The Tcpdump Group git mirrors - tcpdump/blob - README.md

   1 # TCPDUMP 4.x.y by [The Tcpdump Group](https://www.tcpdump.org/)
   2
   3 **To report a security issue please send an e-mail to security@tcpdump.org.**
   4
   5 To report bugs and other problems, contribute patches, request a
   6 feature, provide generic feedback etc please see the
   7 [guidelines for contributing](CONTRIBUTING.md) in the tcpdump source tree root.
   8
   9 Anonymous Git is available via
  10
  11         https://github.com/the-tcpdump-group/tcpdump.git
  12
  13 This directory contains source code for tcpdump, a tool for network
  14 monitoring and data acquisition.
  15
  16 Over the past few years, tcpdump has been steadily improved by the
  17 excellent contributions from the Internet community (just browse
  18 through the [change log](CHANGES)).  We are grateful for all the input.
  19
  20 ### Supported platforms
  21 In many operating systems tcpdump is available as a native package or port,
  22 which simplifies installation of updates and long-term maintenance. However,
  23 the native packages are sometimes a few versions behind and to try a more
  24 recent snapshot it will take to compile tcpdump from the source code.
  25
  26 tcpdump compiles and works on at least the following platforms:
  27
  28 * [AIX](./doc/README.aix.md)
  29 * DragonFly BSD
  30 * FreeBSD
  31 * [Haiku](./doc/README.haiku.md)
  32 * HP-UX 11i
  33 * [illumos](./doc/README.solaris.md) (OmniOS, OpenIndiana)
  34 * GNU/Hurd
  35 * GNU/Linux
  36 * {Mac} OS X / macOS
  37 * [NetBSD](./doc/README.NetBSD.md)
  38 * OpenBSD
  39 * [Solaris](./doc/README.solaris.md)
  40 * [Windows](./doc/README.windows.md) (requires WinPcap or Npcap, and Visual
  41   Studio with CMake)
  42
  43 In the past tcpdump certainly or likely worked on the following platforms:
  44
  45 * 4.3BSD
  46 * BSD/386, later BSD/OS
  47 * DEC OSF/1, later Digital UNIX, later Tru64 UNIX
  48 * DOS
  49 * IRIX
  50 * LynxOS
  51 * SINIX
  52 * SunOS
  53 * Ultrix
  54 * UnixWare
  55
  56 ### Dependency on libpcap
  57 tcpdump uses libpcap, a system-independent interface for user-level
  58 packet capture.  Before building tcpdump, you must first retrieve and
  59 build libpcap.
  60
  61 Once libpcap is built (either install it or make sure it's in
  62 `../libpcap`), you can build tcpdump using the procedure in the
  63 [installation notes](INSTALL.md).
  64
  65 ### Origins of tcpdump
  66 The program is loosely based on SMI's "etherfind" although none of the
  67 etherfind code remains.  It was originally written by Van Jacobson as
  68 part of an ongoing research project to investigate and improve TCP and
  69 Internet gateway performance.  The parts of the program originally
  70 taken from Sun's etherfind were later re-written by Steven McCanne of
  71 LBL.  To insure that there would be no vestige of proprietary code in
  72 tcpdump, Steve wrote these pieces from the specification given by the
  73 manual entry, with no access to the source of tcpdump or etherfind.
  74 ```text
  75 formerly from   Lawrence Berkeley National Laboratory
  76                 Network Research Group <tcpdump@ee.lbl.gov>
  77                 ftp://ftp.ee.lbl.gov/old/tcpdump.tar.Z (3.4)
  78 ```
  79
  80 ### See also
  81 Richard Stevens gives an excellent treatment of the Internet protocols
  82 in his book *"TCP/IP Illustrated, Volume 1"*. If you want to learn more
  83 about tcpdump and how to interpret its output, pick up this book.
  84
  85 Another tool that tcpdump users might find useful is
  86 [tcpslice](https://github.com/the-tcpdump-group/tcpslice).
  87 It is a program that can be used to extract portions of tcpdump binary
  88 trace files.
  89
  90 ### The original LBL README by Steve McCanne, Craig Leres and Van Jacobson
  91 ```
  92 This directory also contains some short awk programs intended as
  93 examples of ways to reduce tcpdump data when you're tracking
  94 particular network problems:
  95
  96 send-ack.awk
  97         Simplifies the tcpdump trace for an ftp (or other unidirectional
  98         tcp transfer).  Since we assume that one host only sends and
  99         the other only acks, all address information is left off and
 100         we just note if the packet is a "send" or an "ack".
 101
 102         There is one output line per line of the original trace.
 103         Field 1 is the packet time in decimal seconds, relative
 104         to the start of the conversation.  Field 2 is delta-time
 105         from last packet.  Field 3 is packet type/direction.
 106         "Send" means data going from sender to receiver, "ack"
 107         means an ack going from the receiver to the sender.  A
 108         preceding "*" indicates that the data is a retransmission.
 109         A preceding "-" indicates a hole in the sequence space
 110         (i.e., missing packet(s)), a "#" means an odd-size (not max
 111         seg size) packet.  Field 4 has the packet flags
 112         (same format as raw trace).  Field 5 is the sequence
 113         number (start seq. num for sender, next expected seq number
 114         for acks).  The number in parens following an ack is
 115         the delta-time from the first send of the packet to the
 116         ack.  A number in parens following a send is the
 117         delta-time from the first send of the packet to the
 118         current send (on duplicate packets only).  Duplicate
 119         sends or acks have a number in square brackets showing
 120         the number of duplicates so far.
 121
 122         Here is a short sample from near the start of an ftp:
 123                 3.00    0.20   send . 512
 124                 3.20    0.20    ack . 1024  (0.20)
 125                 3.20    0.00   send P 1024
 126                 3.40    0.20    ack . 1536  (0.20)
 127                 3.80    0.40 * send . 0  (3.80) [2]
 128                 3.82    0.02 *  ack . 1536  (0.62) [2]
 129         Three seconds into the conversation, bytes 512 through 1023
 130         were sent.  200ms later they were acked.  Shortly thereafter
 131         bytes 1024-1535 were sent and again acked after 200ms.
 132         Then, for no apparent reason, 0-511 is retransmitted, 3.8
 133         seconds after its initial send (the round trip time for this
 134         ftp was 1sec, +-500ms).  Since the receiver is expecting
 135         1536, 1536 is re-acked when 0 arrives.
 136
 137 packetdat.awk
 138         Computes chunk summary data for an ftp (or similar
 139         unidirectional tcp transfer). [A "chunk" refers to
 140         a chunk of the sequence space -- essentially the packet
 141         sequence number divided by the max segment size.]
 142
 143         A summary line is printed showing the number of chunks,
 144         the number of packets it took to send that many chunks
 145         (if there are no lost or duplicated packets, the number
 146         of packets should equal the number of chunks) and the
 147         number of acks.
 148
 149         Following the summary line is one line of information
 150         per chunk.  The line contains eight fields:
 151            1 - the chunk number
 152            2 - the start sequence number for this chunk
 153            3 - time of first send
 154            4 - time of last send
 155            5 - time of first ack
 156            6 - time of last ack
 157            7 - number of times chunk was sent
 158            8 - number of times chunk was acked
 159         (all times are in decimal seconds, relative to the start
 160         of the conversation.)
 161
 162         As an example, here is the first part of the output for
 163         an ftp trace:
 164
 165         # 134 chunks.  536 packets sent.  508 acks.
 166         1       1       0.00    5.80    0.20    0.20    4       1
 167         2       513     0.28    6.20    0.40    0.40    4       1
 168         3       1025    1.16    6.32    1.20    1.20    4       1
 169         4       1561    1.86    15.00   2.00    2.00    6       1
 170         5       2049    2.16    15.44   2.20    2.20    5       1
 171         6       2585    2.64    16.44   2.80    2.80    5       1
 172         7       3073    3.00    16.66   3.20    3.20    4       1
 173         8       3609    3.20    17.24   3.40    5.82    4       11
 174         9       4097    6.02    6.58    6.20    6.80    2       5
 175
 176         This says that 134 chunks were transferred (about 70K
 177         since the average packet size was 512 bytes).  It took
 178         536 packets to transfer the data (i.e., on the average
 179         each chunk was transmitted four times).  Looking at,
 180         say, chunk 4, we see it represents the 512 bytes of
 181         sequence space from 1561 to 2048.  It was first sent
 182         1.86 seconds into the conversation.  It was last
 183         sent 15 seconds into the conversation and was sent
 184         a total of 6 times (i.e., it was retransmitted every
 185         2 seconds on the average).  It was acked once, 140ms
 186         after it first arrived.
 187
 188 stime.awk
 189 atime.awk
 190         Output one line per send or ack, respectively, in the form
 191                 <time> <seq. number>
 192         where <time> is the time in seconds since the start of the
 193         transfer and <seq. number> is the sequence number being sent
 194         or acked.  I typically plot this data looking for suspicious
 195         patterns.
 196
 197
 198 The problem I was looking at was the bulk-data-transfer
 199 throughput of medium delay network paths (1-6 sec.  round trip
 200 time) under typical DARPA Internet conditions.  The trace of the
 201 ftp transfer of a large file was used as the raw data source.
 202 The method was:
 203
 204   - On a local host (but not the Sun running tcpdump), connect to
 205     the remote ftp.
 206
 207   - On the monitor Sun, start the trace going.  E.g.,
 208       tcpdump host local-host and remote-host and port ftp-data >tracefile
 209
 210   - On local, do either a get or put of a large file (~500KB),
 211     preferably to the null device (to minimize effects like
 212     closing the receive window while waiting for a disk write).
 213
 214   - When transfer is finished, stop tcpdump.  Use awk to make up
 215     two files of summary data (maxsize is the maximum packet size,
 216     tracedata is the file of tcpdump tracedata):
 217       awk -f send-ack.awk packetsize=avgsize tracedata >sa
 218       awk -f packetdat.awk packetsize=avgsize tracedata >pd
 219
 220   - While the summary data files are printing, take a look at
 221     how the transfer behaved:
 222       awk -f stime.awk tracedata | xgraph
 223     (90% of what you learn seems to happen in this step).
 224
 225   - Do all of the above steps several times, both directions,
 226     at different times of day, with different protocol
 227     implementations on the other end.
 228
 229   - Using one of the Unix data analysis packages (in my case,
 230     S and Gary Perlman's Unix|Stat), spend a few months staring
 231     at the data.
 232
 233   - Change something in the local protocol implementation and
 234     redo the steps above.
 235
 236   - Once a week, tell your funding agent that you're discovering
 237     wonderful things and you'll write up that research report
 238     "real soon now".
 239 ```