualifier:
437 qualifiers say what kind of thing the id name or number refers to.
443 E.g., `host foo', `net 128.3', `port 20'.
449 qualifiers specify a particular transfer direction to and/or from
451 Possible directions are
458 E.g., `src foo', `dst net 128.3', `src or dst port ftp-data'.
460 there is no dir qualifier,
463 For `null' link layers (i.e. point to point protocols such as slip) the
467 qualifiers can be used to specify a desired direction.
469 qualifiers restrict the match to a particular protocol.
483 E.g., `ether src foo', `arp net 128.3', `tcp port 21'.
485 no proto qualifier, all protocols consistent with the type are
487 E.g., `src foo' means `(ip or arp or rarp) src foo'
488 (except the latter is not legal syntax), `net bar' means `(ip or
489 arp or rarp) net bar' and `port 53' means `(tcp or udp) port 53'.
491 [`fddi' is actually an alias for `ether'; the parser treats them
492 identically as meaning ``the data link level used on the specified
493 network interface.'' FDDI headers contain Ethernet-like source
494 and destination addresses, and often contain Ethernet-like packet
495 types, so you can filter on these FDDI fields just as with the
496 analogous Ethernet fields.
497 FDDI headers also contain other fields,
498 but you cannot name them explicitly in a filter expression.
500 Similarly, `tr' is an alias for `ether'; the previous paragraph's
501 statements about FDDI headers also apply to Token Ring headers.]
503 In addition to the above, there are some special `primitive' keywords
504 that don't follow the pattern:
509 and arithmetic expressions.
510 All of these are described below.
512 More complex filter expressions are built up by using the words
517 to combine primitives.
518 E.g., `host foo and not port ftp and not port ftp-data'.
519 To save typing, identical qualifier lists can be omitted.
521 `tcp dst port ftp or ftp-data or domain' is exactly the same as
522 `tcp dst port ftp or tcp dst port ftp-data or tcp dst port domain'.
524 Allowable primitives are:
525 .IP "\fBdst host \fIhost\fR"
526 True if the IPv4/v6 destination field of the packet is \fIhost\fP,
527 which may be either an address or a name.
528 .IP "\fBsrc host \fIhost\fR"
529 True if the IPv4/v6 source field of the packet is \fIhost\fP.
530 .IP "\fBhost \fIhost\fP
531 True if either the IPv4/v6 source or destination of the packet is \fIhost\fP.
532 Any of the above host expressions can be prepended with the keywords,
533 \fBip\fP, \fBarp\fP, \fBrarp\fP, or \fBip6\fP as in:
536 \fBip host \fIhost\fR
539 which is equivalent to:
542 \fBether proto \fI\\ip\fB and host \fIhost\fR
545 If \fIhost\fR is a name with multiple IP addresses, each address will
546 be checked for a match.
547 .IP "\fBether dst \fIehost\fP
548 True if the ethernet destination address is \fIehost\fP.
550 may be either a name from /etc/ethers or a number (see
553 .IP "\fBether src \fIehost\fP
554 True if the ethernet source address is \fIehost\fP.
555 .IP "\fBether host \fIehost\fP
556 True if either the ethernet source or destination address is \fIehost\fP.
557 .IP "\fBgateway\fP \fIhost\fP
558 True if the packet used \fIhost\fP as a gateway.
560 source or destination address was \fIhost\fP but neither the IP source
561 nor the IP destination was \fIhost\fP.
562 \fIHost\fP must be a name and
563 must be found both by the machine's host-name-to-IP-address resolution
564 mechanisms (host name file, DNS, NIS, etc.) and by the machine's
565 host-name-to-Ethernet-address resolution mechanism (/etc/ethers, etc.).
566 (An equivalent expression is
569 \fBether host \fIehost \fBand not host \fIhost\fR
572 which can be used with either names or numbers for \fIhost / ehost\fP.)
573 This syntax does not work in IPv6-enabled configuration at this moment.
574 .IP "\fBdst net \fInet\fR"
575 True if the IPv4/v6 destination address of the packet has a network
577 \fINet\fP may be either a name from /etc/networks
578 or a network number (see \fInetworks(4)\fP for details).
579 .IP "\fBsrc net \fInet\fR"
580 True if the IPv4/v6 source address of the packet has a network
582 .IP "\fBnet \fInet\fR"
583 True if either the IPv4/v6 source or destination address of the packet has a network
585 .IP "\fBnet \fInet\fR \fBmask \fInetmask\fR"
586 True if the IP address matches \fInet\fR with the specific \fInetmask\fR.
587 May be qualified with \fBsrc\fR or \fBdst\fR.
588 Note that this syntax is not valid for IPv6 \fInet\fR.
589 .IP "\fBnet \fInet\fR/\fIlen\fR"
590 True if the IPv4/v6 address matches \fInet\fR with a netmask \fIlen\fR
592 May be qualified with \fBsrc\fR or \fBdst\fR.
593 .IP "\fBdst port \fIport\fR"
594 True if the packet is ip/tcp, ip/udp, ip6/tcp or ip6/udp and has a
595 destination port value of \fIport\fP.
596 The \fIport\fP can be a number or a name used in /etc/services (see
600 If a name is used, both the port
601 number and protocol are checked.
602 If a number or ambiguous name is used,
603 only the port number is checked (e.g., \fBdst port 513\fR will print both
604 tcp/login traffic and udp/who traffic, and \fBport domain\fR will print
605 both tcp/domain and udp/domain traffic).
606 .IP "\fBsrc port \fIport\fR"
607 True if the packet has a source port value of \fIport\fP.
608 .IP "\fBport \fIport\fR"
609 True if either the source or destination port of the packet is \fIport\fP.
610 Any of the above port expressions can be prepended with the keywords,
611 \fBtcp\fP or \fBudp\fP, as in:
614 \fBtcp src port \fIport\fR
617 which matches only tcp packets whose source port is \fIport\fP.
618 .IP "\fBless \fIlength\fR"
619 True if the packet has a length less than or equal to \fIlength\fP.
620 This is equivalent to:
623 \fBlen <= \fIlength\fP.
626 .IP "\fBgreater \fIlength\fR"
627 True if the packet has a length greater than or equal to \fIlength\fP.
628 This is equivalent to:
631 \fBlen >= \fIlength\fP.
634 .IP "\fBip proto \fIprotocol\fR"
635 True if the packet is an IP packet (see
637 of protocol type \fIprotocol\fP.
638 \fIProtocol\fP can be a number or one of the names
639 \fIicmp\fP, \fIicmp6\fP, \fIigmp\fP, \fIigrp\fP, \fIpim\fP, \fIah\fP,
640 \fIesp\fP, \fIvrrp\fP, \fIudp\fP, or \fItcp\fP.
641 Note that the identifiers \fItcp\fP, \fIudp\fP, and \fIicmp\fP are also
642 keywords and must be escaped via backslash (\\), which is \\\\ in the C-shell.
643 Note that this primitive does not chase the protocol header chain.
644 .IP "\fBip6 proto \fIprotocol\fR"
645 True if the packet is an IPv6 packet of protocol type \fIprotocol\fP.
646 Note that this primitive does not chase the protocol header chain.
647 .IP "\fBip6 protochain \fIprotocol\fR"
648 True if the packet is IPv6 packet,
649 and contains protocol header with type \fIprotocol\fR
650 in its protocol header chain.
654 \fBip6 protochain 6\fR
657 matches any IPv6 packet with TCP protocol header in the protocol header chain.
658 The packet may contain, for example,
659 authentication header, routing header, or hop-by-hop option header,
660 between IPv6 header and TCP header.
661 The BPF code emitted by this primitive is complex and
662 cannot be optimized by BPF optimizer code in \fItcpdump\fP,
663 so this can be somewhat slow.
664 .IP "\fBip protochain \fIprotocol\fR"
665 Equivalent to \fBip6 protochain \fIprotocol\fR, but this is for IPv4.
666 .IP "\fBether broadcast\fR"
667 True if the packet is an ethernet broadcast packet.
670 .IP "\fBip broadcast\fR"
671 True if the packet is an IP broadcast packet.
673 the all-zeroes and all-ones broadcast conventions, and looks up
674 the local subnet mask.
675 .IP "\fBether multicast\fR"
676 True if the packet is an ethernet multicast packet.
679 This is shorthand for `\fBether[0] & 1 != 0\fP'.
680 .IP "\fBip multicast\fR"
681 True if the packet is an IP multicast packet.
682 .IP "\fBip6 multicast\fR"
683 True if the packet is an IPv6 multicast packet.
684 .IP "\fBether proto \fIprotocol\fR"
685 True if the packet is of ether type \fIprotocol\fR.
686 \fIProtocol\fP can be a number or one of the names
687 \fIip\fP, \fIip6\fP, \fIarp\fP, \fIrarp\fP, \fIatalk\fP, \fIaarp\fP,
688 \fIdecnet\fP, \fIsca\fP, \fIlat\fP, \fImopdl\fP, \fImoprc\fP,
689 \fIiso\fP, \fIstp\fP, \fIipx\fP, or \fInetbeui\fP.
690 Note these identifiers are also keywords
691 and must be escaped via backslash (\\).
693 [In the case of FDDI (e.g., `\fBfddi protocol arp\fR') and Token Ring
694 (e.g., `\fBtr protocol arp\fR'), for most of those protocols, the
695 protocol identification comes from the 802.2 Logical Link Control (LLC)
696 header, which is usually layered on top of the FDDI or Token Ring
699 When filtering for most protocol identifiers on FDDI or Token Ring,
700 \fItcpdump\fR checks only the protocol ID field of an LLC header in
701 so-called SNAP format with an Organizational Unit Identifier (OUI) of
702 0x000000, for encapsulated Ethernet; it doesn't check whether the packet
703 is in SNAP format with an OUI of 0x000000.
705 The exceptions are \fIiso\fP, for which it checks the DSAP (Destination
706 Service Access Point) and SSAP (Source Service Access Point) fields of
707 the LLC header, \fIstp\fP and \fInetbeui\fP, where it checks the DSAP of
708 the LLC header, and \fIatalk\fP, where it checks for a SNAP-format
709 packet with an OUI of 0x080007 and the Appletalk etype.
711 In the case of Ethernet, \fItcpdump\fR checks the Ethernet type field
712 for most of those protocols; the exceptions are \fIiso\fP, \fIsap\fP,
713 and \fInetbeui\fP, for which it checks for an 802.3 frame and then
714 checks the LLC header as it does for FDDI and Token Ring, \fIatalk\fP,
715 where it checks both for the Appletalk etype in an Ethernet frame and
716 for a SNAP-format packet as it does for FDDI and Token Ring, \fIaarp\fP,
717 where it checks for the Appletalk ARP etype in either an Ethernet frame
718 or an 802.2 SNAP frame with an OUI of 0x000000, and \fIipx\fP, where it
719 checks for the IPX etype in an Ethernet frame, the IPX DSAP in the LLC
720 header, the 802.3 with no LLC header encapsulation of IPX, and the IPX
721 etype in a SNAP frame.]
722 .IP "\fBdecnet src \fIhost\fR"
723 True if the DECNET source address is
725 which may be an address of the form ``10.123'', or a DECNET host
727 [DECNET host name support is only available on Ultrix systems
728 that are configured to run DECNET.]
729 .IP "\fBdecnet dst \fIhost\fR"
730 True if the DECNET destination address is
732 .IP "\fBdecnet host \fIhost\fR"
733 True if either the DECNET source or destination address is
735 .IP "\fBip\fR, \fBip6\fR, \fBarp\fR, \fBrarp\fR, \fBatalk\fR, \fBaarp\fR, \fBdecnet\fR, \fBiso\fR, \fBstp\fR, \fBipx\fR, \fInetbeui\fP"
739 \fBether proto \fIp\fR
742 where \fIp\fR is one of the above protocols.
743 .IP "\fBlat\fR, \fBmoprc\fR, \fBmopdl\fR"
747 \fBether proto \fIp\fR
750 where \fIp\fR is one of the above protocols.
752 \fItcpdump\fP does not currently know how to parse these protocols.
753 .IP "\fBvlan \fI[vlan_id]\fR"
754 True if the packet is an IEEE 802.1Q VLAN packet.
755 If \fI[vlan_id]\fR is specified, only true is the packet has the specified
757 Note that the first \fBvlan\fR keyword encountered in \fIexpression\fR
758 changes the decoding offsets for the remainder of \fIexpression\fR
759 on the assumption that the packet is a VLAN packet.
760 .IP "\fBtcp\fR, \fBudp\fR, \fBicmp\fR"
764 \fBip proto \fIp\fR\fB or ip6 proto \fIp\fR
767 where \fIp\fR is one of the above protocols.
768 .IP "\fBiso proto \fIprotocol\fR"
769 True if the packet is an OSI packet of protocol type \fIprotocol\fP.
770 \fIProtocol\fP can be a number or one of the names
771 \fIclnp\fP, \fIesis\fP, or \fIisis\fP.
772 .IP "\fBclnp\fR, \fBesis\fR, \fBisis\fR"
779 where \fIp\fR is one of the above protocols.
780 Note that \fItcpdump\fR does an incomplete job of parsing these protocols.
781 .IP "\fIexpr relop expr\fR"
782 True if the relation holds, where \fIrelop\fR is one of >, <, >=, <=, =, !=,
783 and \fIexpr\fR is an arithmetic expression composed of integer constants
784 (expressed in standard C syntax), the normal binary operators
785 [+, -, *, /, &, |], a length operator, and special packet data accessors.
787 data inside the packet, use the following syntax:
790 \fIproto\fB [ \fIexpr\fB : \fIsize\fB ]\fR
793 \fIProto\fR is one of \fBether, fddi, tr, ppp, slip, link,
794 ip, arp, rarp, tcp, udp, icmp\fR or \fBip6\fR, and
795 indicates the protocol layer for the index operation.
796 (\fBether, fddi, tr, ppp, slip\fR and \fBlink\fR all refer to the link
798 Note that \fItcp, udp\fR and other upper-layer protocol types only
799 apply to IPv4, not IPv6 (this will be fixed in the future).
800 The byte offset, relative to the indicated protocol layer, is
802 \fISize\fR is optional and indicates the number of bytes in the
803 field of interest; it can be either one, two, or four, and defaults to one.
804 The length operator, indicated by the keyword \fBlen\fP, gives the
805 length of the packet.
807 For example, `\fBether[0] & 1 != 0\fP' catches all multicast traffic.
808 The expression `\fBip[0] & 0xf != 5\fP'
809 catches all IP packets with options.
811 `\fBip[6:2] & 0x1fff = 0\fP'
812 catches only unfragmented datagrams and frag zero of fragmented datagrams.
813 This check is implicitly applied to the \fBtcp\fP and \fBudp\fP
815 For instance, \fBtcp[0]\fP always means the first
816 byte of the TCP \fIheader\fP, and never means the first byte of an
817 intervening fragment.
819 Some offsets and field values may be expressed as names rather than
821 The following protocol header field offsets are
822 available: \fBicmptype\fP (ICMP type field), \fBicmpcode\fP (ICMP
823 code field), and \fBtcpflags\fP (TCP flags field).
825 The following ICMP type field values are available: \fBicmp-echoreply\fP,
826 \fBicmp-unreach\fP, \fBicmp-sourcequench\fP, \fBicmp-redirect\fP,
827 \fBicmp-echo\fP, \fBicmp-routeradvert\fP, \fBicmp-routersolicit\fP,
828 \fBicmp-timxceed\fP, \fBicmp-paramprob\fP, \fBicmp-tstamp\fP,
829 \fBicmp-tstampreply\fP, \fBicmp-ireq\fP, \fBicmp-ireqreply\fP,
830 \fBicmp-maskreq\fP, \fBicmp-maskreply\fP.
832 The following TCP flags field values are available: \fBtcp-fin\fP,
833 \fBtcp-syn\fP, \fBtcp-rst\fP, \fBtcp-push\fP, \fBtcp-push\fP,
834 \fBtcp-ack\fP, \fBtcp-urg\fP.
836 Primitives may be combined using:
838 A parenthesized group of primitives and operators
839 (parentheses are special to the Shell and must be escaped).
841 Negation (`\fB!\fP' or `\fBnot\fP').
843 Concatenation (`\fB&&\fP' or `\fBand\fP').
845 Alternation (`\fB||\fP' or `\fBor\fP').
847 Negation has highest precedence.
848 Alternation and concatenation have equal precedence and associate
850 Note that explicit \fBand\fR tokens, not juxtaposition,
851 are now required for concatenation.
853 If an identifier is given without a keyword, the most recent keyword
858 \fBnot host vs and ace\fR
864 \fBnot host vs and host ace\fR
867 which should not be confused with
870 \fBnot ( host vs or ace )\fR
874 Expression arguments can be passed to \fItcpdump\fP as either a single
875 argument or as multiple arguments, whichever is more convenient.
876 Generally, if the expression contains Shell metacharacters, it is
877 easier to pass it as a single, quoted argument.
878 Multiple arguments are concatenated with spaces before being parsed.
881 To print all packets arriving at or departing from \fIsundown\fP:
884 \fBtcpdump host sundown\fP
888 To print traffic between \fIhelios\fR and either \fIhot\fR or \fIace\fR:
891 \fBtcpdump host helios and \\( hot or ace \\)\fP
895 To print all IP packets between \fIace\fR and any host except \fIhelios\fR:
898 \fBtcpdump ip host ace and not helios\fP
902 To print all traffic between local hosts and hosts at Berkeley:
906 tcpdump net ucb-ether
910 To print all ftp traffic through internet gateway \fIsnup\fP:
911 (note that the expression is quoted to prevent the shell from
912 (mis-)interpreting the parentheses):
916 tcpdump 'gateway snup and (port ftp or ftp-data)'
920 To print traffic neither sourced from nor destined for local hosts
921 (if you gateway to one other net, this stuff should never make it
922 onto your local net).
926 tcpdump ip and not net \fIlocalnet\fP
930 To print the start and end packets (the SYN and FIN packets) of each
931 TCP conversation that involves a non-local host.
935 tcpdump 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0 and not src and dst net \fIlocalnet\fP'
939 To print IP packets longer than 576 bytes sent through gateway \fIsnup\fP:
943 tcpdump 'gateway snup and ip[2:2] > 576'
947 To print IP broadcast or multicast packets that were
949 sent via ethernet broadcast or multicast:
953 tcpdump 'ether[0] & 1 = 0 and ip[16] >= 224'
957 To print all ICMP packets that are not echo requests/replies (i.e., not
962 tcpdump 'icmp[icmptype] != icmp-echo and icmp[icmptype] != icmp-echoreply'
967 The output of \fItcpdump\fP is protocol dependent.
969 gives a brief description and examples of most of the formats.
977 If the '-e' option is given, the link level header is printed out.
978 On ethernets, the source and destination addresses, protocol,
979 and packet length are printed.
981 On FDDI networks, the '-e' option causes \fItcpdump\fP to print
982 the `frame control' field, the source and destination addresses,
983 and the packet length.
984 (The `frame control' field governs the
985 interpretation of the rest of the packet.
987 as those containing IP datagrams) are `async' packets, with a priority
988 value between 0 and 7; for example, `\fBasync4\fR'.
990 are assumed to contain an 802.2 Logical Link Control (LLC) packet;
991 the LLC header is printed if it is \fInot\fR an ISO datagram or a
992 so-called SNAP packet.
994 On Token Ring networks, the '-e' option causes \fItcpdump\fP to print
995 the `access control' and `frame control' fields, the source and
996 destination addresses, and the packet length.
998 packets are assumed to contain an LLC packet.
999 Regardless of whether
1000 the '-e' option is specified or not, the source routing information is
1001 printed for source-routed packets.
1003 \fI(N.B.: The following description assumes familiarity with
1004 the SLIP compression algorithm described in RFC-1144.)\fP
1006 On SLIP links, a direction indicator (``I'' for inbound, ``O'' for outbound),
1007 packet type, and compression information are printed out.
1008 The packet type is printed first.
1009 The three types are \fIip\fP, \fIutcp\fP, and \fIctcp\fP.
1010 No further link information is printed for \fIip\fR packets.
1011 For TCP packets, the connection identifier is printed following the type.
1012 If the packet is compressed, its encoded header is printed out.
1013 The special cases are printed out as
1014 \fB*S+\fIn\fR and \fB*SA+\fIn\fR, where \fIn\fR is the amount by which
1015 the sequence number (or sequence number and ack) has changed.
1016 If it is not a special case,
1017 zero or more changes are printed.
1018 A change is indicated by U (urgent pointer), W (window), A (ack),
1019 S (sequence number), and I (packet ID), followed by a delta (+n or -n),
1020 or a new value (=n).
1021 Finally, the amount of data in the packet and compressed header length
1024 For example, the following line shows an outbound compressed TCP packet,
1025 with an implicit connection identifier; the ack has changed by 6,
1026 the sequence number by 49, and the packet ID by 6; there are 3 bytes of
1027 data and 6 bytes of compressed header:
1030 \fBO ctcp * A+6 S+49 I+6 3 (6)\fP
1036 Arp/rarp output shows the type of request and its arguments.
1038 format is intended to be self explanatory.
1039 Here is a short sample taken from the start of an `rlogin' from
1040 host \fIrtsg\fP to host \fIcsam\fP:
1044 \f(CWarp who-has csam tell rtsg
1045 arp reply csam is-at CSAM\fR
1049 The first line says that rtsg sent an arp packet asking
1050 for the ethernet address of internet host csam.
1052 replies with its ethernet address (in this example, ethernet addresses
1053 are in caps and internet addresses in lower case).
1055 This would look less redundant if we had done \fItcpdump \-n\fP:
1059 \f(CWarp who-has 128.3.254.6 tell 128.3.254.68
1060 arp reply 128.3.254.6 is-at 02:07:01:00:01:c4\fP
1064 If we had done \fItcpdump \-e\fP, the fact that the first packet is
1065 broadcast and the second is point-to-point would be visible:
1069 \f(CWRTSG Broadcast 0806 64: arp who-has csam tell rtsg
1070 CSAM RTSG 0806 64: arp reply csam is-at CSAM\fR
1074 For the first packet this says the ethernet source address is RTSG, the
1075 destination is the ethernet broadcast address, the type field
1076 contained hex 0806 (type ETHER_ARP) and the total length was 64 bytes.
1080 \fI(N.B.:The following description assumes familiarity with
1081 the TCP protocol described in RFC-793.
1082 If you are not familiar
1083 with the protocol, neither this description nor \fItcpdump\fP will
1084 be of much use to you.)\fP
1086 The general format of a tcp protocol line is:
1090 \fIsrc > dst: flags data-seqno ack window urgent options\fP
1094 \fISrc\fP and \fIdst\fP are the source and destination IP
1095 addresses and ports.
1096 \fIFlags\fP are some combination of S (SYN),
1097 F (FIN), P (PUSH) or R (RST) or a single `.' (no flags).
1098 \fIData-seqno\fP describes the portion of sequence space covered
1099 by the data in this packet (see example below).
1100 \fIAck\fP is sequence number of the next data expected the other
1101 direction on this connection.
1102 \fIWindow\fP is the number of bytes of receive buffer space available
1103 the other direction on this connection.
1104 \fIUrg\fP indicates there is `urgent' data in the packet.
1105 \fIOptions\fP are tcp options enclosed in angle brackets (e.g., <mss 1024>).
1107 \fISrc, dst\fP and \fIflags\fP are always present.
1109 depend on the contents of the packet's tcp protocol header and
1110 are output only if appropriate.
1112 Here is the opening portion of an rlogin from host \fIrtsg\fP to
1117 \s-2\f(CWrtsg.1023 > csam.login: S 768512:768512(0) win 4096 <mss 1024>
1118 csam.login > rtsg.1023: S 947648:947648(0) ack 768513 win 4096 <mss 1024>
1119 rtsg.1023 > csam.login: . ack 1 win 4096
1120 rtsg.1023 > csam.login: P 1:2(1) ack 1 win 4096
1121 csam.login > rtsg.1023: . ack 2 win 4096
1122 rtsg.1023 > csam.login: P 2:21(19) ack 1 win 4096
1123 csam.login > rtsg.1023: P 1:2(1) ack 21 win 4077
1124 csam.login > rtsg.1023: P 2:3(1) ack 21 win 4077 urg 1
1125 csam.login > rtsg.1023: P 3:4(1) ack 21 win 4077 urg 1\fR\s+2
1129 The first line says that tcp port 1023 on rtsg sent a packet
1132 The \fBS\fP indicates that the \fISYN\fP flag was set.
1133 The packet sequence number was 768512 and it contained no data.
1134 (The notation is `first:last(nbytes)' which means `sequence
1136 up to but not including \fIlast\fP which is \fInbytes\fP bytes of user data'.)
1137 There was no piggy-backed ack, the available receive window was 4096
1138 bytes and there was a max-segment-size option requesting an mss of
1141 Csam replies with a similar packet except it includes a piggy-backed
1143 Rtsg then acks csam's SYN.
1146 The packet contained no data so there is no data sequence number.
1147 Note that the ack sequence
1148 number is a small integer (1).
1149 The first time \fItcpdump\fP sees a
1150 tcp `conversation', it prints the sequence number from the packet.
1151 On subsequent packets of the conversation, the difference between
1152 the current packet's sequence number and this initial sequence number
1154 This means that sequence numbers after the
1155 first can be interpreted
1156 as relative byte positions in the conversation's data stream (with the
1157 first data byte each direction being `1').
1158 `-S' will override this
1159 feature, causing the original sequence numbers to be output.
1161 On the 6th line, rtsg sends csam 19 bytes of data (bytes 2 through 20
1162 in the rtsg \(-> csam side of the conversation).
1163 The PUSH flag is set in the packet.
1164 On the 7th line, csam says it's received data sent by rtsg up to
1165 but not including byte 21.
1166 Most of this data is apparently sitting in the
1167 socket buffer since csam's receive window has gotten 19 bytes smaller.
1168 Csam also sends one byte of data to rtsg in this packet.
1169 On the 8th and 9th lines,
1170 csam sends two bytes of urgent, pushed data to rtsg.
1172 If the snapshot was small enough that \fItcpdump\fP didn't capture
1173 the full TCP header, it interprets as much of the header as it can
1174 and then reports ``[|\fItcp\fP]'' to indicate the remainder could not
1176 If the header contains a bogus option (one with a length
1177 that's either too small or beyond the end of the header), \fItcpdump\fP
1178 reports it as ``[\fIbad opt\fP]'' and does not interpret any further
1179 options (since it's impossible to tell where they start).
1181 length indicates options are present but the IP datagram length is not
1182 long enough for the options to actually be there, \fItcpdump\fP reports
1183 it as ``[\fIbad hdr length\fP]''.
1185 .B Capturing TCP packets with particular flag combinations (SYN-ACK, URG-ACK, etc.)
1187 There are 8 bits in the control bits section of the TCP header:
1189 .I CWR | ECE | URG | ACK | PSH | RST | SYN | FIN
1191 Let's assume that we want to watch packets used in establishing
1193 Recall that TCP uses a 3-way handshake protocol
1194 when it initializes a new connection; the connection sequence with
1195 regard to the TCP control bits is
1201 2) Recipient responds with SYN, ACK
1207 Now we're interested in capturing packets that have only the
1208 SYN bit set (Step 1).
1209 Note that we don't want packets from step 2
1210 (SYN-ACK), just a plain initial SYN.
1211 What we need is a correct filter
1212 expression for \fItcpdump\fP.
1214 Recall the structure of a TCP header without options:
1218 -----------------------------------------------------------------
1219 | source port | destination port |
1220 -----------------------------------------------------------------
1222 -----------------------------------------------------------------
1223 | acknowledgment number |
1224 -----------------------------------------------------------------
1225 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
1226 -----------------------------------------------------------------
1227 | TCP checksum | urgent pointer |
1228 -----------------------------------------------------------------
1231 A TCP header usually holds 20 octets of data, unless options are
1233 The first line of the graph contains octets 0 - 3, the
1234 second line shows octets 4 - 7 etc.
1236 Starting to count with 0, the relevant TCP control bits are contained
1241 ----------------|---------------|---------------|----------------
1242 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
1243 ----------------|---------------|---------------|----------------
1244 | | 13th octet | | |
1247 Let's have a closer look at octet no. 13:
1257 These are the TCP control bits we are interested
1259 We have numbered the bits in this octet from 0 to 7, right to
1260 left, so the PSH bit is bit number 3, while the URG bit is number 5.
1262 Recall that we want to capture packets with only SYN set.
1263 Let's see what happens to octet 13 if a TCP datagram arrives
1264 with the SYN bit set in its header:
1275 control bits section we see that only bit number 1 (SYN) is set.
1277 Assuming that octet number 13 is an 8-bit unsigned integer in
1278 network byte order, the binary value of this octet is
1282 and its decimal representation is
1286 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 1*2 + 0*2 = 2
1289 We're almost done, because now we know that if only SYN is set,
1290 the value of the 13th octet in the TCP header, when interpreted
1291 as a 8-bit unsigned integer in network byte order, must be exactly 2.
1293 This relationship can be expressed as
1299 We can use this expression as the filter for \fItcpdump\fP in order
1300 to watch packets which have only SYN set:
1303 tcpdump -i xl0 tcp[13] == 2
1306 The expression says "let the 13th octet of a TCP datagram have
1307 the decimal value 2", which is exactly what we want.
1309 Now, let's assume that we need to capture SYN packets, but we
1310 don't care if ACK or any other TCP control bit is set at the
1312 Let's see what happens to octet 13 when a TCP datagram
1313 with SYN-ACK set arrives:
1323 Now bits 1 and 4 are set in the 13th octet.
1329 which translates to decimal
1333 0*2 + 0*2 + 0*2 + 1*2 + 0*2 + 0*2 + 1*2 + 0*2 = 18
1336 Now we can't just use 'tcp[13] == 18' in the \fItcpdump\fP filter
1337 expression, because that would select only those packets that have
1338 SYN-ACK set, but not those with only SYN set.
1339 Remember that we don't care
1340 if ACK or any other control bit is set as long as SYN is set.
1342 In order to achieve our goal, we need to logically AND the
1343 binary value of octet 13 with some other value to preserve
1345 We know that we want SYN to be set in any case,
1346 so we'll logically AND the value in the 13th octet with
1347 the binary value of a SYN:
1351 00010010 SYN-ACK 00000010 SYN
1352 AND 00000010 (we want SYN) AND 00000010 (we want SYN)
1354 = 00000010 = 00000010
1357 We see that this AND operation delivers the same result
1358 regardless whether ACK or another TCP control bit is set.
1359 The decimal representation of the AND value as well as
1360 the result of this operation is 2 (binary 00000010),
1361 so we know that for packets with SYN set the following
1362 relation must hold true:
1364 ( ( value of octet 13 ) AND ( 2 ) ) == ( 2 )
1366 This points us to the \fItcpdump\fP filter expression
1369 tcpdump -i xl0 'tcp[13] & 2 == 2'
1372 Note that you should use single quotes or a backslash
1373 in the expression to hide the AND ('&') special character
1379 UDP format is illustrated by this rwho packet:
1383 \f(CWactinide.who > broadcast.who: udp 84\fP
1387 This says that port \fIwho\fP on host \fIactinide\fP sent a udp
1388 datagram to port \fIwho\fP on host \fIbroadcast\fP, the Internet
1390 The packet contained 84 bytes of user data.
1392 Some UDP services are recognized (from the source or destination
1393 port number) and the higher level protocol information printed.
1394 In particular, Domain Name service requests (RFC-1034/1035) and Sun
1395 RPC calls (RFC-1050) to NFS.
1397 UDP Name Server Requests
1399 \fI(N.B.:The following description assumes familiarity with
1400 the Domain Service protocol described in RFC-1035.
1401 If you are not familiar
1402 with the protocol, the following description will appear to be written
1405 Name server requests are formatted as
1409 \fIsrc > dst: id op? flags qtype qclass name (len)\fP
1411 \f(CWh2opolo.1538 > helios.domain: 3+ A? ucbvax.berkeley.edu. (37)\fR
1415 Host \fIh2opolo\fP asked the domain server on \fIhelios\fP for an
1416 address record (qtype=A) associated with the name \fIucbvax.berkeley.edu.\fP
1417 The query id was `3'.
1418 The `+' indicates the \fIrecursion desired\fP flag
1420 The query length was 37 bytes, not including the UDP and
1421 IP protocol headers.
1422 The query operation was the normal one, \fIQuery\fP,
1423 so the op field was omitted.
1424 If the op had been anything else, it would
1425 have been printed between the `3' and the `+'.
1426 Similarly, the qclass was the normal one,
1427 \fIC_IN\fP, and omitted.
1428 Any other qclass would have been printed
1429 immediately after the `A'.
1431 A few anomalies are checked and may result in extra fields enclosed in
1432 square brackets: If a query contains an answer, authority records or
1433 additional records section,
1438 are printed as `[\fIn\fPa]', `[\fIn\fPn]' or `[\fIn\fPau]' where \fIn\fP
1439 is the appropriate count.
1440 If any of the response bits are set (AA, RA or rcode) or any of the
1441 `must be zero' bits are set in bytes two and three, `[b2&3=\fIx\fP]'
1442 is printed, where \fIx\fP is the hex value of header bytes two and three.
1444 UDP Name Server Responses
1446 Name server responses are formatted as
1450 \fIsrc > dst: id op rcode flags a/n/au type class data (len)\fP
1452 \f(CWhelios.domain > h2opolo.1538: 3 3/3/7 A 128.32.137.3 (273)
1453 helios.domain > h2opolo.1537: 2 NXDomain* 0/1/0 (97)\fR
1457 In the first example, \fIhelios\fP responds to query id 3 from \fIh2opolo\fP
1458 with 3 answer records, 3 name server records and 7 additional records.
1459 The first answer record is type A (address) and its data is internet
1460 address 128.32.137.3.
1461 The total size of the response was 273 bytes,
1462 excluding UDP and IP headers.
1463 The op (Query) and response code
1464 (NoError) were omitted, as was the class (C_IN) of the A record.
1466 In the second example, \fIhelios\fP responds to query 2 with a
1467 response code of non-existent domain (NXDomain) with no answers,
1468 one name server and no authority records.
1469 The `*' indicates that
1470 the \fIauthoritative answer\fP bit was set.
1472 answers, no type, class or data were printed.
1474 Other flag characters that might appear are `\-' (recursion available,
1475 RA, \fInot\fP set) and `|' (truncated message, TC, set).
1477 `question' section doesn't contain exactly one entry, `[\fIn\fPq]'
1480 Note that name server requests and responses tend to be large and the
1481 default \fIsnaplen\fP of 68 bytes may not capture enough of the packet
1483 Use the \fB\-s\fP flag to increase the snaplen if you
1484 need to seriously investigate name server traffic.
1486 has worked well for me.
1491 \fItcpdump\fP now includes fairly extensive SMB/CIFS/NBT decoding for data
1492 on UDP/137, UDP/138 and TCP/139.
1493 Some primitive decoding of IPX and
1494 NetBEUI SMB data is also done.
1496 By default a fairly minimal decode is done, with a much more detailed
1497 decode done if -v is used.
1498 Be warned that with -v a single SMB packet
1499 may take up a page or more, so only use -v if you really want all the
1502 If you are decoding SMB sessions containing unicode strings then you
1503 may wish to set the environment variable USE_UNICODE to 1.
1505 auto-detect unicode srings would be welcome.
1507 For information on SMB packet formats and what all te fields mean see
1508 www.cifs.org or the pub/samba/specs/ directory on your favourite
1509 samba.org mirror site.
1510 The SMB patches were written by Andrew Tridgell
1514 NFS Requests and Replies
1516 Sun NFS (Network File System) requests and replies are printed as:
1520 \fIsrc.xid > dst.nfs: len op args\fP
1521 \fIsrc.nfs > dst.xid: reply stat len op results\fP
1524 sushi.6709 > wrl.nfs: 112 readlink fh 21,24/10.73165
1525 wrl.nfs > sushi.6709: reply ok 40 readlink "../var"
1526 sushi.201b > wrl.nfs:
1527 144 lookup fh 9,74/4096.6878 "xcolors"
1528 wrl.nfs > sushi.201b:
1529 reply ok 128 lookup fh 9,74/4134.3150
1534 In the first line, host \fIsushi\fP sends a transaction with id \fI6709\fP
1535 to \fIwrl\fP (note that the number following the src host is a
1536 transaction id, \fInot\fP the source port).
1537 The request was 112 bytes,
1538 excluding the UDP and IP headers.
1539 The operation was a \fIreadlink\fP
1540 (read symbolic link) on file handle (\fIfh\fP) 21,24/10.731657119.
1541 (If one is lucky, as in this case, the file handle can be interpreted
1542 as a major,minor device number pair, followed by the inode number and
1544 \fIWrl\fP replies `ok' with the contents of the link.
1546 In the third line, \fIsushi\fP asks \fIwrl\fP to lookup the name
1547 `\fIxcolors\fP' in directory file 9,74/4096.6878.
1548 Note that the data printed
1549 depends on the operation type.
1550 The format is intended to be self
1551 explanatory if read in conjunction with
1552 an NFS protocol spec.
1554 If the \-v (verbose) flag is given, additional information is printed.
1560 sushi.1372a > wrl.nfs:
1561 148 read fh 21,11/12.195 8192 bytes @ 24576
1562 wrl.nfs > sushi.1372a:
1563 reply ok 1472 read REG 100664 ids 417/0 sz 29388
1568 (\-v also prints the IP header TTL, ID, length, and fragmentation fields,
1569 which have been omitted from this example.) In the first line,
1570 \fIsushi\fP asks \fIwrl\fP to read 8192 bytes from file 21,11/12.195,
1571 at byte offset 24576.
1572 \fIWrl\fP replies `ok'; the packet shown on the
1573 second line is the first fragment of the reply, and hence is only 1472
1574 bytes long (the other bytes will follow in subsequent fragments, but
1575 these fragments do not have NFS or even UDP headers and so might not be
1576 printed, depending on the filter expression used).
1577 Because the \-v flag
1578 is given, some of the file attributes (which are returned in addition
1579 to the file data) are printed: the file type (``REG'', for regular file),
1580 the file mode (in octal), the uid and gid, and the file size.
1582 If the \-v flag is given more than once, even more details are printed.
1584 Note that NFS requests are very large and much of the detail won't be printed
1585 unless \fIsnaplen\fP is increased.
1586 Try using `\fB\-s 192\fP' to watch
1589 NFS reply packets do not explicitly identify the RPC operation.
1591 \fItcpdump\fP keeps track of ``recent'' requests, and matches them to the
1592 replies using the transaction ID.
1593 If a reply does not closely follow the
1594 corresponding request, it might not be parsable.
1596 AFS Requests and Replies
1598 Transarc AFS (Andrew File System) requests and replies are printed
1604 \fIsrc.sport > dst.dport: rx packet-type\fP
1605 \fIsrc.sport > dst.dport: rx packet-type service call call-name args\fP
1606 \fIsrc.sport > dst.dport: rx packet-type service reply call-name args\fP
1609 elvis.7001 > pike.afsfs:
1610 rx data fs call rename old fid 536876964/1/1 ".newsrc.new"
1611 new fid 536876964/1/1 ".newsrc"
1612 pike.afsfs > elvis.7001: rx data fs reply rename
1617 In the first line, host elvis sends a RX packet to pike.
1619 a RX data packet to the fs (fileserver) service, and is the start of
1621 The RPC call was a rename, with the old directory file id
1622 of 536876964/1/1 and an old filename of `.newsrc.new', and a new directory
1623 file id of 536876964/1/1 and a new filename of `.newsrc'.
1625 responds with a RPC reply to the rename call (which was successful, because
1626 it was a data packet and not an abort packet).
1628 In general, all AFS RPCs are decoded at least by RPC call name.
1630 AFS RPCs have at least some of the arguments decoded (generally only
1631 the `interesting' arguments, for some definition of interesting).
1633 The format is intended to be self-describing, but it will probably
1634 not be useful to people who are not familiar with the workings of
1637 If the -v (verbose) flag is given twice, acknowledgement packets and
1638 additional header information is printed, such as the the RX call ID,
1639 call number, sequence number, serial number, and the RX packet flags.
1641 If the -v flag is given twice, additional information is printed,
1642 such as the the RX call ID, serial number, and the RX packet flags.
1643 The MTU negotiation information is also printed from RX ack packets.
1645 If the -v flag is given three times, the security index and service id
1648 Error codes are printed for abort packets, with the exception of Ubik
1649 beacon packets (because abort packets are used to signify a yes vote
1650 for the Ubik protocol).
1652 Note that AFS requests are very large and many of the arguments won't
1653 be printed unless \fIsnaplen\fP is increased.
1654 Try using `\fB-s 256\fP'
1655 to watch AFS traffic.
1657 AFS reply packets do not explicitly identify the RPC operation.
1659 \fItcpdump\fP keeps track of ``recent'' requests, and matches them to the
1660 replies using the call number and service ID.
1661 If a reply does not closely
1663 corresponding request, it might not be parsable.
1666 KIP Appletalk (DDP in UDP)
1668 Appletalk DDP packets encapsulated in UDP datagrams are de-encapsulated
1669 and dumped as DDP packets (i.e., all the UDP header information is
1673 is used to translate appletalk net and node numbers to names.
1674 Lines in this file have the form
1686 The first two lines give the names of appletalk networks.
1688 line gives the name of a particular host (a host is distinguished
1689 from a net by the 3rd octet in the number \-
1690 a net number \fImust\fP have two octets and a host number \fImust\fP
1691 have three octets.) The number and name should be separated by
1692 whitespace (blanks or tabs).
1695 file may contain blank lines or comment lines (lines starting with
1698 Appletalk addresses are printed in the form
1704 \f(CW144.1.209.2 > icsd-net.112.220
1705 office.2 > icsd-net.112.220
1706 jssmag.149.235 > icsd-net.2\fR
1712 doesn't exist or doesn't contain an entry for some appletalk
1713 host/net number, addresses are printed in numeric form.)
1714 In the first example, NBP (DDP port 2) on net 144.1 node 209
1715 is sending to whatever is listening on port 220 of net icsd node 112.
1716 The second line is the same except the full name of the source node
1717 is known (`office').
1718 The third line is a send from port 235 on
1719 net jssmag node 149 to broadcast on the icsd-net NBP port (note that
1720 the broadcast address (255) is indicated by a net name with no host
1721 number \- for this reason it's a good idea to keep node names and
1722 net names distinct in /etc/atalk.names).
1724 NBP (name binding protocol) and ATP (Appletalk transaction protocol)
1725 packets have their contents interpreted.
1726 Other protocols just dump
1727 the protocol name (or number if no name is registered for the
1728 protocol) and packet size.
1730 \fBNBP packets\fP are formatted like the following examples:
1734 \s-2\f(CWicsd-net.112.220 > jssmag.2: nbp-lkup 190: "=:LaserWriter@*"
1735 jssmag.209.2 > icsd-net.112.220: nbp-reply 190: "RM1140:LaserWriter@*" 250
1736 techpit.2 > icsd-net.112.220: nbp-reply 190: "techpit:LaserWriter@*" 186\fR\s+2
1740 The first line is a name lookup request for laserwriters sent by net icsd host
1741 112 and broadcast on net jssmag.
1742 The nbp id for the lookup is 190.
1743 The second line shows a reply for this request (note that it has the
1744 same id) from host jssmag.209 saying that it has a laserwriter
1745 resource named "RM1140" registered on port 250.
1747 another reply to the same request saying host techpit has laserwriter
1748 "techpit" registered on port 186.
1750 \fBATP packet\fP formatting is demonstrated by the following example:
1754 \s-2\f(CWjssmag.209.165 > helios.132: atp-req 12266<0-7> 0xae030001
1755 helios.132 > jssmag.209.165: atp-resp 12266:0 (512) 0xae040000
1756 helios.132 > jssmag.209.165: atp-resp 12266:1 (512) 0xae040000
1757 helios.132 > jssmag.209.165: atp-resp 12266:2 (512) 0xae040000
1758 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1759 helios.132 > jssmag.209.165: atp-resp 12266:4 (512) 0xae040000
1760 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1761 helios.132 > jssmag.209.165: atp-resp 12266:6 (512) 0xae040000
1762 helios.132 > jssmag.209.165: atp-resp*12266:7 (512) 0xae040000
1763 jssmag.209.165 > helios.132: atp-req 12266<3,5> 0xae030001
1764 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1765 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1766 jssmag.209.165 > helios.132: atp-rel 12266<0-7> 0xae030001
1767 jssmag.209.133 > helios.132: atp-req* 12267<0-7> 0xae030002\fR\s+2
1771 Jssmag.209 initiates transaction id 12266 with host helios by requesting
1772 up to 8 packets (the `<0-7>').
1773 The hex number at the end of the line
1774 is the value of the `userdata' field in the request.
1776 Helios responds with 8 512-byte packets.
1777 The `:digit' following the
1778 transaction id gives the packet sequence number in the transaction
1779 and the number in parens is the amount of data in the packet,
1780 excluding the atp header.
1781 The `*' on packet 7 indicates that the
1784 Jssmag.209 then requests that packets 3 & 5 be retransmitted.
1786 resends them then jssmag.209 releases the transaction.
1788 jssmag.209 initiates the next request.
1789 The `*' on the request
1790 indicates that XO (`exactly once') was \fInot\fP set.
1795 Fragmented Internet datagrams are printed as
1799 \fB(frag \fIid\fB:\fIsize\fB@\fIoffset\fB+)\fR
1800 \fB(frag \fIid\fB:\fIsize\fB@\fIoffset\fB)\fR
1804 (The first form indicates there are more fragments.
1806 indicates this is the last fragment.)
1808 \fIId\fP is the fragment id.
1809 \fISize\fP is the fragment
1810 size (in bytes) excluding the IP header.
1811 \fIOffset\fP is this
1812 fragment's offset (in bytes) in the original datagram.
1814 The fragment information is output for each fragment.
1816 fragment contains the higher level protocol header and the frag
1817 info is printed after the protocol info.
1819 after the first contain no higher level protocol header and the
1820 frag info is printed after the source and destination addresses.
1821 For example, here is part of an ftp from arizona.edu to lbl-rtsg.arpa
1822 over a CSNET connection that doesn't appear to handle 576 byte datagrams:
1826 \s-2\f(CWarizona.ftp-data > rtsg.1170: . 1024:1332(308) ack 1 win 4096 (frag 595a:328@0+)
1827 arizona > rtsg: (frag 595a:204@328)
1828 rtsg.1170 > arizona.ftp-data: . ack 1536 win 2560\fP\s+2
1832 There are a couple of things to note here: First, addresses in the
1833 2nd line don't include port numbers.
1834 This is because the TCP
1835 protocol information is all in the first fragment and we have no idea
1836 what the port or sequence numbers are when we print the later fragments.
1837 Second, the tcp sequence information in the first line is printed as if there
1838 were 308 bytes of user data when, in fact, there are 512 bytes (308 in
1839 the first frag and 204 in the second).
1840 If you are looking for holes
1841 in the sequence space or trying to match up acks
1842 with packets, this can fool you.
1844 A packet with the IP \fIdon't fragment\fP flag is marked with a
1845 trailing \fB(DF)\fP.
1849 By default, all output lines are preceded by a timestamp.
1851 is the current clock time in the form
1857 and is as accurate as the kernel's clock.
1858 The timestamp reflects the time the kernel first saw the packet.
1860 is made to account for the time lag between when the
1861 ethernet interface removed the packet from the wire and when the kernel
1862 serviced the `new packet' interrupt.
1864 traffic(1C), nit(4P), bpf(4), pcap(3)
1866 The original authors are:
1870 Steven McCanne, all of the
1871 Lawrence Berkeley National Laboratory, University of California, Berkeley, CA.
1873 It is currently being maintained by tcpdump.org.
1875 The current version is available via http:
1878 .I http://www.tcpdump.org/
1881 The original distribution is available via anonymous ftp:
1884 .I ftp://ftp.ee.lbl.gov/tcpdump.tar.Z
1887 IPv6/IPsec support is added by WIDE/KAME project.
1888 This program uses Eric Young's SSLeay library, under specific configuration.
1890 Please send problems, bugs, questions, desirable enhancements, etc. to:
1893 tcpdump-workers@tcpdump.org
1896 Please send source code contributions, etc. to:
1902 NIT doesn't let you watch your own outbound traffic, BPF will.
1903 We recommend that you use the latter.
1905 On Linux systems with 2.0[.x] kernels:
1907 packets on the loopback device will be seen twice;
1909 packet filtering cannot be done in the kernel, so that all packets must
1910 be copied from the kernel in order to be filtered in user mode;
1912 all of a packet, not just the part that's within the snapshot length,
1913 will be copied from the kernel (the 2.0[.x] packet capture mechanism, if
1914 asked to copy only part of a packet to userland, will not report the
1915 true length of the packet; this would cause most IP packets to get an
1919 capturing on some PPP devices won't work correctly.
1921 We recommend that you upgrade to a 2.2 or later kernel.
1923 Some attempt should be made to reassemble IP fragments or, at least
1924 to compute the right length for the higher level protocol.
1926 Name server inverse queries are not dumped correctly: the (empty)
1927 question section is printed rather than real query in the answer
1929 Some believe that inverse queries are themselves a bug and
1930 prefer to fix the program generating them rather than \fItcpdump\fP.
1932 A packet trace that crosses a daylight savings time change will give
1933 skewed time stamps (the time change is ignored).
1935 Filter expressions that manipulate FDDI or Token Ring headers assume
1936 that all FDDI and Token Ring packets are SNAP-encapsulated Ethernet
1938 This is true for IP, ARP, and DECNET Phase IV, but is not true
1939 for protocols such as ISO CLNS.
1940 Therefore, the filter may inadvertently
1941 accept certain packets that do not properly match the filter expression.
1943 Filter expressions on fields other than those that manipulate Token Ring
1944 headers will not correctly handle source-routed Token Ring packets.
1947 should chase header chain, but at this moment it does not.
1948 .BR "ip6 protochain"
1949 is supplied for this behavior.
1951 Arithmetic expression against transport layer headers, like \fBtcp[0]\fP,
1952 does not work against IPv6 packets.
1953 It only looks at IPv4 packets.
projects / tcpdump / blob