1TCPDUMP(8) System Manager's Manual TCPDUMP(8)
2
6 tcpdump - dump traffic on a network
7
9 tcpdump [ -AbdDefhHIJKlLnNOpqStuUvxX# ] [ -B buffer_size ]
10 [ -c count ] [ --count ] [ -C file_size ]
11 [ -E spi@ipaddr algo:secret,... ]
12 [ -F file ] [ -G rotate_seconds ] [ -i interface ]
13 [ --immediate-mode ] [ -j tstamp_type ] [ -m module ]
14 [ -M secret ] [ --number ] [ --print ] [ -Q in|out|inout ]
15 [ -r file ] [ -s snaplen ] [ -T type ] [ --version ]
16 [ -V file ] [ -w file ] [ -W filecount ] [ -y datalinktype ]
17 [ -z postrotate-command ] [ -Z user ]
18 [ --time-stamp-precision=tstamp_precision ]
19 [ --micro ] [ --nano ]
20 [ expression ]
21
23 Tcpdump prints out a description of the contents of packets on a net‐
24 work interface that match the Boolean expression; the description is
25 preceded by a time stamp, printed, by default, as hours, minutes, sec‐
26 onds, and fractions of a second since midnight. It can also be run
27 with the -w flag, which causes it to save the packet data to a file for
28 later analysis, and/or with the -r flag, which causes it to read from a
29 saved packet file rather than to read packets from a network interface.
30 It can also be run with the -V flag, which causes it to read a list of
31 saved packet files. In all cases, only packets that match expression
32 will be processed by tcpdump.
33 Tcpdump will, if not run with the -c flag, continue capturing packets
35 until it is interrupted by a SIGINT signal (generated, for example, by
36 typing your interrupt character, typically control-C) or a SIGTERM sig‐
37 nal (typically generated with the kill(1) command); if run with the -c
38 flag, it will capture packets until it is interrupted by a SIGINT or
39 SIGTERM signal or the specified number of packets have been processed.
40 When tcpdump finishes capturing packets, it will report counts of:
42 packets ``captured'' (this is the number of packets that tcpdump
44 has received and processed);
45 packets ``received by filter'' (the meaning of this depends on
47 the OS on which you're running tcpdump, and possibly on the way
48 the OS was configured - if a filter was specified on the command
49 line, on some OSes it counts packets regardless of whether they
50 were matched by the filter expression and, even if they were
51 matched by the filter expression, regardless of whether tcpdump
52 has read and processed them yet, on other OSes it counts only
53 packets that were matched by the filter expression regardless of
54 whether tcpdump has read and processed them yet, and on other
55 OSes it counts only packets that were matched by the filter ex‐
56 pression and were processed by tcpdump);
57 packets ``dropped by kernel'' (this is the number of packets
59 that were dropped, due to a lack of buffer space, by the packet
60 capture mechanism in the OS on which tcpdump is running, if the
61 OS reports that information to applications; if not, it will be
62 reported as 0).
63 On platforms that support the SIGINFO signal, such as most BSDs (in‐
65 cluding macOS) and Digital/Tru64 UNIX, it will report those counts when
66 it receives a SIGINFO signal (generated, for example, by typing your
67 ``status'' character, typically control-T, although on some platforms,
68 such as macOS, the ``status'' character is not set by default, so you
69 must set it with stty(1) in order to use it) and will continue captur‐
70 ing packets. On platforms that do not support the SIGINFO signal, the
71 same can be achieved by using the SIGUSR1 signal.
72 Using the SIGUSR2 signal along with the -w flag will forcibly flush the
74 packet buffer into the output file.
75 Reading packets from a network interface may require that you have spe‐
77 cial privileges; see the pcap(3PCAP) man page for details. Reading a
78 saved packet file doesn't require special privileges.
79
81 -A Print each packet (minus its link level header) in ASCII. Handy
82 for capturing web pages.
83 -b Print the AS number in BGP packets in ASDOT notation rather than
85 ASPLAIN notation.
86 -B buffer_size
88 --buffer-size=buffer_size
89 Set the operating system capture buffer size to buffer_size, in
90 units of KiB (1024 bytes).
91 -c count
93 Exit after receiving count packets.
94 --count
96 Print only on stderr the packet count when reading capture
97 file(s) instead of parsing/printing the packets. If a filter is
98 specified on the command line, tcpdump counts only packets that
99 were matched by the filter expression.
100 -C file_size
102 Before writing a raw packet to a savefile, check whether the
103 file is currently larger than file_size and, if so, close the
104 current savefile and open a new one. Savefiles after the first
105 savefile will have the name specified with the -w flag, with a
106 number after it, starting at 1 and continuing upward. The units
107 of file_size are millions of bytes (1,000,000 bytes, not
108 1,048,576 bytes).
109 Note that when used with -Z option (enabled by default), privi‐
111 leges are dropped before opening the first savefile.
112 -d Dump the compiled packet-matching code in a human readable form
114 to standard output and stop.
115 Please mind that although code compilation is always DLT-spe‐
117 cific, typically it is impossible (and unnecessary) to specify
118 which DLT to use for the dump because tcpdump uses either the
119 DLT of the input pcap file specified with -r, or the default DLT
120 of the network interface specified with -i, or the particular
121 DLT of the network interface specified with -y and -i respec‐
122 tively. In these cases the dump shows the same exact code that
123 would filter the input file or the network interface without -d.
124 However, when neither -r nor -i is specified, specifying -d pre‐
126 vents tcpdump from guessing a suitable network interface (see
127 -i). In this case the DLT defaults to EN10MB and can be set to
128 another valid value manually with -y.
129 -dd Dump packet-matching code as a C program fragment.
131 -ddd Dump packet-matching code as decimal numbers (preceded with a
133 count).
134 -D
136 --list-interfaces
137 Print the list of the network interfaces available on the system
138 and on which tcpdump can capture packets. For each network in‐
139 terface, a number and an interface name, possibly followed by a
140 text description of the interface, are printed. The interface
141 name or the number can be supplied to the -i flag to specify an
142 interface on which to capture.
143 This can be useful on systems that don't have a command to list
145 them (e.g., Windows systems, or UNIX systems lacking ifconfig
146 -a); the number can be useful on Windows 2000 and later systems,
147 where the interface name is a somewhat complex string.
148 The -D flag will not be supported if tcpdump was built with an
150 older version of libpcap that lacks the pcap_findalldevs(3PCAP)
151 function.
152 -e Print the link-level header on each dump line. This can be
154 used, for example, to print MAC layer addresses for protocols
155 such as Ethernet and IEEE 802.11.
156 -E Use spi@ipaddr algo:secret for decrypting IPsec ESP packets that
158 are addressed to addr and contain Security Parameter Index value
159 spi. This combination may be repeated with comma or newline sep‐
160 aration.
161 Note that setting the secret for IPv4 ESP packets is supported
163 at this time.
164 Algorithms may be des-cbc, 3des-cbc, blowfish-cbc, rc3-cbc,
166 cast128-cbc, or none. The default is des-cbc. The ability to
167 decrypt packets is only present if tcpdump was compiled with
168 cryptography enabled.
169 secret is the ASCII text for ESP secret key. If preceded by 0x,
171 then a hex value will be read.
172 The option assumes RFC2406 ESP, not RFC1827 ESP. The option is
174 only for debugging purposes, and the use of this option with a
175 true `secret' key is discouraged. By presenting IPsec secret
176 key onto command line you make it visible to others, via ps(1)
177 and other occasions.
178 In addition to the above syntax, the syntax file name may be
180 used to have tcpdump read the provided file in. The file is
181 opened upon receiving the first ESP packet, so any special per‐
182 missions that tcpdump may have been given should already have
183 been given up.
184 -f Print `foreign' IPv4 addresses numerically rather than symboli‐
186 cally (this option is intended to get around serious brain dam‐
187 age in Sun's NIS server — usually it hangs forever translating
188 non-local internet numbers).
189 The test for `foreign' IPv4 addresses is done using the IPv4 ad‐
191 dress and netmask of the interface on which capture is being
192 done. If that address or netmask are not available, available,
193 either because the interface on which capture is being done has
194 no address or netmask or because the capture is being done on
195 the Linux "any" interface, which can capture on more than one
196 interface, this option will not work correctly.
197 -F file
199 Use file as input for the filter expression. An additional ex‐
200 pression given on the command line is ignored.
201 -G rotate_seconds
203 If specified, rotates the dump file specified with the -w option
204 every rotate_seconds seconds. Savefiles will have the name
205 specified by -w which should include a time format as defined by
206 strftime(3). If no time format is specified, each new file will
207 overwrite the previous. Whenever a generated filename is not
208 unique, tcpdump will overwrite the pre-existing data; providing
209 a time specification that is coarser than the capture period is
210 therefore not advised.
211 If used in conjunction with the -C option, filenames will take
213 the form of `file<count>'.
214 Note that when used with -Z option (enabled by default), privi‐
216 leges are dropped before opening the first savefile.
217 -h
219 --help Print the tcpdump and libpcap version strings, print a usage
220 message, and exit.
221 --version
223 Print the tcpdump and libpcap version strings and exit.
224 -H Attempt to detect 802.11s draft mesh headers.
226 -i interface
228 --interface=interface
229 Listen, report the list of link-layer types, report the list of
230 time stamp types, or report the results of compiling a filter
231 expression on interface. If unspecified and if the -d flag is
232 not given, tcpdump searches the system interface list for the
233 lowest numbered, configured up interface (excluding loopback),
234 which may turn out to be, for example, ``eth0''.
235 On Linux systems with 2.2 or later kernels, an interface argu‐
237 ment of ``any'' can be used to capture packets from all inter‐
238 faces. Note that captures on the ``any'' device will not be
239 done in promiscuous mode.
240 If the -D flag is supported, an interface number as printed by
242 that flag can be used as the interface argument, if no interface
243 on the system has that number as a name.
244 -I
246 --monitor-mode
247 Put the interface in "monitor mode"; this is supported only on
248 IEEE 802.11 Wi-Fi interfaces, and supported only on some operat‐
249 ing systems.
250 Note that in monitor mode the adapter might disassociate from
252 the network with which it's associated, so that you will not be
253 able to use any wireless networks with that adapter. This could
254 prevent accessing files on a network server, or resolving host
255 names or network addresses, if you are capturing in monitor mode
256 and are not connected to another network with another adapter.
257 This flag will affect the output of the -L flag. If -I isn't
259 specified, only those link-layer types available when not in
260 monitor mode will be shown; if -I is specified, only those link-
261 layer types available when in monitor mode will be shown.
262 --immediate-mode
264 Capture in "immediate mode". In this mode, packets are deliv‐
265 ered to tcpdump as soon as they arrive, rather than being
266 buffered for efficiency. This is the default when printing
267 packets rather than saving packets to a ``savefile'' if the
268 packets are being printed to a terminal rather than to a file or
269 pipe.
270 -j tstamp_type
272 --time-stamp-type=tstamp_type
273 Set the time stamp type for the capture to tstamp_type. The
274 names to use for the time stamp types are given in
275 pcap-tstamp(7); not all the types listed there will necessarily
276 be valid for any given interface.
277 -J
279 --list-time-stamp-types
280 List the supported time stamp types for the interface and exit.
281 If the time stamp type cannot be set for the interface, no time
282 stamp types are listed.
283 --time-stamp-precision=tstamp_precision
285 When capturing, set the time stamp precision for the capture to
286 tstamp_precision. Note that availability of high precision time
287 stamps (nanoseconds) and their actual accuracy is platform and
288 hardware dependent. Also note that when writing captures made
289 with nanosecond accuracy to a savefile, the time stamps are
290 written with nanosecond resolution, and the file is written with
291 a different magic number, to indicate that the time stamps are
292 in seconds and nanoseconds; not all programs that read pcap
293 savefiles will be able to read those captures.
294 When reading a savefile, convert time stamps to the precision
296 specified by timestamp_precision, and display them with that
297 resolution. If the precision specified is less than the preci‐
298 sion of time stamps in the file, the conversion will lose preci‐
299 sion.
300 The supported values for timestamp_precision are micro for mi‐
302 crosecond resolution and nano for nanosecond resolution. The
303 default is microsecond resolution.
304 --micro
306 --nano Shorthands for --time-stamp-precision=micro or --time-stamp-pre‐
307 cision=nano, adjusting the time stamp precision accordingly.
308 When reading packets from a savefile, using --micro truncates
309 time stamps if the savefile was created with nanosecond preci‐
310 sion. In contrast, a savefile created with microsecond preci‐
311 sion will have trailing zeroes added to the time stamp when
312 --nano is used.
313 -K
315 --dont-verify-checksums
316 Don't attempt to verify IP, TCP, or UDP checksums. This is use‐
317 ful for interfaces that perform some or all of those checksum
318 calculation in hardware; otherwise, all outgoing TCP checksums
319 will be flagged as bad.
320 -l Make stdout line buffered. Useful if you want to see the data
322 while capturing it. E.g.,
323 tcpdump -l | tee dat
325 or
327 tcpdump -l > dat & tail -f dat
329 Note that on Windows,``line buffered'' means ``unbuffered'', so
331 that WinDump will write each character individually if -l is
332 specified.
333 -U is similar to -l in its behavior, but it will cause output to
335 be ``packet-buffered'', so that the output is written to stdout
336 at the end of each packet rather than at the end of each line;
337 this is buffered on all platforms, including Windows.
338 -L
340 --list-data-link-types
341 List the known data link types for the interface, in the speci‐
342 fied mode, and exit. The list of known data link types may be
343 dependent on the specified mode; for example, on some platforms,
344 a Wi-Fi interface might support one set of data link types when
345 not in monitor mode (for example, it might support only fake
346 Ethernet headers, or might support 802.11 headers but not sup‐
347 port 802.11 headers with radio information) and another set of
348 data link types when in monitor mode (for example, it might sup‐
349 port 802.11 headers, or 802.11 headers with radio information,
350 only in monitor mode).
351 -m module
353 Load SMI MIB module definitions from file module. This option
354 can be used several times to load several MIB modules into tcp‐
355 dump.
356 -M secret
358 Use secret as a shared secret for validating the digests found
359 in TCP segments with the TCP-MD5 option (RFC 2385), if present.
360 -n Don't convert host addresses to names. This can be used to
362 avoid DNS lookups.
363 -nn Don't convert protocol and port numbers etc. to names either.
365 -N Don't print domain name qualification of host names. E.g., if
367 you give this flag then tcpdump will print ``nic'' instead of
368 ``nic.ddn.mil''.
369 -#
371 --number
372 Print an optional packet number at the beginning of the line.
373 -O
375 --no-optimize
376 Do not run the packet-matching code optimizer. This is useful
377 only if you suspect a bug in the optimizer.
378 -p
380 --no-promiscuous-mode
381 Don't put the interface into promiscuous mode. Note that the
382 interface might be in promiscuous mode for some other reason;
383 hence, `-p' cannot be used as an abbreviation for `ether host
384 {local-hw-addr} or ether broadcast'.
385 --print
387 Print parsed packet output, even if the raw packets are being
388 saved to a file with the -w flag.
389 -Q direction
391 --direction=direction
392 Choose send/receive direction direction for which packets should
393 be captured. Possible values are `in', `out' and `inout'. Not
394 available on all platforms.
395 -q Quick (quiet?) output. Print less protocol information so out‐
397 put lines are shorter.
398 -r file
400 Read packets from file (which was created with the -w option or
401 by other tools that write pcap or pcapng files). Standard input
402 is used if file is ``-''.
403 -S
405 --absolute-tcp-sequence-numbers
406 Print absolute, rather than relative, TCP sequence numbers.
407 -s snaplen
409 --snapshot-length=snaplen
410 Snarf snaplen bytes of data from each packet rather than the de‐
411 fault of 262144 bytes. Packets truncated because of a limited
412 snapshot are indicated in the output with ``[|proto]'', where
413 proto is the name of the protocol level at which the truncation
414 has occurred.
415 Note that taking larger snapshots both increases the amount of
417 time it takes to process packets and, effectively, decreases the
418 amount of packet buffering. This may cause packets to be lost.
419 Note also that taking smaller snapshots will discard data from
420 protocols above the transport layer, which loses information
421 that may be important. NFS and AFS requests and replies, for
422 example, are very large, and much of the detail won't be avail‐
423 able if a too-short snapshot length is selected.
424 If you need to reduce the snapshot size below the default, you
426 should limit snaplen to the smallest number that will capture
427 the protocol information you're interested in. Setting snaplen
428 to 0 sets it to the default of 262144, for backwards compatibil‐
429 ity with recent older versions of tcpdump.
430 -T type
432 Force packets selected by "expression" to be interpreted the
433 specified type. Currently known types are aodv (Ad-hoc On-de‐
434 mand Distance Vector protocol), carp (Common Address Redundancy
435 Protocol), cnfp (Cisco NetFlow protocol), domain (Domain Name
436 System), lmp (Link Management Protocol), pgm (Pragmatic General
437 Multicast), pgm_zmtp1 (ZMTP/1.0 inside PGM/EPGM), ptp (Precision
438 Time Protocol), radius (RADIUS), resp (REdis Serialization Pro‐
439 tocol), rpc (Remote Procedure Call), rtcp (Real-Time Applica‐
440 tions control protocol), rtp (Real-Time Applications protocol),
441 snmp (Simple Network Management Protocol), someip (SOME/IP),
442 tftp (Trivial File Transfer Protocol), vat (Visual Audio Tool),
443 vxlan (Virtual eXtensible Local Area Network), wb (distributed
444 White Board) and zmtp1 (ZeroMQ Message Transport Protocol 1.0).
445 Note that the pgm type above affects UDP interpretation only,
447 the native PGM is always recognised as IP protocol 113 regard‐
448 less. UDP-encapsulated PGM is often called "EPGM" or "PGM/UDP".
449 Note that the pgm_zmtp1 type above affects interpretation of
451 both native PGM and UDP at once. During the native PGM decoding
452 the application data of an ODATA/RDATA packet would be decoded
453 as a ZeroMQ datagram with ZMTP/1.0 frames. During the UDP de‐
454 coding in addition to that any UDP packet would be treated as an
455 encapsulated PGM packet.
456 -t Don't print a timestamp on each dump line.
458 -tt Print the timestamp, as seconds since January 1, 1970, 00:00:00,
460 UTC, and fractions of a second since that time, on each dump
461 line.
462 -ttt Print a delta (microsecond or nanosecond resolution depending on
464 the --time-stamp-precision option) between current and previous
465 line on each dump line. The default is microsecond resolution.
466 -tttt Print a timestamp, as hours, minutes, seconds, and fractions of
468 a second since midnight, preceded by the date, on each dump
469 line.
470 -ttttt Print a delta (microsecond or nanosecond resolution depending on
472 the --time-stamp-precision option) between current and first
473 line on each dump line. The default is microsecond resolution.
474 -u Print undecoded NFS handles.
476 -U
478 --packet-buffered
479 If the -w option is not specified, or if it is specified but the
480 --print flag is also specified, make the printed packet output
481 ``packet-buffered''; i.e., as the description of the contents of
482 each packet is printed, it will be written to the standard out‐
483 put, rather than, when not writing to a terminal, being written
484 only when the output buffer fills.
485 If the -w option is specified, make the saved raw packet output
487 ``packet-buffered''; i.e., as each packet is saved, it will be
488 written to the output file, rather than being written only when
489 the output buffer fills.
490 The -U flag will not be supported if tcpdump was built with an
492 older version of libpcap that lacks the pcap_dump_flush(3PCAP)
493 function.
494 -v When parsing and printing, produce (slightly more) verbose out‐
496 put. For example, the time to live, identification, total
497 length and options in an IP packet are printed. Also enables
498 additional packet integrity checks such as verifying the IP and
499 ICMP header checksum.
500 When writing to a file with the -w option and at the same time
502 not reading from a file with the -r option, report to stderr,
503 once per second, the number of packets captured. In Solaris,
504 FreeBSD and possibly other operating systems this periodic up‐
505 date currently can cause loss of captured packets on their way
506 from the kernel to tcpdump.
507 -vv Even more verbose output. For example, additional fields are
509 printed from NFS reply packets, and SMB packets are fully de‐
510 coded.
511 -vvv Even more verbose output. For example, telnet SB ... SE options
513 are printed in full. With -X Telnet options are printed in hex
514 as well.
515 -V file
517 Read a list of filenames from file. Standard input is used if
518 file is ``-''.
519 -w file
521 Write the raw packets to file rather than parsing and printing
522 them out. They can later be printed with the -r option. Stan‐
523 dard output is used if file is ``-''.
524 This output will be buffered if written to a file or pipe, so a
526 program reading from the file or pipe may not see packets for an
527 arbitrary amount of time after they are received. Use the -U
528 flag to cause packets to be written as soon as they are re‐
529 ceived.
530 The MIME type application/vnd.tcpdump.pcap has been registered
532 with IANA for pcap files. The filename extension .pcap appears
533 to be the most commonly used along with .cap and .dmp. Tcpdump
534 itself doesn't check the extension when reading capture files
535 and doesn't add an extension when writing them (it uses magic
536 numbers in the file header instead). However, many operating
537 systems and applications will use the extension if it is present
538 and adding one (e.g. .pcap) is recommended.
539 See pcap-savefile(5) for a description of the file format.
541 -W filecount
543 Used in conjunction with the -C option, this will limit the num‐
544 ber of files created to the specified number, and begin over‐
545 writing files from the beginning, thus creating a 'rotating'
546 buffer. In addition, it will name the files with enough leading
547 0s to support the maximum number of files, allowing them to sort
548 correctly.
549 Used in conjunction with the -G option, this will limit the num‐
551 ber of rotated dump files that get created, exiting with status
552 0 when reaching the limit.
553 If used in conjunction with both -C and -G, the -W option will
555 currently be ignored, and will only affect the file name.
556 -x When parsing and printing, in addition to printing the headers
558 of each packet, print the data of each packet (minus its link
559 level header) in hex. The smaller of the entire packet or
560 snaplen bytes will be printed. Note that this is the entire
561 link-layer packet, so for link layers that pad (e.g. Ethernet),
562 the padding bytes will also be printed when the higher layer
563 packet is shorter than the required padding. In the current im‐
564 plementation this flag may have the same effect as -xx if the
565 packet is truncated.
566 -xx When parsing and printing, in addition to printing the headers
568 of each packet, print the data of each packet, including its
569 link level header, in hex.
570 -X When parsing and printing, in addition to printing the headers
572 of each packet, print the data of each packet (minus its link
573 level header) in hex and ASCII. This is very handy for
574 analysing new protocols. In the current implementation this
575 flag may have the same effect as -XX if the packet is truncated.
576 -XX When parsing and printing, in addition to printing the headers
578 of each packet, print the data of each packet, including its
579 link level header, in hex and ASCII.
580 -y datalinktype
582 --linktype=datalinktype
583 Set the data link type to use while capturing packets (see -L)
584 or just compiling and dumping packet-matching code (see -d) to
585 datalinktype.
586 -z postrotate-command
588 Used in conjunction with the -C or -G options, this will make
589 tcpdump run " postrotate-command file " where file is the save‐
590 file being closed after each rotation. For example, specifying
591 -z gzip or -z bzip2 will compress each savefile using gzip or
592 bzip2.
593 Note that tcpdump will run the command in parallel to the cap‐
595 ture, using the lowest priority so that this doesn't disturb the
596 capture process.
597 And in case you would like to use a command that itself takes
599 flags or different arguments, you can always write a shell
600 script that will take the savefile name as the only argument,
601 make the flags & arguments arrangements and execute the command
602 that you want.
603 -Z user
605 --relinquish-privileges=user
606 If tcpdump is running as root, after opening the capture device
607 or input savefile, but before opening any savefiles for output,
608 change the user ID to user and the group ID to the primary group
609 of user.
610 This behavior is enabled by default (-Z tcpdump), and can be
612 disabled by -Z root.
613 expression
616 selects which packets will be dumped. If no expression is
617 given, all packets on the net will be dumped. Otherwise, only
618 packets for which expression is `true' will be dumped.
619 For the expression syntax, see pcap-filter(7).
621 The expression argument can be passed to tcpdump as either a
623 single Shell argument, or as multiple Shell arguments, whichever
624 is more convenient. Generally, if the expression contains Shell
625 metacharacters, such as backslashes used to escape protocol
626 names, it is easier to pass it as a single, quoted argument
627 rather than to escape the Shell metacharacters. Multiple argu‐
628 ments are concatenated with spaces before being parsed.
629
631 To print all packets arriving at or departing from sundown:
632 tcpdump host sundown
633 To print traffic between helios and either hot or ace:
635 tcpdump host helios and \( hot or ace \)
636 To print all IP packets between ace and any host except helios:
638 tcpdump ip host ace and not helios
639 To print all traffic between local hosts and hosts at Berkeley:
641 tcpdump net ucb-ether
642 To print all ftp traffic through internet gateway snup: (note that the
644 expression is quoted to prevent the shell from (mis-)interpreting the
645 parentheses):
646 tcpdump 'gateway snup and (port ftp or ftp-data)'
647 To print traffic neither sourced from nor destined for local hosts (if
649 you gateway to one other net, this stuff should never make it onto your
650 local net).
651 tcpdump ip and not net localnet
652 To print the start and end packets (the SYN and FIN packets) of each
654 TCP conversation that involves a non-local host.
655 tcpdump 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0 and not src and dst net localnet'
656 To print the TCP packets with flags RST and ACK both set. (i.e. select
658 only the RST and ACK flags in the flags field, and if the result is
659 "RST and ACK both set", match)
660 tcpdump 'tcp[tcpflags] & (tcp-rst|tcp-ack) == (tcp-rst|tcp-ack)'
661 To print all IPv4 HTTP packets to and from port 80, i.e. print only
663 packets that contain data, not, for example, SYN and FIN packets and
664 ACK-only packets. (IPv6 is left as an exercise for the reader.)
665 tcpdump 'tcp port 80 and (((ip[2:2] - ((ip[0]&0xf)<<2)) - ((tcp[12]&0xf0)>>2)) != 0)'
666 To print IP packets longer than 576 bytes sent through gateway snup:
668 tcpdump 'gateway snup and ip[2:2] > 576'
669 To print IP broadcast or multicast packets that were not sent via Eth‐
671 ernet broadcast or multicast:
672 tcpdump 'ether[0] & 1 = 0 and ip[16] >= 224'
673 To print all ICMP packets that are not echo requests/replies (i.e., not
675 ping packets):
676 tcpdump 'icmp[icmptype] != icmp-echo and icmp[icmptype] != icmp-echoreply'
677
679 The output of tcpdump is protocol dependent. The following gives a
680 brief description and examples of most of the formats.
681 Timestamps
683 By default, all output lines are preceded by a timestamp. The time‐
685 stamp is the current clock time in the form
686 hh:mm:ss.frac
687 and is as accurate as the kernel's clock. The timestamp reflects the
688 time the kernel applied a time stamp to the packet. No attempt is made
689 to account for the time lag between when the network interface finished
690 receiving the packet from the network and when the kernel applied a
691 time stamp to the packet; that time lag could include a delay between
692 the time when the network interface finished receiving a packet from
693 the network and the time when an interrupt was delivered to the kernel
694 to get it to read the packet and a delay between the time when the ker‐
695 nel serviced the `new packet' interrupt and the time when it applied a
696 time stamp to the packet.
697 Link Level Headers
699 If the '-e' option is given, the link level header is printed out. On
701 Ethernets, the source and destination addresses, protocol, and packet
702 length are printed.
703 On FDDI networks, the '-e' option causes tcpdump to print the `frame
705 control' field, the source and destination addresses, and the packet
706 length. (The `frame control' field governs the interpretation of the
707 rest of the packet. Normal packets (such as those containing IP data‐
708 grams) are `async' packets, with a priority value between 0 and 7; for
709 example, `async4'. Such packets are assumed to contain an 802.2 Logi‐
710 cal Link Control (LLC) packet; the LLC header is printed if it is not
711 an ISO datagram or a so-called SNAP packet.
712 On Token Ring networks, the '-e' option causes tcpdump to print the
714 `access control' and `frame control' fields, the source and destination
715 addresses, and the packet length. As on FDDI networks, packets are as‐
716 sumed to contain an LLC packet. Regardless of whether the '-e' option
717 is specified or not, the source routing information is printed for
718 source-routed packets.
719 On 802.11 networks, the '-e' option causes tcpdump to print the `frame
721 control' fields, all of the addresses in the 802.11 header, and the
722 packet length. As on FDDI networks, packets are assumed to contain an
723 LLC packet.
724 (N.B.: The following description assumes familiarity with the SLIP com‐
726 pression algorithm described in RFC-1144.)
727 On SLIP links, a direction indicator (``I'' for inbound, ``O'' for out‐
729 bound), packet type, and compression information are printed out. The
730 packet type is printed first. The three types are ip, utcp, and ctcp.
731 No further link information is printed for ip packets. For TCP pack‐
732 ets, the connection identifier is printed following the type. If the
733 packet is compressed, its encoded header is printed out. The special
734 cases are printed out as *S+n and *SA+n, where n is the amount by which
735 the sequence number (or sequence number and ack) has changed. If it is
736 not a special case, zero or more changes are printed. A change is in‐
737 dicated by U (urgent pointer), W (window), A (ack), S (sequence num‐
738 ber), and I (packet ID), followed by a delta (+n or -n), or a new value
739 (=n). Finally, the amount of data in the packet and compressed header
740 length are printed.
741 For example, the following line shows an outbound compressed TCP
743 packet, with an implicit connection identifier; the ack has changed by
744 6, the sequence number by 49, and the packet ID by 6; there are 3 bytes
745 of data and 6 bytes of compressed header:
746 O ctcp * A+6 S+49 I+6 3 (6)
747 ARP/RARP Packets
749 ARP/RARP output shows the type of request and its arguments. The for‐
751 mat is intended to be self explanatory. Here is a short sample taken
752 from the start of an `rlogin' from host rtsg to host csam:
753 arp who-has csam tell rtsg
754 arp reply csam is-at CSAM
755 The first line says that rtsg sent an ARP packet asking for the Ether‐
756 net address of internet host csam. Csam replies with its Ethernet ad‐
757 dress (in this example, Ethernet addresses are in caps and internet ad‐
758 dresses in lower case).
759 This would look less redundant if we had done tcpdump -n:
761 arp who-has 128.3.254.6 tell 128.3.254.68
762 arp reply 128.3.254.6 is-at 02:07:01:00:01:c4
763 If we had done tcpdump -e, the fact that the first packet is broadcast
765 and the second is point-to-point would be visible:
766 RTSG Broadcast 0806 64: arp who-has csam tell rtsg
767 CSAM RTSG 0806 64: arp reply csam is-at CSAM
768 For the first packet this says the Ethernet source address is RTSG, the
769 destination is the Ethernet broadcast address, the type field contained
770 hex 0806 (type ETHER_ARP) and the total length was 64 bytes.
771 IPv4 Packets
773 If the link-layer header is not being printed, for IPv4 packets, IP is
775 printed after the time stamp.
776 If the -v flag is specified, information from the IPv4 header is shown
778 in parentheses after the IP or the link-layer header. The general for‐
779 mat of this information is:
780 tos tos, ttl ttl, id id, offset offset, flags [flags], proto proto, length length, options (options)
781 tos is the type of service field; if the ECN bits are non-zero, those
782 are reported as ECT(1), ECT(0), or CE. ttl is the time-to-live; it is
783 not reported if it is zero. id is the IP identification field. offset
784 is the fragment offset field; it is printed whether this is part of a
785 fragmented datagram or not. flags are the MF and DF flags; + is re‐
786 ported if MF is set, and DF is reported if F is set. If neither are
787 set, . is reported. proto is the protocol ID field. length is the to‐
788 tal length field. options are the IP options, if any.
789 Next, for TCP and UDP packets, the source and destination IP addresses
791 and TCP or UDP ports, with a dot between each IP address and its corre‐
792 sponding port, will be printed, with a > separating the source and des‐
793 tination. For other protocols, the addresses will be printed, with a >
794 separating the source and destination. Higher level protocol informa‐
795 tion, if any, will be printed after that.
796 For fragmented IP datagrams, the first fragment contains the higher
798 level protocol header; fragments after the first contain no higher
799 level protocol header. Fragmentation information will be printed only
800 with the -v flag, in the IP header information, as described above.
801 TCP Packets
803 (N.B.:The following description assumes familiarity with the TCP proto‐
805 col described in RFC-793. If you are not familiar with the protocol,
806 this description will not be of much use to you.)
807 The general format of a TCP protocol line is:
809 src > dst: Flags [tcpflags], seq data-seqno, ack ackno, win window, urg urgent, options [opts], length len
810 Src and dst are the source and destination IP addresses and ports.
811 Tcpflags are some combination of S (SYN), F (FIN), P (PUSH), R (RST), U
812 (URG), W (ECN CWR), E (ECN-Echo) or `.' (ACK), or `none' if no flags
813 are set. Data-seqno describes the portion of sequence space covered by
814 the data in this packet (see example below). Ackno is sequence number
815 of the next data expected the other direction on this connection. Win‐
816 dow is the number of bytes of receive buffer space available the other
817 direction on this connection. Urg indicates there is `urgent' data in
818 the packet. Opts are TCP options (e.g., mss 1024). Len is the length
819 of payload data.
820 Iptype, Src, dst, and flags are always present. The other fields de‐
822 pend on the contents of the packet's TCP protocol header and are output
823 only if appropriate.
824 Here is the opening portion of an rlogin from host rtsg to host csam.
826 IP rtsg.1023 > csam.login: Flags [S], seq 768512:768512, win 4096, opts [mss 1024]
827 IP csam.login > rtsg.1023: Flags [S.], seq, 947648:947648, ack 768513, win 4096, opts [mss 1024]
828 IP rtsg.1023 > csam.login: Flags [.], ack 1, win 4096
829 IP rtsg.1023 > csam.login: Flags [P.], seq 1:2, ack 1, win 4096, length 1
830 IP csam.login > rtsg.1023: Flags [.], ack 2, win 4096
831 IP rtsg.1023 > csam.login: Flags [P.], seq 2:21, ack 1, win 4096, length 19
832 IP csam.login > rtsg.1023: Flags [P.], seq 1:2, ack 21, win 4077, length 1
833 IP csam.login > rtsg.1023: Flags [P.], seq 2:3, ack 21, win 4077, urg 1, length 1
834 IP csam.login > rtsg.1023: Flags [P.], seq 3:4, ack 21, win 4077, urg 1, length 1
835 The first line says that TCP port 1023 on rtsg sent a packet to port
836 login on csam. The S indicates that the SYN flag was set. The packet
837 sequence number was 768512 and it contained no data. (The notation is
838 `first:last' which means `sequence numbers first up to but not includ‐
839 ing last'.) There was no piggy-backed ACK, the available receive win‐
840 dow was 4096 bytes and there was a max-segment-size option requesting
841 an MSS of 1024 bytes.
842 Csam replies with a similar packet except it includes a piggy-backed
844 ACK for rtsg's SYN. Rtsg then ACKs csam's SYN. The `.' means the ACK
845 flag was set. The packet contained no data so there is no data se‐
846 quence number or length. Note that the ACK sequence number is a small
847 integer (1). The first time tcpdump sees a TCP `conversation', it
848 prints the sequence number from the packet. On subsequent packets of
849 the conversation, the difference between the current packet's sequence
850 number and this initial sequence number is printed. This means that
851 sequence numbers after the first can be interpreted as relative byte
852 positions in the conversation's data stream (with the first data byte
853 each direction being `1'). `-S' will override this feature, causing
854 the original sequence numbers to be output.
855 On the 6th line, rtsg sends csam 19 bytes of data (bytes 2 through 20
857 in the rtsg → csam side of the conversation). The PUSH flag is set in
858 the packet. On the 7th line, csam says it's received data sent by rtsg
859 up to but not including byte 21. Most of this data is apparently sit‐
860 ting in the socket buffer since csam's receive window has gotten 19
861 bytes smaller. Csam also sends one byte of data to rtsg in this
862 packet. On the 8th and 9th lines, csam sends two bytes of urgent,
863 pushed data to rtsg.
864 If the snapshot was small enough that tcpdump didn't capture the full
866 TCP header, it interprets as much of the header as it can and then re‐
867 ports ``[|tcp]'' to indicate the remainder could not be interpreted.
868 If the header contains a bogus option (one with a length that's either
869 too small or beyond the end of the header), tcpdump reports it as
870 ``[bad opt]'' and does not interpret any further options (since it's
871 impossible to tell where they start). If the header length indicates
872 options are present but the IP datagram length is not long enough for
873 the options to actually be there, tcpdump reports it as ``[bad hdr
874 length]''.
875 Capturing TCP packets with particular flag combinations (SYN-ACK, URG-
877 ACK, etc.)
878 There are 8 bits in the control bits section of the TCP header:
880 CWR | ECE | URG | ACK | PSH | RST | SYN | FIN
882 Let's assume that we want to watch packets used in establishing a TCP
884 connection. Recall that TCP uses a 3-way handshake protocol when it
885 initializes a new connection; the connection sequence with regard to
886 the TCP control bits is
887 1) Caller sends SYN
889 2) Recipient responds with SYN, ACK
890 3) Caller sends ACK
891 Now we're interested in capturing packets that have only the SYN bit
893 set (Step 1). Note that we don't want packets from step 2 (SYN-ACK),
894 just a plain initial SYN. What we need is a correct filter expression
895 for tcpdump.
896 Recall the structure of a TCP header without options:
898 0 15 31
900 -----------------------------------------------------------------
901 | source port | destination port |
902 -----------------------------------------------------------------
903 | sequence number |
904 -----------------------------------------------------------------
905 | acknowledgment number |
906 -----------------------------------------------------------------
907 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
908 -----------------------------------------------------------------
909 | TCP checksum | urgent pointer |
910 -----------------------------------------------------------------
911 A TCP header usually holds 20 octets of data, unless options are
913 present. The first line of the graph contains octets 0 - 3, the second
914 line shows octets 4 - 7 etc.
915 Starting to count with 0, the relevant TCP control bits are contained
917 in octet 13:
918 0 7| 15| 23| 31
920 ----------------|---------------|---------------|----------------
921 | HL | rsvd |C|E|U|A|P|R|S|F| window size |
922 ----------------|---------------|---------------|----------------
923 | | 13th octet | | |
924 Let's have a closer look at octet no. 13:
926 | |
928 |---------------|
929 |C|E|U|A|P|R|S|F|
930 |---------------|
931 |7 5 3 0|
932 These are the TCP control bits we are interested in. We have numbered
934 the bits in this octet from 0 to 7, right to left, so the PSH bit is
935 bit number 3, while the URG bit is number 5.
936 Recall that we want to capture packets with only SYN set. Let's see
938 what happens to octet 13 if a TCP datagram arrives with the SYN bit set
939 in its header:
940 |C|E|U|A|P|R|S|F|
942 |---------------|
943 |0 0 0 0 0 0 1 0|
944 |---------------|
945 |7 6 5 4 3 2 1 0|
946 Looking at the control bits section we see that only bit number 1 (SYN)
948 is set.
949 Assuming that octet number 13 is an 8-bit unsigned integer in network
951 byte order, the binary value of this octet is
952 00000010
954 and its decimal representation is
956 7 6 5 4 3 2 1 0
958 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 1*2 + 0*2 = 2
959 We're almost done, because now we know that if only SYN is set, the
961 value of the 13th octet in the TCP header, when interpreted as a 8-bit
962 unsigned integer in network byte order, must be exactly 2.
963 This relationship can be expressed as
965 tcp[13] == 2
966 We can use this expression as the filter for tcpdump in order to watch
968 packets which have only SYN set:
969 tcpdump -i xl0 tcp[13] == 2
970 The expression says "let the 13th octet of a TCP datagram have the dec‐
972 imal value 2", which is exactly what we want.
973 Now, let's assume that we need to capture SYN packets, but we don't
975 care if ACK or any other TCP control bit is set at the same time.
976 Let's see what happens to octet 13 when a TCP datagram with SYN-ACK set
977 arrives:
978 |C|E|U|A|P|R|S|F|
980 |---------------|
981 |0 0 0 1 0 0 1 0|
982 |---------------|
983 |7 6 5 4 3 2 1 0|
984 Now bits 1 and 4 are set in the 13th octet. The binary value of octet
986 13 is
987 00010010
989 which translates to decimal
991 7 6 5 4 3 2 1 0
993 0*2 + 0*2 + 0*2 + 1*2 + 0*2 + 0*2 + 1*2 + 0*2 = 18
994 Now we can't just use 'tcp[13] == 18' in the tcpdump filter expression,
996 because that would select only those packets that have SYN-ACK set, but
997 not those with only SYN set. Remember that we don't care if ACK or any
998 other control bit is set as long as SYN is set.
999 In order to achieve our goal, we need to logically AND the binary value
1001 of octet 13 with some other value to preserve the SYN bit. We know
1002 that we want SYN to be set in any case, so we'll logically AND the
1003 value in the 13th octet with the binary value of a SYN:
1004 00010010 SYN-ACK 00000010 SYN
1006 AND 00000010 (we want SYN) AND 00000010 (we want SYN)
1007 -------- --------
1008 = 00000010 = 00000010
1009 We see that this AND operation delivers the same result regardless
1011 whether ACK or another TCP control bit is set. The decimal representa‐
1012 tion of the AND value as well as the result of this operation is 2 (bi‐
1013 nary 00000010), so we know that for packets with SYN set the following
1014 relation must hold true:
1015 ( ( value of octet 13 ) AND ( 2 ) ) == ( 2 )
1017 This points us to the tcpdump filter expression
1019 tcpdump -i xl0 'tcp[13] & 2 == 2'
1020 Some offsets and field values may be expressed as names rather than as
1022 numeric values. For example tcp[13] may be replaced with tcp[tcpflags].
1023 The following TCP flag field values are also available: tcp-fin, tcp-
1024 syn, tcp-rst, tcp-push, tcp-ack, tcp-urg.
1025 This can be demonstrated as:
1027 tcpdump -i xl0 'tcp[tcpflags] & tcp-push != 0'
1028 Note that you should use single quotes or a backslash in the expression
1030 to hide the AND ('&') special character from the shell.
1031 UDP Packets
1033 UDP format is illustrated by this rwho packet:
1035 actinide.who > broadcast.who: udp 84
1036 This says that port who on host actinide sent a UDP datagram to port
1037 who on host broadcast, the Internet broadcast address. The packet con‐
1038 tained 84 bytes of user data.
1039 Some UDP services are recognized (from the source or destination port
1041 number) and the higher level protocol information printed. In particu‐
1042 lar, Domain Name service requests (RFC-1034/1035) and Sun RPC calls
1043 (RFC-1050) to NFS.
1044 TCP or UDP Name Server Requests
1046 (N.B.:The following description assumes familiarity with the Domain
1048 Service protocol described in RFC-1035. If you are not familiar with
1049 the protocol, the following description will appear to be written in
1050 Greek.)
1051 Name server requests are formatted as
1053 src > dst: id op? flags qtype qclass name (len)
1054 h2opolo.1538 > helios.domain: 3+ A? ucbvax.berkeley.edu. (37)
1055 Host h2opolo asked the domain server on helios for an address record
1056 (qtype=A) associated with the name ucbvax.berkeley.edu. The query id
1057 was `3'. The `+' indicates the recursion desired flag was set. The
1058 query length was 37 bytes, excluding the TCP or UDP and IP protocol
1059 headers. The query operation was the normal one, Query, so the op
1060 field was omitted. If the op had been anything else, it would have
1061 been printed between the `3' and the `+'. Similarly, the qclass was
1062 the normal one, C_IN, and omitted. Any other qclass would have been
1063 printed immediately after the `A'.
1064 A few anomalies are checked and may result in extra fields enclosed in
1066 square brackets: If a query contains an answer, authority records or
1067 additional records section, ancount, nscount, or arcount are printed as
1068 `[na]', `[nn]' or `[nau]' where n is the appropriate count. If any of
1069 the response bits are set (AA, RA or rcode) or any of the `must be
1070 zero' bits are set in bytes two and three, `[b2&3=x]' is printed, where
1071 x is the hex value of header bytes two and three.
1072 TCP or UDP Name Server Responses
1074 Name server responses are formatted as
1076 src > dst: id op rcode flags a/n/au type class data (len)
1077 helios.domain > h2opolo.1538: 3 3/3/7 A 128.32.137.3 (273)
1078 helios.domain > h2opolo.1537: 2 NXDomain* 0/1/0 (97)
1079 In the first example, helios responds to query id 3 from h2opolo with 3
1080 answer records, 3 name server records and 7 additional records. The
1081 first answer record is type A (address) and its data is internet ad‐
1082 dress 128.32.137.3. The total size of the response was 273 bytes, ex‐
1083 cluding TCP or UDP and IP headers. The op (Query) and response code
1084 (NoError) were omitted, as was the class (C_IN) of the A record.
1085 In the second example, helios responds to query 2 with a response code
1087 of non-existent domain (NXDomain) with no answers, one name server and
1088 no authority records. The `*' indicates that the authoritative answer
1089 bit was set. Since there were no answers, no type, class or data were
1090 printed.
1091 Other flag characters that might appear are `-' (recursion available,
1093 RA, not set) and `|' (truncated message, TC, set). If the `question'
1094 section doesn't contain exactly one entry, `[nq]' is printed.
1095 SMB/CIFS decoding
1097 tcpdump now includes fairly extensive SMB/CIFS/NBT decoding for data on
1099 UDP/137, UDP/138 and TCP/139. Some primitive decoding of IPX and Net‐
1100 BEUI SMB data is also done.
1101 By default a fairly minimal decode is done, with a much more detailed
1103 decode done if -v is used. Be warned that with -v a single SMB packet
1104 may take up a page or more, so only use -v if you really want all the
1105 gory details.
1106 For information on SMB packet formats and what all the fields mean see
1108 https://download.samba.org/pub/samba/specs/ and other online resources.
1109 The SMB patches were written by Andrew Tridgell (tridge@samba.org).
1110 NFS Requests and Replies
1112 Sun NFS (Network File System) requests and replies are printed as:
1114 src.sport > dst.nfs: NFS request xid xid len op args
1115 src.nfs > dst.dport: NFS reply xid xid reply stat len op results
1116 sushi.1023 > wrl.nfs: NFS request xid 26377
1117 112 readlink fh 21,24/10.73165
1118 wrl.nfs > sushi.1023: NFS reply xid 26377
1119 reply ok 40 readlink "../var"
1120 sushi.1022 > wrl.nfs: NFS request xid 8219
1121 144 lookup fh 9,74/4096.6878 "xcolors"
1122 wrl.nfs > sushi.1022: NFS reply xid 8219
1123 reply ok 128 lookup fh 9,74/4134.3150
1124 In the first line, host sushi sends a transaction with id 26377 to wrl.
1125 The request was 112 bytes, excluding the UDP and IP headers. The oper‐
1126 ation was a readlink (read symbolic link) on file handle (fh)
1127 21,24/10.731657119. (If one is lucky, as in this case, the file handle
1128 can be interpreted as a major,minor device number pair, followed by the
1129 inode number and generation number.) In the second line, wrl replies
1130 `ok' with the same transaction id and the contents of the link.
1131 In the third line, sushi asks (using a new transaction id) wrl to
1133 lookup the name `xcolors' in directory file 9,74/4096.6878. In the
1134 fourth line, wrl sends a reply with the respective transaction id.
1135 Note that the data printed depends on the operation type. The format
1137 is intended to be self explanatory if read in conjunction with an NFS
1138 protocol spec. Also note that older versions of tcpdump printed NFS
1139 packets in a slightly different format: the transaction id (xid) would
1140 be printed instead of the non-NFS port number of the packet.
1141 If the -v (verbose) flag is given, additional information is printed.
1143 For example:
1144 sushi.1023 > wrl.nfs: NFS request xid 79658
1145 148 read fh 21,11/12.195 8192 bytes @ 24576
1146 wrl.nfs > sushi.1023: NFS reply xid 79658
1147 reply ok 1472 read REG 100664 ids 417/0 sz 29388
1148 (-v also prints the IP header TTL, ID, length, and fragmentation
1149 fields, which have been omitted from this example.) In the first line,
1150 sushi asks wrl to read 8192 bytes from file 21,11/12.195, at byte off‐
1151 set 24576. Wrl replies `ok'; the packet shown on the second line is
1152 the first fragment of the reply, and hence is only 1472 bytes long (the
1153 other bytes will follow in subsequent fragments, but these fragments do
1154 not have NFS or even UDP headers and so might not be printed, depending
1155 on the filter expression used). Because the -v flag is given, some of
1156 the file attributes (which are returned in addition to the file data)
1157 are printed: the file type (``REG'', for regular file), the file mode
1158 (in octal), the UID and GID, and the file size.
1159 If the -v flag is given more than once, even more details are printed.
1161 NFS reply packets do not explicitly identify the RPC operation. In‐
1163 stead, tcpdump keeps track of ``recent'' requests, and matches them to
1164 the replies using the transaction ID. If a reply does not closely fol‐
1165 low the corresponding request, it might not be parsable.
1166 AFS Requests and Replies
1168 Transarc AFS (Andrew File System) requests and replies are printed as:
1170 src.sport > dst.dport: rx packet-type
1172 src.sport > dst.dport: rx packet-type service call call-name args
1173 src.sport > dst.dport: rx packet-type service reply call-name args
1174 elvis.7001 > pike.afsfs:
1175 rx data fs call rename old fid 536876964/1/1 ".newsrc.new"
1176 new fid 536876964/1/1 ".newsrc"
1177 pike.afsfs > elvis.7001: rx data fs reply rename
1178 In the first line, host elvis sends a RX packet to pike. This was a RX
1179 data packet to the fs (fileserver) service, and is the start of an RPC
1180 call. The RPC call was a rename, with the old directory file id of
1181 536876964/1/1 and an old filename of `.newsrc.new', and a new directory
1182 file id of 536876964/1/1 and a new filename of `.newsrc'. The host
1183 pike responds with a RPC reply to the rename call (which was success‐
1184 ful, because it was a data packet and not an abort packet).
1185 In general, all AFS RPCs are decoded at least by RPC call name. Most
1187 AFS RPCs have at least some of the arguments decoded (generally only
1188 the `interesting' arguments, for some definition of interesting).
1189 The format is intended to be self-describing, but it will probably not
1191 be useful to people who are not familiar with the workings of AFS and
1192 RX.
1193 If the -v (verbose) flag is given twice, acknowledgement packets and
1195 additional header information is printed, such as the RX call ID, call
1196 number, sequence number, serial number, and the RX packet flags.
1197 If the -v flag is given twice, additional information is printed, such
1199 as the RX call ID, serial number, and the RX packet flags. The MTU ne‐
1200 gotiation information is also printed from RX ack packets.
1201 If the -v flag is given three times, the security index and service id
1203 are printed.
1204 Error codes are printed for abort packets, with the exception of Ubik
1206 beacon packets (because abort packets are used to signify a yes vote
1207 for the Ubik protocol).
1208 AFS reply packets do not explicitly identify the RPC operation. In‐
1210 stead, tcpdump keeps track of ``recent'' requests, and matches them to
1211 the replies using the call number and service ID. If a reply does not
1212 closely follow the corresponding request, it might not be parsable.
1213 KIP AppleTalk (DDP in UDP)
1216 AppleTalk DDP packets encapsulated in UDP datagrams are de-encapsulated
1218 and dumped as DDP packets (i.e., all the UDP header information is dis‐
1219 carded). The file /etc/atalk.names is used to translate AppleTalk net
1220 and node numbers to names. Lines in this file have the form
1221 number name
1222 1.254 ether
1224 16.1 icsd-net
1225 1.254.110 ace
1226 The first two lines give the names of AppleTalk networks. The third
1227 line gives the name of a particular host (a host is distinguished from
1228 a net by the 3rd octet in the number - a net number must have two
1229 octets and a host number must have three octets.) The number and name
1230 should be separated by whitespace (blanks or tabs). The
1231 /etc/atalk.names file may contain blank lines or comment lines (lines
1232 starting with a `#').
1233 AppleTalk addresses are printed in the form
1235 net.host.port
1236 144.1.209.2 > icsd-net.112.220
1238 office.2 > icsd-net.112.220
1239 jssmag.149.235 > icsd-net.2
1240 (If the /etc/atalk.names doesn't exist or doesn't contain an entry for
1241 some AppleTalk host/net number, addresses are printed in numeric form.)
1242 In the first example, NBP (DDP port 2) on net 144.1 node 209 is sending
1243 to whatever is listening on port 220 of net icsd node 112. The second
1244 line is the same except the full name of the source node is known (`of‐
1245 fice'). The third line is a send from port 235 on net jssmag node 149
1246 to broadcast on the icsd-net NBP port (note that the broadcast address
1247 (255) is indicated by a net name with no host number - for this reason
1248 it's a good idea to keep node names and net names distinct in
1249 /etc/atalk.names).
1250 NBP (name binding protocol) and ATP (AppleTalk transaction protocol)
1252 packets have their contents interpreted. Other protocols just dump the
1253 protocol name (or number if no name is registered for the protocol) and
1254 packet size.
1255 NBP packets are formatted like the following examples:
1257 icsd-net.112.220 > jssmag.2: nbp-lkup 190: "=:LaserWriter@*"
1258 jssmag.209.2 > icsd-net.112.220: nbp-reply 190: "RM1140:LaserWriter@*" 250
1259 techpit.2 > icsd-net.112.220: nbp-reply 190: "techpit:LaserWriter@*" 186
1260 The first line is a name lookup request for laserwriters sent by net
1261 icsd host 112 and broadcast on net jssmag. The nbp id for the lookup
1262 is 190. The second line shows a reply for this request (note that it
1263 has the same id) from host jssmag.209 saying that it has a laserwriter
1264 resource named "RM1140" registered on port 250. The third line is an‐
1265 other reply to the same request saying host techpit has laserwriter
1266 "techpit" registered on port 186.
1267 ATP packet formatting is demonstrated by the following example:
1269 jssmag.209.165 > helios.132: atp-req 12266<0-7> 0xae030001
1270 helios.132 > jssmag.209.165: atp-resp 12266:0 (512) 0xae040000
1271 helios.132 > jssmag.209.165: atp-resp 12266:1 (512) 0xae040000
1272 helios.132 > jssmag.209.165: atp-resp 12266:2 (512) 0xae040000
1273 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1274 helios.132 > jssmag.209.165: atp-resp 12266:4 (512) 0xae040000
1275 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1276 helios.132 > jssmag.209.165: atp-resp 12266:6 (512) 0xae040000
1277 helios.132 > jssmag.209.165: atp-resp*12266:7 (512) 0xae040000
1278 jssmag.209.165 > helios.132: atp-req 12266<3,5> 0xae030001
1279 helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
1280 helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
1281 jssmag.209.165 > helios.132: atp-rel 12266<0-7> 0xae030001
1282 jssmag.209.133 > helios.132: atp-req* 12267<0-7> 0xae030002
1283 Jssmag.209 initiates transaction id 12266 with host helios by request‐
1284 ing up to 8 packets (the `<0-7>'). The hex number at the end of the
1285 line is the value of the `userdata' field in the request.
1286 Helios responds with 8 512-byte packets. The `:digit' following the
1288 transaction id gives the packet sequence number in the transaction and
1289 the number in parens is the amount of data in the packet, excluding the
1290 ATP header. The `*' on packet 7 indicates that the EOM bit was set.
1291 Jssmag.209 then requests that packets 3 & 5 be retransmitted. Helios
1293 resends them then jssmag.209 releases the transaction. Finally, jss‐
1294 mag.209 initiates the next request. The `*' on the request indicates
1295 that XO (`exactly once') was not set.
1296
1299 stty(1), pcap(3PCAP), bpf(4), nit(4P), pcap-savefile(5),
1300 pcap-filter(7), pcap-tstamp(7)
1301 https://www.iana.org/assignments/media-types/applica‐
1303 tion/vnd.tcpdump.pcap
1304
1306 The original authors are:
1307 Van Jacobson, Craig Leres and Steven McCanne, all of the Lawrence
1309 Berkeley National Laboratory, University of California, Berkeley, CA.
1310 It is currently being maintained by tcpdump.org.
1312 The current version is available via HTTPS:
1314 https://www.tcpdump.org/
1316 The original distribution is available via anonymous ftp:
1318 ftp://ftp.ee.lbl.gov/old/tcpdump.tar.Z
1320 IPv6/IPsec support is added by WIDE/KAME project. This program uses
1322 OpenSSL/LibreSSL, under specific configurations.
1323
1325 To report a security issue please send an e-mail to
1326 security@tcpdump.org.
1327 To report bugs and other problems, contribute patches, request a fea‐
1329 ture, provide generic feedback etc. please see the file CONTRIBUTING in
1330 the tcpdump source tree root.
1331 NIT doesn't let you watch your own outbound traffic, BPF will. We rec‐
1333 ommend that you use the latter.
1334 On Linux systems with 2.0[.x] kernels:
1336 packets on the loopback device will be seen twice;
1338 packet filtering cannot be done in the kernel, so that all pack‐
1340 ets must be copied from the kernel in order to be filtered in
1341 user mode;
1342 all of a packet, not just the part that's within the snapshot
1344 length, will be copied from the kernel (the 2.0[.x] packet cap‐
1345 ture mechanism, if asked to copy only part of a packet to
1346 userspace, will not report the true length of the packet; this
1347 would cause most IP packets to get an error from tcpdump);
1348 capturing on some PPP devices won't work correctly.
1350 We recommend that you upgrade to a 2.2 or later kernel.
1352 Some attempt should be made to reassemble IP fragments or, at least to
1354 compute the right length for the higher level protocol.
1355 Name server inverse queries are not dumped correctly: the (empty) ques‐
1357 tion section is printed rather than real query in the answer section.
1358 Some believe that inverse queries are themselves a bug and prefer to
1359 fix the program generating them rather than tcpdump.
1360 A packet trace that crosses a daylight savings time change will give
1362 skewed time stamps (the time change is ignored).
1363 Filter expressions on fields other than those in Token Ring headers
1365 will not correctly handle source-routed Token Ring packets.
1366 Filter expressions on fields other than those in 802.11 headers will
1368 not correctly handle 802.11 data packets with both To DS and From DS
1369 set.
1370 ip6 proto should chase header chain, but at this moment it does not.
1372 ip6 protochain is supplied for this behavior.
1373 Arithmetic expression against transport layer headers, like tcp[0],
1375 does not work against IPv6 packets. It only looks at IPv4 packets.
1376 21 December 2020 TCPDUMP(8)