TCPDUMP(8)

📦

version 4.99.6 · 9 versions

4.99.6 · fedora 43, fedora 44 (latest)
4.99.0 · centos-stream 9
4.99.4 · centos-stream 10
4.99.1 · ubuntu jammy
4.99.5 · fedora 43
4.99.4 · ubuntu noble
4.99.3 · debian bookworm
4.99.6 · ubuntu resolute, ubuntu stonking
4.99.5 · debian trixie, ubuntu questing

✨ TCPDUMP(8) ✨

okay bestie, gather round 💖 this is tcpdump, straight out of the System Manager's Manual (section 8, the big-kid one), and it is literally the main character of network debugging. she's been iconic for decades and she's about to tell you everything 💅

🌸 NAME 🌸

tcpdump — dump traffic on a network 🦋

no cap, that's it. that's the whole vibe.

⭐ SYNOPSIS ⭐

       tcpdump [ -AbdDefghHIJKlLnNOpqStuUvxX# ] [ -B buffer_size ]
               [ -c count ] [ --count ] [ -C file_size ]
               [ -E spi@ipaddr algo:secret,...  ]
               [ -F file ] [ -G rotate_seconds ] [ -i interface ]
               [ --immediate-mode ] [ -j tstamp_type ] [ -m module ]
               [ -M secret ] [ --number ] [ --print ] [ -Q in|out|inout ]
               [ -r file ] [ -s snaplen ] [ -T type ] [ --version ] [ -V file ]
       [ -w file ] [ -W filecount ]
               [ -y datalinktype ] [ -z postrotate-command ] [ -Z user ]
               [ --time-stamp-precision=tstamp_precision ] [ --micro ] [ --nano
       ]
               [ expression ]

like. she has OPTIONS. she has RANGE. she contains multitudes 🫶

🎀 DESCRIPTION 🎀

so what does she do? tcpdump prints out a description of the contents of packets on a network interface that match the Boolean expression (peep pcap-filter(7) for the expression syntax, iykyk). and that description? it's preceded by a time stamp, printed, by default, as hours, minutes, seconds, and fractions of a second since midnight ⏰

you can also run her with the -w flag, which makes her save the packet data to a file for later analysis, and/or with the -r flag, which makes her read from a saved packet file rather than read packets from a network interface. she can also be run with the -V flag, which makes her read a list of saved packet files. and in all cases, only packets that match expression will be processed by tcpdump. periodt ✨

tcpdump will, if not run with the -c flag, continue capturing packets until it is interrupted by a SIGINT signal (generated, for example, by typing your interrupt character, typically control-C) or a SIGTERM signal (typically generated with the kill(1) command) 🛑 if run with the -c flag, she'll capture packets until interrupted by a SIGINT or SIGTERM signal or until the specified number of packets have been processed. either way, she stops when she stops 💅

when tcpdump finishes capturing packets, she will report counts of:

packets captured — this is the number of packets that tcpdump has received and processed 💖
packets received by filter — okay so the meaning of this one depends on the OS you're running tcpdump on, and possibly on the way the OS was configured. if a filter was specified on the command line, on some OSes it counts packets regardless of whether they were matched by the filter expression and, even if they were matched by the filter expression, regardless of whether tcpdump has read and processed them yet; on other OSes it counts only packets that were matched by the filter expression regardless of whether tcpdump has read and processed them yet; and on other OSes it counts only packets that were matched by the filter expression and were processed by tcpdump. so yeah. it's giving "it depends" 😭
packets dropped by kernel — this is the number of packets that were dropped, due to a lack of buffer space, by the packet capture mechanism in the OS on which tcpdump is running, if the OS reports that information to applications. if not, it will be reported as 0 💀

on platforms that support the SIGINFO signal, such as most BSDs (including macOS) and Digital/Tru64 UNIX, she'll report those counts when she receives a SIGINFO signal (generated, for example, by typing your status character, typically control-T, although on some platforms, such as macOS, the status character is not set by default, so you must set it with stty(1) in order to use it) and will continue capturing packets 🌈 on platforms that do not support the SIGINFO signal, the same can be achieved by using the SIGUSR1 signal 👑

using the SIGUSR2 signal along with the -w flag will forcibly flush the packet buffer into the output file. flush it. flush it real good 🧚

reading packets from a network interface may require that you have special privileges — see the pcap(3PCAP) man page for details. reading a saved packet file doesn't require special privileges 🙈

💅 OPTIONS 💅

here come the flags, bestie. all of them. every single one. no skips, no summarising, we are eating the whole plate 🍓

-A — Print each packet (minus its link level header) in ASCII. Handy for capturing web pages 🌐 No effect when -x[x] or -X[X] options are used.
-b — Print the AS number in BGP packets in ASDOT notation rather than ASPLAIN notation.
-B buffer_size, --buffer-size=buffer_size — Set the operating system capture buffer size to buffer_size, in units of KiB (1024 bytes). she's getting a bigger bucket 🪣
-c count — Exit after receiving count packets.
--count — Print only on stdout the packet count when reading capture file(s) instead of parsing/printing the packets. If a filter is specified on the command line, tcpdump counts only packets that were matched by the filter expression. just the number, ma'am 🔢
-C file_size — Before writing a raw packet to a savefile, check whether the file is currently larger than file_size and, if so, close the current savefile and open a new one. Savefiles after the first savefile will have the name specified with the -w flag, with a number after it, starting at 1 and continuing upward. The default unit of file_size is millions of bytes (1,000,000 bytes, not 1,048,576 bytes) 📏

By adding a suffix of k/K, m/M or g/G to the value, the unit can be changed to 1,024 (KiB), 1,048,576 (MiB), or 1,073,741,824 (GiB) respectively.

note that when used with -Z option (enabled by default), privileges are dropped before opening the first savefile ⚠️

-d — Dump the compiled packet-matching code in a human readable form to standard output and stop.

please mind that although code compilation is always DLT-specific, typically it is impossible (and unnecessary) to specify which DLT to use for the dump because tcpdump uses either the DLT of the input pcap file specified with -r, or the default DLT of the network interface specified with -i, or the particular DLT of the network interface specified with -y and -i respectively. In these cases the dump shows the same exact code that would filter the input file or the network interface without -d. 🧠

However, when neither -r nor -i is specified, specifying -d prevents tcpdump from guessing a suitable network interface (see -i). In this case the DLT defaults to EN10MB and can be set to another valid value manually with -y. no vibes-based guessing allowed, bestie 😌

-dd — Dump packet-matching code as a C program fragment.
-ddd — Dump packet-matching code as decimal numbers (preceded with a count).
-D, --list-interfaces — Print the list of the network interfaces available on the system and on which tcpdump can capture packets 🌈 For each network interface, a number and an interface name, possibly followed by a text description of the interface, are printed. The interface name or the number can be supplied to the -i flag to specify an interface on which to capture.

This can be useful on systems that don't have a command to list them (e.g., Windows systems, or UNIX systems lacking ifconfig -a); the number can be useful on Windows 2000 and later systems, where the interface name is a somewhat complex string 😵‍💫

the -D flag will not be supported if tcpdump was built with an older version of libpcap that lacks the pcap_findalldevs(3PCAP) function 🙈

-e — Print the link-level header on each dump line. This can be used, for example, to print MAC layer addresses for protocols such as Ethernet and IEEE 802.11. show us the receipts, layer 2 edition 📬
-E — Use spi@ipaddr algo:secret for decrypting IPsec ESP packets that are addressed to addr and contain Security Parameter Index value spi. This combination may be repeated with comma or newline separation.

Note that setting the secret for IPv4 ESP packets is supported at this time.

Algorithms may be des-cbc, 3des-cbc, blowfish-cbc, rc3-cbc, cast128-cbc, or none. The default is des-cbc. The ability to decrypt packets is only present if tcpdump was compiled with cryptography enabled 🔐

secret is the ASCII text for ESP secret key. If preceded by 0x, then a hex value will be read.

The option assumes RFC 2406 ESP, not RFC 1827 ESP. The option is only for debugging purposes, and the use of this option with a true secret key is discouraged.

⚠️ by presenting IPsec secret key onto command line you make it visible to others, via ps(1) and other occasions. not the move, bestie. everyone can see it. everyone 💀

In addition to the above syntax, the syntax file name may be used to have tcpdump read the provided file in. The file is opened upon receiving the first ESP packet, so any special permissions that tcpdump may have been given should already have been given up.

-f — Print foreign IPv4 addresses numerically rather than symbolically (this option is intended to get around serious brain damage in Sun's NIS server — usually it hangs forever translating non-local internet numbers 🫠).

The test for foreign IPv4 addresses is done using the IPv4 address and netmask of the interface on that capture is being done. If that address or netmask are not available, either because the interface on that capture is being done has no address or netmask or because it is the any pseudo-interface (see the -i flag below), this option will not work correctly. so she might just ✨not✨ and you'll have to deal 😭

-F file — Use file as input for the filter expression. An additional expression given on the command line is ignored.
-g, --ip-oneline — Do not insert a line break after the IP header in verbose mode. one-line girlies rise 👑
-G rotate_seconds — If specified, rotates the dump file specified with the -w option every rotate_seconds seconds. Savefiles will have the name specified by -w which should include a time format as defined by strftime(3). If no time format is specified, each new file will overwrite the previous. Whenever a generated filename is not unique, tcpdump will overwrite the preexisting data; providing a time specification that is coarser than the capture period is therefore not advised 🚨

If used in conjunction with the -C option, filenames will take the form of file<count>.

note that when used with -Z option (enabled by default), privileges are dropped before opening the first savefile ⚠️

-h, --help — Print the tcpdump and libpcap version strings, print a usage message, and exit. help her help you 🫶

...and that's just part 1, bestie. there's more. there's SO much more 💖

--version — Prints out the version strings for both tcpdump and libpcap, then dips. That's it, that's the whole job. ✨
-H — Tries to sniff out 802.11s draft mesh headers. It's giving "attempting," bestie, so no promises 💅
-i interface, --interface=interface — This one is the main character of interface flags. It tells tcpdump to listen on interface, report the list of link-layer types, report the list of time stamp types, or report the results of compiling a filter expression on interface. If you don't specify it — and you also didn't pass -d — tcpdump goes shopping through the system interface list and picks the lowest numbered, configured up interface (loopback is not invited to this party 🚫), which might turn out to be, like, eth0.

On every supported Linux system, plus recent macOS and Solaris, handing over an interface argument of any means a special pseudo-interface that captures packets from all the regular network interfaces of the OS. What counts as a "regular" network interface is an implementation detail of the OS (on Linux, for example, this includes SocketCAN devices), so packets captured on the any pseudo-interface may represent way more different network protocols than you'd expect. Also, all regular network interfaces are just a subset of all available capture devices as printed by the -D flag. That bigger list may also include capture devices libpcap implements by means other than the OS network stack: Bluetooth, DAG, D-Bus, SNF and USB. And heads up — captures on the any pseudo-interface will not be done in promiscuous mode. Not the vibe for promiscuous mode, sorry. 🙈

If the -D flag is supported, an interface number as printed by that flag can be used as the interface argument — but only if no interface on the system already has that number as a name.

-I, --monitor-mode — Puts the interface into "monitor mode". This is supported only on IEEE 802.11 Wi-Fi interfaces, and only on some operating systems. Lowkey exclusive. 👑

Real talk: in monitor mode the adapter might disassociate from the network it's associated with, which means you won't be able to use any wireless networks with that adapter. That could stop you from accessing files on a network server, or resolving host names or network addresses, if you're capturing in monitor mode and you're not connected to another network with another adapter. So like… plan accordingly 🫶

This flag will affect the output of the -L flag. If -I isn't specified, only those link-layer types available when not in monitor mode get shown; if -I is specified, only those link-layer types available when in monitor mode get shown. It's giving either/or energy.

--immediate-mode — Capture in "immediate mode". In this mode, packets are delivered to tcpdump the instant they arrive, instead of being buffered for efficiency. This is the default when printing packets rather than saving packets to a savefile, if the packets are being printed to a terminal rather than to a file or pipe. ✨
-j tstamp_type, --time-stamp-type=tstamp_type — Sets the time stamp type for the capture to tstamp_type. The names to use for the time stamp types are listed in pcap-tstamp(7); not all the types listed there will necessarily be valid for any given interface, so don't come for the manpage when one doesn't work 😭
-J, --list-time-stamp-types — Lists the supported time stamp types for the interface and exits. If the time stamp type cannot be set for the interface, no time stamp types are listed at all. Empty list, no notes. 💀
--time-stamp-precision=tstamp_precision — When capturing, this sets the time stamp precision for the capture to tstamp_precision. Note that availability of high precision time stamps (nanoseconds) and their actual accuracy is platform and hardware dependent. Also note: when writing captures made with nanosecond accuracy to a savefile, the time stamps are written with nanosecond resolution, and the file is written with a different magic number, to indicate that the time stamps are in seconds and nanoseconds — and not all programs that read pcap savefiles will be able to read those captures. So, y'know, compatibility check first 🧚

When reading a savefile, this converts time stamps to the precision specified by timestamp_precision, and displays them with that resolution. If the precision specified is less than the precision of time stamps in the file, the conversion will lose precision. Sad but true.

The supported values for timestamp_precision are micro for microsecond resolution and nano for nanosecond resolution. The default is microsecond resolution.

--micro, --nano — Shorthands for --time-stamp-precision=micro or --time-stamp-precision=nano, adjusting the time stamp precision accordingly. When reading packets from a savefile, using --micro truncates time stamps if the savefile was created with nanosecond precision. In contrast, a savefile created with microsecond precision will have trailing zeroes added to the time stamp when --nano is used. Symmetry? Never heard of her. 💅
-K, --dont-verify-checksums — Don't attempt to verify IP, TCP, or UDP checksums. This is useful for interfaces that perform some or all of those checksum calculations in hardware; otherwise, all outgoing TCP checksums will be flagged as bad, and you'll be sitting there like "why is everything bad" when actually it's fine. 😭
-l — Makes stdout line buffered. Useful if you want to see the data while capturing it. E.g.,

tcpdump -l | tee dat

tcpdump -l > dat & tail -f dat

Note that on Windows, "line buffered" means "unbuffered", so WinDump will write each character individually if -l is specified. One. Character. At. A. Time. 💀

-U is similar to -l in its behavior, but it will cause output to be "packet-buffered", so the output is written to stdout at the end of each packet rather than at the end of each line; this is buffered on all platforms, including Windows.

-L, --list-data-link-types — Lists the known data link types for the interface, in the specified mode, and exits. The list of known data link types may be dependent on the specified mode; for example, on some platforms a Wi-Fi interface might support one set of data link types when not in monitor mode (maybe only fake Ethernet headers, or maybe 802.11 headers but not 802.11 headers with radio information) and a whole different set of data link types when in monitor mode (maybe 802.11 headers, or 802.11 headers with radio information, only in monitor mode). Mode check fr fr 🌈
-m module — Loads SMI MIB module definitions from file module. This option can be used several times to load several MIB modules into tcpdump. Stack 'em up 🍓
-M secret — Uses secret as a shared secret for validating the digests found in TCP segments with the TCP-MD5 option (RFC 2385), if present.
-n — Don't convert host addresses to names. This can be used to avoid DNS lookups. No lookups, no waiting, no drama. ✨
-nn — Don't convert protocol and port numbers etc. to names either. Double n, double refusal 💅
-N — Don't print domain name qualification of host names. E.g., if you give this flag then tcpdump will print nic instead of nic.ddn.mil. Short and sweet, the way that she abbreviates 👑
-#, --number — Prints a packet number at the beginning of the line. Numbering the girlies 1, 2, 3… 🔥
-O, --no-optimize — Do not run the packet-matching code optimizer. This is useful only if you suspect a bug in the optimizer. So like, unless you're hunting a bug, leave it alone bestie 🙈
-p, --no-promiscuous-mode — Don't put the interface into promiscuous mode. Note that the interface might be in promiscuous mode for some other reason; hence, -p cannot be used as an abbreviation for ether host {local-hw-addr} or ether broadcast. Important distinction, no cap.
--print — Prints parsed packet output, even if the raw packets are being saved to a file with the -w flag. Both. At once. Multitasking icon. ✨
-Q direction, --direction=direction — Chooses the send/receive direction direction for which packets should be captured. Possible values are in, out and inout. Not available on all platforms. 🦋
-q — Quick (quiet?) output. Prints less protocol information so output lines are shorter. It's giving minimalism, but for packets.
-r file — Reads packets from file (which was created with the -w option or by other tools that write pcap or pcapng files). Standard input is used if file is -. Glow up your old captures 💖
-S, --absolute-tcp-sequence-numbers — Prints absolute, rather than relative, TCP sequence numbers.
-s snaplen, --snapshot-length=snaplen — Snarfs snaplen bytes of data from each packet rather than the default of 262144 bytes. Packets truncated because of a limited snapshot are indicated in the output with [|proto], where proto is the name of the protocol level at which the truncation has occurred.

Note that taking larger snapshots both increases the amount of time it takes to process packets and, effectively, decreases the amount of packet buffering. This may cause packets to be lost. Note also that taking smaller snapshots will discard data from protocols above the transport layer, which loses information that may be important. NFS and AFS requests and replies, for example, are very large, and much of the detail won't be available if a too-short snapshot length is selected. So pick your struggle 😭

If you need to reduce the snapshot size below the default, you should limit snaplen to the smallest number that will capture the protocol information you're interested in. Setting snaplen to 0 sets it to the default of 262144, for backwards compatibility with recent older versions of tcpdump.

-T type — Forces packets selected by "expression" to be interpreted as the specified type. Currently known types are:
- aodv — Ad-hoc On-demand Distance Vector protocol
- carp — Common Address Redundancy Protocol
- cnfp — Cisco NetFlow protocol
- domain — Domain Name System
- lmp — Link Management Protocol
- pgm — Pragmatic General Multicast
- pgm_zmtp1 — ZMTP/1.0 inside PGM/EPGM
- ptp — Precision Time Protocol
- radius — RADIUS
- resp — REdis Serialization Protocol
- rpc — Remote Procedure Call
- rtcp — Real-Time Applications control protocol
- rtp — Real-Time Applications protocol
- snmp — Simple Network Management Protocol
- someip — SOME/IP
- tftp — Trivial File Transfer Protocol
- vat — Visual Audio Tool
- vxlan — Virtual eXtensible Local Area Network
- wb — distributed White Board
- zmtp1 — ZeroMQ Message Transport Protocol 1.0

Look at that roster, she's got range 👑

Okay so quick tea about the pgm type — it only messes with how we read UDP. The native PGM is always recognised as IP protocol 113 no matter what. And UDP-encapsulated PGM? That's the thing people call "EPGM" or "PGM/UDP". ✨

And the pgm_zmtp1 type up there? She's doing double duty — she affects interpretation of native PGM and UDP at the same time, bestie. During native PGM decoding, the application data of an ODATA/RDATA packet gets decoded as a ZeroMQ datagram with ZMTP/1.0 frames. During UDP decoding, on top of all that, any UDP packet gets treated as an encapsulated PGM packet. 🫶

-t — No timestamp on each dump line. Simple as that, we're keeping it clean. 💅
-tt — Print the timestamp as seconds since January 1, 1970, 00:00:00, UTC, plus fractions of a second since then, on each dump line. Epoch girlie behaviour. ⏰
-ttt — Print a delta between the current and previous line on each dump line. Resolution is microsecond or nanosecond depending on the --time-stamp-precision option. Default is microsecond resolution. ✨
-tttt — Print a timestamp as hours, minutes, seconds, and fractions of a second since midnight, with the date in front, on each dump line. 🌸
-ttttt — Print a delta between the current and first line on each dump line. Again, resolution is microsecond or nanosecond depending on the --time-stamp-precision option, and the default is microsecond resolution. ⭐
-u — Print undecoded NFS handles. No decoding, we show it raw. 🙈
-U, --packet-buffered — Okay so this one has two whole moods depending on your flags:
- If the -w option is not specified, or if it is specified but the --print flag is also specified: make the printed packet output packet-buffered. Meaning — as the description of the contents of each packet is printed, it gets written to standard output, rather than (when not writing to a terminal) only being written when the output buffer fills. 💖
- If the -w option is specified: make the saved raw packet output packet-buffered. Meaning — as each packet is saved, it gets written to the output file, rather than only being written when the output buffer fills. 🎀

Heads up bestie: the -U flag will not be supported if tcpdump was built with an older version of libpcap that lacks the pcap_dump_flush(3PCAP) function. 😭

-v — When parsing and printing, produce (slightly more) verbose output. For example, the time to live, identification, total length and options in an IP packet get printed. It also enables additional packet integrity checks, like verifying the IP and ICMP header checksum. Extra detail, extra slay.

Also! When writing to a file with the -w option and at the same time not reading from a file with the -r option, it reports to stderr, once per second, the number of packets captured. In Solaris, FreeBSD and possibly other operating systems, this periodic update currently can cause loss of captured packets on their way from the kernel to tcpdump. 💀

-vv — Even more verbose output. For example, additional fields get printed from NFS reply packets, and SMB packets are fully decoded. She's thorough. 👑
-vvv — Even more verbose output. For example, telnet SB ... SE options are printed in full. With -X, Telnet options are printed in hex as well. 🔥
-V file — Read a list of filenames from file. Standard input is used if file is -. 📋
-w file — Write the raw packets to file rather than parsing and printing them out. They can later be printed with the -r option. Standard output is used if file is -. 💾

This output will be buffered if written to a file or pipe, so a program reading from the file or pipe may not see packets for an arbitrary amount of time after they are received. Use the -U flag to cause packets to be written as soon as they are received. ⚡

The MIME type application/vnd.tcpdump.pcap has been registered with IANA for pcap files. The filename extension .pcap appears to be the most commonly used, along with .cap and .dmp. tcpdump itself doesn't check the extension when reading capture files and doesn't add an extension when writing them (it uses magic numbers in the file header instead). However, many operating systems and applications will use the extension if it is present, and adding one (e.g. .pcap) is recommended. 🌈

See pcap-savefile(5) for a description of the file format. 📖

-W filecount — Used together with the -C option, this limits the number of files created to the specified number, and begins overwriting files from the beginning, thus creating a 'rotating' buffer. On top of that, it names the files with enough leading 0s to support the maximum number of files, allowing them to sort correctly. 🔄

Used together with the -G option, this limits the number of rotated dump files that get created, exiting with status 0 when reaching the limit. ✅

If used with both -C and -G, the -W option will currently be ignored, and will only affect the file name. 🙃

-x — When parsing and printing, in addition to printing the headers of each packet, print the data of each packet (minus its link level header) in hex. The smaller of the entire packet or snaplen bytes will be printed. Note that this is the entire link-layer packet, so for link layers that pad (e.g. Ethernet), the padding bytes will also be printed when the higher layer packet is shorter than the required padding. In the current implementation this flag may have the same effect as -xx if the packet is truncated. No effect when -X[X] option is used. 🔍
-xx — When parsing and printing, in addition to printing the headers of each packet, print the data of each packet, including its link level header, in hex. No effect when -X[X] option is used. 🔍
-X — When parsing and printing, in addition to printing the headers of each packet, print the data of each packet (minus its link level header) in hex and ASCII. This is very handy for analysing new protocols. In the current implementation this flag may have the same effect as -XX if the packet is truncated. 🧐
-XX — When parsing and printing, in addition to printing the headers of each packet, print the data of each packet, including its link level header, in hex and ASCII. 🧐
-y datalinktype, --linktype=datalinktype — Set the data link type to use while capturing packets (see -L) or just compiling and dumping packet-matching code (see -d) to datalinktype. 🔗
-z postrotate-command — Used together with the -C or -G options, this makes tcpdump run " postrotate-command file " where file is the savefile being closed after each rotation. For example, specifying -z gzip or -z bzip2 will compress each savefile using gzip or bzip2. 🗜️

This option can only be used if fork subprocess is implemented (e.g. not on Windows systems). 🚫

Note that tcpdump will run the command in parallel to the capture, using the lowest priority so that this doesn't disturb the capture process. 🫧

And in case you would like to use a command that itself takes flags or different arguments, you can always write a shell script that will take the savefile name as the only argument, make the flags & arguments arrangements and execute the command that you want. 💅

-Z user, --relinquish-privileges=user — If tcpdump is running as root, after opening the capture device or input savefile, but before opening any savefiles for output, change the user ID to user and the group ID to the primary group of user. 🔐

This behavior is enabled by default (-Z tcpdump), and can be disabled by -Z root. ✨

expression — selects which packets will be dumped. If no expression is given, all packets on the net will be dumped. Otherwise, only packets for which expression is true will be dumped. 🎯

For the expression syntax, see pcap-filter(7). 📚

The expression argument can be passed to tcpdump as either a single Shell argument, or as multiple Shell arguments, whichever is more convenient. Generally, if the expression contains Shell metacharacters, such as backslashes used to escape protocol names, it is easier to pass it as a single, quoted argument rather than to escape the Shell metacharacters. Multiple arguments are concatenated with spaces before being parsed. 🌸

💖 EXAMPLES 💖

This is the part where we actually do the thing, bestie. Let's gooo.

To print all packets arriving at or departing from sundown:

tcpdump host sundown

To print traffic between helios and either hot or ace:

tcpdump host helios and \( hot or ace \)

To print all IP packets between ace and any host except helios:

tcpdump ip host ace and not helios

To print all traffic between local hosts and hosts at Berkeley:

tcpdump net ucb-ether

To print all ftp traffic through internet gateway snup: (note that the expression is quoted to prevent the shell from (mis-)interpreting the parentheses):

tcpdump 'gateway snup and (port ftp or ftp-data)'

To print traffic neither sourced from nor destined for local hosts (if you gateway to one other net, this stuff should never make it onto your local net):

tcpdump ip and not net localnet

To print the start and end packets (the SYN and FIN packets) of each TCP conversation that involves a non-local host:

tcpdump 'tcp[tcpflags] & (tcp-syn|tcp-fin) != 0 and not src and dst net localnet'

To print the TCP packets with flags RST and ACK both set. (i.e. select only the RST and ACK flags in the flags field, and if the result is "RST and ACK both set", match):

tcpdump 'tcp[tcpflags] & (tcp-rst|tcp-ack) == (tcp-rst|tcp-ack)'

To print all IPv4 HTTP packets to and from port 80, i.e. print only packets that contain data, not, for example, SYN and FIN packets and ACK-only packets. (IPv6 is left as an exercise for the reader.)

tcpdump 'tcp port 80 and (((ip[2:2] - ((ip[0]&0xf)<<2)) - ((tcp[12]&0xf0)>>2)) != 0)'

To print IP packets longer than 576 bytes sent through gateway snup:

tcpdump 'gateway snup and ip[2:2] > 576'

To print IP broadcast or multicast packets that were not sent via Ethernet broadcast or multicast:

tcpdump 'ether[0] & 1 = 0 and ip[16] >= 224'

To print all ICMP packets that are not echo requests/replies (i.e., not ping packets):

tcpdump 'icmp[icmptype] != icmp-echo and icmp[icmptype] != icmp-echoreply'

🌈 OUTPUT FORMAT 🌈

okay bestie so here's the tea: the output of tcpdump is protocol dependent, fr fr. What you get on screen depends on what you're sniffing. Below is a quick rundown plus examples of most of the formats you'll see. 💅

⏰ Timestamps

By default, every single output line gets a timestamp glued to the front. It's the current clock time, looking like this:

hh:mm:ss.frac

and it's as accurate as the kernel's clock is — that's the whole vibe. 🕐

The timestamp reflects the time the kernel slapped a time stamp on the packet. And ngl, no attempt is made to account for the time lag between when the network interface finished receiving the packet from the network and when the kernel applied a time stamp to it. That lag could include a delay between when the network interface finished receiving a packet from the network and when an interrupt got delivered to the kernel to make it go read the packet, plus a delay between when the kernel serviced the new packet interrupt and when it applied a time stamp to the packet. So like. it's a whole journey. 😭

🌐 Interface

When the any interface is selected on capture, or when a link-type LINUX_SLL2 capture file is read, the interface name gets printed after the timestamp. ✨

Then comes the packet type, where In and Out mean a packet destined for this host or originating from this host respectively. Other possible values are:

B — broadcast packets
M — multicast packets
P — packets destined for other hosts

🔗 Link Level Headers

If you pass the -e option, the link level header gets printed out. On Ethernets, that means the source and destination addresses, protocol, and packet length are all printed. iconic.

On FDDI networks, the -e option makes tcpdump print the frame control field, the source and destination addresses, and the packet length. (Quick aside: the frame control field governs how the rest of the packet gets interpreted. Normal packets — like the ones carrying IP datagrams — are async packets, with a priority value between 0 and 7; for example, async4. Those packets are assumed to contain an 802.2 Logical Link Control (LLC) packet, and the LLC header is printed if it's not an ISO datagram or a so-called SNAP packet.)

On Token Ring networks, the -e option makes tcpdump print the access control and frame control fields, the source and destination addresses, and the packet length. Just like on FDDI networks, packets are assumed to contain an LLC packet. And here's the thing — whether or not you specify -e, the source routing information is printed for source-routed packets. no opting out of that one. 💀

On 802.11 networks, the -e option makes tcpdump print the frame control fields, all of the addresses in the 802.11 header, and the packet length. And yeah, same as FDDI — packets are assumed to contain an LLC packet.

N.B.: the following description assumes you're familiar with the SLIP compression algorithm described in RFC 1144. if you're not, this next bit is gonna be a lot. 🫶

On SLIP links, a direction indicator (I for inbound, O for outbound), packet type, and compression information all get printed out. The packet type comes first. The three types are ip, utcp, and ctcp. No further link information is printed for ip packets. For TCP packets, the connection identifier is printed following the type. If the packet is compressed, its encoded header is printed out. The special cases get printed out as *S+n and *SA+n, where n is the amount by which the sequence number (or sequence number and ack) has changed. If it's not a special case, zero or more changes are printed. A change is indicated by U (urgent pointer), W (window), A (ack), S (sequence number), and I (packet ID), followed by a delta (+n or -n), or a new value (=n). And finally, the amount of data in the packet and the compressed header length are printed. periodt. 💅

For example, the following line shows an outbound compressed TCP packet, with an implicit connection identifier; the ack has changed by 6, the sequence number by 49, and the packet ID by 6; there are 3 bytes of data and 6 bytes of compressed header:

O ctcp * A+6 S+49 I+6 3 (6)

📮 ARP/RARP Packets

ARP/RARP output shows the type of request and its arguments. The format is meant to be self explanatory, honestly. Here's a short sample taken from the start of an rlogin from host rtsg to host csam:

arp who-has csam tell rtsg
arp reply csam is-at CSAM

The first line says that rtsg sent an ARP packet asking for the Ethernet address of internet host csam. Csam replies with its Ethernet address — and in this example, Ethernet addresses are in caps while internet addresses are in lower case. cute. 🎀

This would look less redundant if we had done tcpdump -n:

arp who-has 128.3.254.6 tell 128.3.254.68
arp reply 128.3.254.6 is-at 02:07:01:00:01:c4

And if we had done tcpdump -e, the fact that the first packet is broadcast and the second is point-to-point would be visible:

RTSG Broadcast 0806  64: arp who-has csam tell rtsg
CSAM RTSG 0806  64: arp reply csam is-at CSAM

For the first packet this says the Ethernet source address is RTSG, the destination is the Ethernet broadcast address, the type field contained hex 0806 (type ETHER_ARP) and the total length was 64 bytes. ✨

🧊 IPv4 Packets

If the link-layer header isn't being printed, then for IPv4 packets, IP is printed after the time stamp.

If the -v flag is specified, information from the IPv4 header is shown in parentheses after the IP or the link-layer header. The general format of this information is:

tos tos, ttl ttl, id id, offset offset, flags [flags], proto proto, length length, options (options)

Let's break that down bestie:

tos is the type of service field; if the ECN bits are non-zero, those get reported as ECT(1), ECT(0), or CE.
ttl is the time-to-live; it is not reported if it's zero.
id is the IP identification field.
offset is the fragment offset field; it gets printed whether or not this is part of a fragmented datagram.
flags are the MF and DF flags; + is reported if MF is set, and DF is reported if F is set. If neither are set, . is reported.
proto is the protocol ID field.
length is the total length field; if the packet is a presumed TSO (TCP Segmentation Offload) send, [was 0, presumed TSO] is reported.
options are the IP options, if there are any.

Next, for TCP and UDP packets, the source and destination IP addresses and TCP or UDP ports get printed, with a dot between each IP address and its corresponding port, and a > separating the source and destination. For other protocols, the addresses get printed, with a > separating the source and destination. Higher level protocol information, if any, gets printed after that. 🦋

For fragmented IP datagrams, the first fragment contains the higher level protocol header; fragments after the first contain no higher level protocol header. Fragmentation information is printed only with the -v flag, in the IP header information, as described above. don't forget the -v, iykyk.

🔥 TCP Packets

N.B.: the following description assumes familiarity with the TCP protocol described in RFC 793. If you are not familiar with the protocol, this description will not be of much use to you. like, at all. 😭

The general format of a TCP protocol line is:

src > dst: Flags [tcpflags], seq data-seqno, ack ackno, win window, urg urgent, options [opts], length len

Src and dst are the source and destination IP addresses and ports. Tcpflags are some combination of S (SYN), F (FIN), P (PSH), R (RST), U (URG), W (CWR), E (ECE) or . (ACK), or none if no flags are set. Data-seqno describes the portion of sequence space covered by the data in this packet (see the example below). Ackno is the sequence number of the next data expected the other direction on this connection. Window is the number of bytes of receive buffer space available the other direction on this connection. Urg indicates there's urgent data in the packet. Opts are TCP options (e.g., mss 1024). Len is the length of payload data. 💅

Iptype, Src, dst, and flags are always present. The other fields depend on the contents of the packet's TCP protocol header and are output only if appropriate. no notes.

Here's the opening portion of an rlogin from host rtsg to host csam:

IP rtsg.1023 > csam.login: Flags [S], seq 768512:768512, win 4096, opts [mss 1024]
IP csam.login > rtsg.1023: Flags [S.], seq, 947648:947648, ack 768513, win 4096, opts [mss 1024]
IP rtsg.1023 > csam.login: Flags [.], ack 1, win 4096
IP rtsg.1023 > csam.login: Flags [P.], seq 1:2, ack 1, win 4096, length 1
IP csam.login > rtsg.1023: Flags [.], ack 2, win 4096
IP rtsg.1023 > csam.login: Flags [P.], seq 2:21, ack 1, win 4096, length 19
IP csam.login > rtsg.1023: Flags [P.], seq 1:2, ack 21, win 4077, length 1
IP csam.login > rtsg.1023: Flags [P.], seq 2:3, ack 21, win 4077, urg 1, length 1
IP csam.login > rtsg.1023: Flags [P.], seq 3:4, ack 21, win 4077, urg 1, length 1

okay so let's actually read this, because it's giving textbook opening-handshake energy:

The first line says that TCP port 1023 on rtsg sent a packet to port login on csam. The S indicates that the SYN flag was set. The packet sequence number was 768512 and it contained no data. (The notation is first:last which means sequence numbers first up to but not including last.) There was no piggy-backed ACK, the available receive window was 4096 bytes and there was a max-segment-size option requesting an MSS of 1024 bytes. ✨

Csam replies with a similar packet, except it includes a piggy-backed ACK for rtsg's SYN. Rtsg then ACKs csam's SYN. The . means the ACK flag was set. The packet contained no data, so there's no data sequence number or length. Note that the ACK sequence number is a small integer (1). Here's the important bit, bestie: the first time tcpdump sees a TCP conversation, it prints the sequence number from the packet. On subsequent packets of the conversation, the difference between the current packet's sequence number and this initial sequence number is printed. That means sequence numbers after the first can be interpreted as relative byte positions in the conversation's data stream (with the first data byte each direction being 1). And -S will override this feature, causing the original sequence numbers to be output. 🔥

On the 6th line, rtsg sends csam 19 bytes of data (bytes 2 through 20 in the rtsg → csam side of the conversation). The PSH flag is set in the packet. On the 7th line, csam says it's received data sent by rtsg up to but not including byte 21. Most of this data is apparently sitting in the socket buffer since csam's receive window has gotten 19 bytes smaller. Csam also sends one byte of data to rtsg in this packet. And on the 8th and 9th lines, csam sends two bytes of urgent, pushed data to rtsg. urgent pushed data?? the drama. 💅

If the snapshot was sooo tiny that tcpdump couldn't even grab the full TCP header, it does its best with whatever it got and then hits you with [|tcp] — basically "bestie, this rest is unreadable, I tried 💀".

And if the header has a bogus option (one whose length is either too small or goes past the end of the header), tcpdump says [bad opt] and just stops interpreting options from there, because honestly there's no way to tell where the next one even starts. And if the header length says "hey, options are present!" but the IP datagram length is like... not long enough for those options to actually exist, tcpdump reports [bad hdr length]. Caught in a lie 💅.

🚩 Particular TCP Flag Combinations (SYN-ACK, URG-ACK, etc.)

Okay so the control bits section of the TCP header has exactly 8 bits:

CWR | ECE | URG | ACK | PSH | RST | SYN | FIN

Let's say we wanna watch packets from the start of a TCP connection. Remember, TCP does the whole 3-way handshake thing when it spins up a new connection, and with the TCP control bits it goes:

1) Caller sends SYN
2) Recipient responds with SYN, ACK
3) Caller sends ACK

Now bestie, we want only the packets where the SYN bit is set (that's Step 1). And no, we do not want the step 2 SYN-ACK packets — just a plain, unbothered, moisturized initial SYN. So we need a correct filter expression for tcpdump. ✨

Recall the structure of a TCP header without options:

 0                            15                              31
-----------------------------------------------------------------
|          source port          |       destination port        |
-----------------------------------------------------------------
|                        sequence number                        |
-----------------------------------------------------------------
|                     acknowledgment number                     |
-----------------------------------------------------------------
|  HL   | rsvd  |C|E|U|A|P|R|S|F|        window size            |
-----------------------------------------------------------------
|         TCP checksum          |       urgent pointer          |
-----------------------------------------------------------------

A TCP header usually holds 20 octets of data, unless options are present (options are the ✨extra✨). The first line of the graph is octets 0 - 3, the second line shows octets 4 - 7, and so on.

Starting to count with 0, the relevant TCP control bits live in octet 13:

 0             7|             15|             23|             31
----------------|---------------|---------------|----------------
|  HL   | rsvd  |C|E|U|A|P|R|S|F|        window size            |
----------------|---------------|---------------|----------------
|               |  13th octet   |               |               |

Let's zoom in on octet no. 13, main character behavior:

                |               |
                |---------------|
                |C|E|U|A|P|R|S|F|
                |---------------|
                |7   5   3     0|

These are the TCP control bits we care about. We numbered the bits in this octet from 0 to 7, right to left, so PSH is bit number 3, and URG is number 5.

Now recall: we want packets with only SYN set. Let's see what happens to octet 13 when a TCP datagram shows up with the SYN bit set in its header:

                |C|E|U|A|P|R|S|F|
                |---------------|
                |0 0 0 0 0 0 1 0|
                |---------------|
                |7 6 5 4 3 2 1 0|

Looking at the control bits section, only bit number 1 (SYN) is set. Slay.

Assuming octet number 13 is an 8-bit unsigned integer in network byte order, the binary value of this octet is

00000010

and its decimal representation is

   7     6     5     4     3     2     1     0
0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 0*2 + 1*2 + 0*2  =  2

We're almost done 🎀 because now we know that if only SYN is set, the value of the 13th octet in the TCP header, interpreted as an 8-bit unsigned integer in network byte order, has to be exactly 2.

This relationship can be expressed as

tcp[13] == 2

We can use that as our filter for tcpdump to watch packets that have only SYN set:

tcpdump -i xl0 'tcp[13] == 2'

The expression says "let the 13th octet of a TCP datagram have the decimal value 2", which is exactly what we want. No notes. 👑

Now, let's assume we need to capture SYN packets, but we don't care if ACK or any other TCP control bit is set at the same time. Let's see what happens to octet 13 when a TCP datagram with SYN-ACK set arrives:

     |C|E|U|A|P|R|S|F|
     |---------------|
     |0 0 0 1 0 0 1 0|
     |---------------|
     |7 6 5 4 3 2 1 0|

Now bits 1 and 4 are set in the 13th octet. The binary value of octet 13 is

00010010

which translates to decimal

   7     6     5     4     3     2     1     0
0*2 + 0*2 + 0*2 + 1*2 + 0*2 + 0*2 + 1*2 + 0*2   = 18

Now we can't just use tcp[13] == 18 in the tcpdump filter expression, because that would select only those packets that have SYN-ACK set — and not the ones with only SYN set. Remember, we don't care if ACK or any other control bit is set, as long as SYN is set. Periodt.

So to get what we want, we need to logically AND the binary value of octet 13 with some other value to preserve the SYN bit. We know we want SYN set no matter what, so we'll logically AND the value in the 13th octet with the binary value of a SYN:

          00010010 SYN-ACK              00000010 SYN
     AND  00000010 (we want SYN)   AND  00000010 (we want SYN)
          --------                      --------
     =    00000010                 =    00000010

See how this AND operation delivers the same result whether ACK or another TCP control bit is set? Iconic. The decimal representation of the AND value, and the result of this operation, is 2 (binary 00000010), so we know that for packets with SYN set, this relation must hold true:

( ( value of octet 13 ) AND ( 2 ) ) == ( 2 )

Which points us right at the tcpdump filter expression

tcpdump -i xl0 'tcp[13] & 2 == 2'

Some offsets and field values can be expressed as names instead of numeric values. For example, tcp[13] can be replaced with tcp[tcpflags]. And these TCP flag field values are available too: tcp-fin, tcp-syn, tcp-rst, tcp-push, tcp-ack, tcp-urg, tcp-ece and tcp-cwr.

Demonstrated like this:

tcpdump -i xl0 'tcp[tcpflags] & tcp-push != 0'

⚠️ Bestie, use single quotes or a backslash in the expression to hide the AND (&) special character from the shell, or the shell will snatch it and ruin your whole vibe.

🦋 UDP Packets

UDP format is illustrated by this rwho packet:

actinide.who > broadcast.who: udp 84

This says that port who on host actinide sent a UDP datagram to port who on host broadcast, the Internet broadcast address. The packet contained 84 bytes of user data. Cute, concise, we love her. 💖

Some UDP services are recognized (from the source or destination port number) and the higher level protocol information gets printed. Specifically: Domain Name service requests (RFC 1034/1035) and Sun RPC calls (RFC 1050) to NFS.

🌸 TCP or UDP Name Server Requests

(N.B.: The following description assumes you're familiar with the Domain Service protocol described in RFC 1035. If you're not familiar with the protocol, the following description will appear to be written in Greek. 😭)

Name server requests are formatted as

src > dst: id op? flags qtype qclass name (len)
h2opolo.1538 > helios.domain: 3+ A? ucbvax.berkeley.edu. (37)

Host h2opolo asked the domain server on helios for an address record (qtype=A) associated with the name ucbvax.berkeley.edu. The query id was 3. The + means the recursion desired flag was set. The query length was 37 bytes — and that's excluding the TCP or UDP and IP protocol headers, so don't @ me. The query operation was the normal one, Query, so the op field was omitted. If the op had been anything else, it would've been printed between the 3 and the +. Same energy with qclass: it was the normal one, C_IN, and omitted. Any other qclass would've been printed immediately after the A.

A few anomalies get checked and may result in extra fields in square brackets: if a query contains an answer, authority records or additional records section, then ancount, nscount, or arcount are printed as [na], [nn] or [nau] where n is the appropriate count. And if any of the response bits are set (AA, RA or rcode) or any of the must be zero bits are set in bytes two and three, [b2&3=x] is printed, where x is the hex value of header bytes two and three. 👀

🧚 TCP or UDP Name Server Responses

Name server responses are formatted as

src > dst:  id op rcode flags a/n/au type class data (len)
helios.domain > h2opolo.1538: 3 3/3/7 A 128.32.137.3 (273)
helios.domain > h2opolo.1537: 2 NXDomain* 0/1/0 (97)

In the first example, helios responds to query id 3 from h2opolo with 3 answer records, 3 name server records and 7 additional records. The first answer record is type A (address) and its data is internet address 128.32.137.3. The total size of the response was 273 bytes, excluding TCP or UDP and IP headers. The op (Query) and response code (NoError) were omitted, as was the class (C_IN) of the A record. She's efficient like that. 💅

In the second example, helios responds to query 2 with a response code of nonexistent domain (NXDomain) with no answers, one name server and no authority records. The * means the authoritative answer bit was set. And since there were no answers, no type, class or data were printed. Nothing to show, so she didn't. ✨

Other flag characters that might show up are - (recursion available, RA, not set) and | (truncated message, TC, set). If the question section doesn't contain exactly one entry, [nq] is printed.

🍓 SMB/CIFS Decoding

tcpdump now includes fairly extensive SMB/CIFS/NBT decoding for data on UDP/137, UDP/138 and TCP/139. Some primitive decoding of IPX and NetBEUI SMB data is also done. Lowkey a whole lil ecosystem in here. 🌈

By default a fairly minimal decode is done, with a much more detailed decode happening if -v is used. And bestie, be warned: with -v a single SMB packet may take up a page or more, so only use -v if you really want all the gory details. 😭

For information on SMB packet formats and what all the fields mean, see https://download.samba.org/pub/samba/specs/ and other online resources. The SMB patches were written by Andrew Tridgell (tridge@samba.org). Shoutout to the legend. 🫶

✨ NFS Requests and Replies ✨

Sun NFS (Network File System) requests and replies get printed like this, bestie:

src.sport > dst.nfs: NFS request xid xid len op args
src.nfs > dst.dport: NFS reply xid xid reply stat len op results

and in real life it looks like 👇

sushi.1023 > wrl.nfs: NFS request xid 26377
     112 readlink fh 21,24/10.73165
wrl.nfs > sushi.1023: NFS reply xid 26377
     reply ok 40 readlink "../var"
sushi.1022 > wrl.nfs: NFS request xid 8219
     144 lookup fh 9,74/4096.6878 "xcolors"
wrl.nfs > sushi.1022: NFS reply xid 8219
     reply ok 128 lookup fh 9,74/4134.3150

So line one? Host sushi is sending a transaction with id 26377 over to wrl. The request was 112 bytes, and that's excluding the UDP and IP headers, okay? The operation was a readlink (that's "read symbolic link" if you were wondering 🫶) on file handle (fh) 21,24/10.731657119. And if you're lucky — like, genuinely blessed — that file handle can be read as a major,minor device number pair, followed by the inode number and generation number. Then line two is wrl replying ok with the exact same transaction id and the contents of the link. Slay. 💅

Line three, sushi asks (with a brand new transaction id, main character energy) for wrl to lookup the name xcolors in directory file 9,74/4096.6878. And line four? wrl sends a reply with the matching transaction id. Periodt.

Heads up though: the data that gets printed depends on the operation type. The format is meant to be self explanatory if you read it alongside an NFS protocol spec. Also ngl, older versions of tcpdump printed NFS packets in a slightly different format — the transaction id (xid) would be printed instead of the non-NFS port number of the packet.

If the -v (verbose) flag is given, you get extra info printed. Like, for example:

sushi.1023 > wrl.nfs: NFS request xid 79658
     148 read fh 21,11/12.195 8192 bytes @ 24576
wrl.nfs > sushi.1023: NFS reply xid 79658
     reply ok 1472 read REG 100664 ids 417/0 sz 29388

(-v also prints the IP header TTL, ID, length, and fragmentation fields, which got omitted from this example, no cap.) In that first line, sushi asks wrl to read 8192 bytes from file 21,11/12.195, at byte offset 24576. Then wrl replies ok; the packet on the second line is the first fragment of the reply, which is why it's only 1472 bytes long (the rest of the bytes will follow in subsequent fragments, but those fragments don't have NFS or even UDP headers and so might not be printed, depending on the filter expression you used). And because the -v flag is there, some of the file attributes (which come back in addition to the file data) get printed: the file type (``REG'', for regular file), the file mode (in octal), the UID and GID, and the file size. Obsessed. 🌸

If the -v flag is given more than once, even more details get printed. ✨

NFS reply packets do not explicitly identify the RPC operation. Instead, tcpdump keeps track of ``recent'' requests, and matches them to the replies using the transaction ID. So if a reply doesn't closely follow the corresponding request, it might not be parsable. 💀

🦋 AFS Requests and Replies 🦋

Transarc AFS (Andrew File System) requests and replies get printed as:

src.sport > dst.dport: rx packet-type
src.sport > dst.dport: rx packet-type service call call-name args
src.sport > dst.dport: rx packet-type service reply call-name args
elvis.7001 > pike.afsfs:
     rx data fs call rename old fid 536876964/1/1 ".newsrc.new"
     new fid 536876964/1/1 ".newsrc"
pike.afsfs > elvis.7001: rx data fs reply rename

First line: host elvis sends a RX packet to pike. It was a RX data packet to the fs (fileserver) service, and it's the start of an RPC call. That RPC call was a rename, with the old directory file id of 536876964/1/1 and an old filename of .newsrc.new, and a new directory file id of 536876964/1/1 and a new filename of .newsrc. Then host pike responds with a RPC reply to the rename call — and it was successful, because it was a data packet and not an abort packet. 🎀

In general, all AFS RPCs are decoded at least by RPC call name. Most AFS RPCs have at least <em>some</em> of the arguments decoded (generally only the interesting arguments, for some definition of interesting, iykyk).

The format is meant to be self-describing, but it will probably not be useful to people who aren't familiar with the workings of AFS and RX. 🙈

If the -v (verbose) flag is given twice, acknowledgement packets and additional header information get printed, such as the RX call ID, call number, sequence number, serial number, and the RX packet flags.

If the -v flag is given twice, additional information is printed, such as the RX call ID, serial number, and the RX packet flags. The MTU negotiation information is also printed from RX ack packets.

If the -v flag is given three times, the security index and service id get printed.

Error codes are printed for abort packets, with the exception of Ubik beacon packets (because abort packets are used to signify a yes vote for the Ubik protocol).

AFS reply packets do not explicitly identify the RPC operation. Instead, tcpdump keeps track of ``recent'' requests, and matches them to the replies using the call number and service ID. If a reply doesn't closely follow the corresponding request, it might not be parsable. 😭

🌈 KIP AppleTalk (DDP in UDP) 🌈

AppleTalk DDP packets that are encapsulated in UDP datagrams get de-encapsulated and dumped as DDP packets (i.e., all the UDP header information is discarded, bye 👋). The file /etc/atalk.names is used to translate AppleTalk net and node numbers to names. Lines in this file have the form

number    name

1.254          ether
16.1      icsd-net
1.254.110 ace

The first two lines give the names of AppleTalk networks. The third line gives the name of a particular host (a host is distinguished from a net by the 3rd octet in the number — a net number must have two octets and a host number must have three octets.) The number and name should be separated by whitespace (blanks or tabs). And the /etc/atalk.names file may contain blank lines or comment lines (lines starting with a #).

AppleTalk addresses get printed in the form

net.host.port

144.1.209.2 > icsd-net.112.220
office.2 > icsd-net.112.220
jssmag.149.235 > icsd-net.2

(If the /etc/atalk.names doesn't exist or doesn't contain an entry for some AppleTalk host/net number, addresses are printed in numeric form.)

In that first example, NBP (DDP port 2) on net 144.1 node 209 is sending to whatever is listening on port 220 of net icsd node 112. The second line is the same except the full name of the source node is known (office). The third line is a send from port 235 on net jssmag node 149 to broadcast on the icsd-net NBP port — and note that the broadcast address (255) is indicated by a net name with no host number. Which is exactly why it's a good idea to keep node names and net names distinct in /etc/atalk.names. 💅

NBP (name binding protocol) and ATP (AppleTalk transaction protocol) packets have their contents interpreted. Other protocols just dump the protocol name (or number if no name is registered for the protocol) and packet size.

🎀 NBP Packets 🎀

NBP packets are formatted like these examples:

icsd-net.112.220 > jssmag.2: nbp-lkup 190: "=:LaserWriter@*"
jssmag.209.2 > icsd-net.112.220: nbp-reply 190: "RM1140:LaserWriter@*" 250
techpit.2 > icsd-net.112.220: nbp-reply 190: "techpit:LaserWriter@*" 186

The first line is a name lookup request for laserwriters sent by net icsd host 112 and broadcast on net jssmag. The nbp id for the lookup is 190. The second line shows a reply for this request (and notice it has the same id) from host jssmag.209 saying it has a laserwriter resource named "RM1140" registered on port 250. The third line is another reply to the same request, saying host techpit has laserwriter "techpit" registered on port 186. Cute, right? 🍓

⭐ ATP Packets ⭐

ATP packet formatting is demonstrated by this example:

jssmag.209.165 > helios.132: atp-req  12266<0-7> 0xae030001
helios.132 > jssmag.209.165: atp-resp 12266:0 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:1 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:2 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:4 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:6 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp*12266:7 (512) 0xae040000
jssmag.209.165 > helios.132: atp-req  12266<3,5> 0xae030001
helios.132 > jssmag.209.165: atp-resp 12266:3 (512) 0xae040000
helios.132 > jssmag.209.165: atp-resp 12266:5 (512) 0xae040000
jssmag.209.165 > helios.132: atp-rel  12266<0-7> 0xae030001
jssmag.209.133 > helios.132: atp-req* 12267<0-7> 0xae030002

Jssmag.209 initiates transaction id 12266 with host helios by requesting up to 8 packets (that's the <0-7>). The hex number at the end of the line is the value of the userdata field in the request.

Helios responds with 8 512-byte packets. The :digit following the transaction id gives the packet sequence number in the transaction, and the number in parens is the amount of data in the packet, excluding the ATP header. The * on packet 7 means the EOM bit was set. ✨

Jssmag.209 then requests that packets 3 & 5 be retransmitted. Helios resends them, then jssmag.209 releases the transaction. Finally, jssmag.209 initiates the next request. And the * on the request means XO (``exactly once'') was not set. 💅

💜 BACKWARD COMPATIBILITY 💜

The TCP flag names tcp-ece and tcp-cwr became available when linking with libpcap 1.9.0 or later.

🫶 SEE ALSO 🫶

stty(1), pcap(3PCAP), bpf(4), nit(4P), pcap-savefile(5), pcap-filter(7), pcap-tstamp(7)

https://www.iana.org/assignments/media-types/application/vnd.tcpdump.pcap

👑 AUTHORS 👑

The original authors are:

Van Jacobson, Craig Leres and Steven McCanne, all of the Lawrence Berkeley National Laboratory, University of California, Berkeley, CA.

It is currently maintained by The Tcpdump Group.

The current version is available via HTTPS:

https://www.tcpdump.org/

The original distribution is available via anonymous ftp:

ftp://ftp.ee.lbl.gov/old/tcpdump.tar.Z

IPv6/IPsec support is added by WIDE/KAME project. This program uses OpenSSL/LibreSSL, under specific configurations.

💀 BUGS 💀

okay bestie, gather round, because this is the part where we're honest about the messy bits. no shade, every icon has a few 🫶

🚨 security issue? do NOT post it in the group chat. send an e-mail to security@tcpdump.org and keep it cute and confidential.
🐛 wanna report bugs, contribute patches, request a feature, or just give generic feedback? go look at the file CONTRIBUTING.md in the tcpdump source tree root. that's the whole vibe.
NIT doesn't let you watch your own outbound traffic. BPF will. and like… we recommend you use the latter. periodt. 💅
some attempt should be made to reassemble IP fragments, or, at the very least, to compute the right length for the higher level protocol. right now? not it. 😭
name server inverse queries are not dumped correctly: the (empty) question section is printed rather than the real query in the answer section. and some people think inverse queries are themselves a bug and would rather fix the program generating them than fix tcpdump. iconic opinion tbh.
a packet trace that crosses a daylight savings time change will give you skewed time stamps — the time change is just ignored. clocks said nope. ⏰
filter expressions on fields other than those in Token Ring headers will not correctly handle source-routed Token Ring packets.
filter expressions on fields other than those in 802.11 headers will not correctly handle 802.11 data packets with both To DS and From DS set.
ip6 proto should chase the header chain, but at this moment it does not. ip6 protochain is supplied for this behavior. iykyk 🧚
arithmetic expression against transport layer headers, like tcp[0], does not work against IPv6 packets. it only looks at IPv4 packets. so yeah, that's a thing. 💀

no cap, none of these are secrets — they're literally in the manual. read them, respect them, work around them like the main character you are. 👑

and that's a wrap on TCPDUMP(8), bestie!! 🌈✨ this one is dated 30 June 2025 and it's the man page for tcpdump, so if you're reading a copy from somewhere else… check the date, babes. now go sniff some packets responsibly, mwah 💖🎀🦋

📦 where this page comes from

fedora 43: tcpdump 14:4.99.6-2.fc43 · BSD-2-Clause AND BSD-3-Clause AND BSD-4-Clause AND BSD-4-Clause-UC AND ISC AND NTP
fedora 44: tcpdump 14:4.99.6-3.fc44 · BSD-2-Clause AND BSD-3-Clause AND BSD-4-Clause AND BSD-4-Clause-UC AND ISC AND NTP
💖 girlified by deepseek-ai/DeepSeek-V4.1-Flash on 2026-10-18: a language model's rewrite of the original, under its terms