PC World 2007 December/PCWorld_2007-12_cd.bin/v cisle/nmap/nmap-4.20-setup.exe/CHANGELOG

Text File | 2006-12-07 | 205KB | 4,568 lines

# Nmap Changelog ($Id: CHANGELOG 4229 2006-12-08 03:02:09Z fyodor $); -*-text-*-
4.20

o Integrated the latest OS fingerprint submissions.  The 2nd
  generation DB size has grown to 231 fingerprints.  Please keep them
  coming!  New fingerprints include Mac OS X Server 10.5 pre-release,
  NetBSD 4.99.4, Windows NT, and much more.

o Fixed a segmentation fault in the new OS detection system
  which was reported by Craig Humphrey and Sebastian Garcia.

o Fixed a TCP sequence prediction difficulty indicator bug. The index
  is supposed to go from 0 ("trivial joke") to about 260 (OpenBSD).
  But some systems generated ISNs so insecurely, that Nmap went
  berserk and reported a negative difficulty index.  This generally
  only affects some printers, crappy consumer devices, and Microsoft
  Windows (old versions).  Thanks to Sebastian Garcia for helping me
  track down the problem.

4.20RC2

o Integrated all of your OS detection submissions since RC1.  The DB
  has increased 13% to 214 fingerprints.  Please keep them coming!
  New fingerprints include versions of z/OS, OpenBSD, Linux, AIX,
  FreeBSD, Cisco CatOS, IPSO firewall, and a slew of printers and
  misc. devices.  We also got our first Windows 95 fingerprint,
  submitted anonymously of course :).

o Fixed (I hope) the "getinterfaces: intf_loop() failed" error which
  was seen on Windows Vista.  The problem was apparently in
  intf-win32.c of libcnet (need to define MIB_IF_TYPE_MAX to
  MAX_IF_TYPE rather than 32).  Thanks to Dan Griffin
  (dan(a)jwsecure.com) for tracking this down!

o Applied a couple minor bug fixes for IP options
  support and packet tracing.  Thanks to Michal Luczaj
  (regenrecht(a)o2.pl) for reporting them.

o Incorporated SLNP (Simple Library Network Protocol) version
  detection support.  Thanks to Tibor Csogor (tibi(a)tiborius.net) for
  the patch.

4.20RC1

o Fixed (I hope) a bug related to Pcap capture on Mac OS X.  Thanks to
  Christophe Thil for reporting the problem and to Kurt Grutzmacher
  and Diman Todorov for helping to track it down.

o Integrated all of your OS detection submissions since ALPHA11.  The
  DB has increased 27% to 189 signatures.  Notable additions include
  the Apple Airport Express, Windows Vista RC1, OpenBSD 4.0, a Sony
  TiVo device, and tons of broadband routers, printers, switches, and
  Linux kernels.  Keep those submissions coming!

o Upgraded the included LibPCRE from version 6.4 to 6.7.  Thanks to
  Jochen Voss (voss(a)seehuhn.de) for the suggestion (he found some bugs
  in 6.4)

4.20ALPHA11

o Integrated all of your OS detection submissions, bringing the
  database up to 149 fingerprints.  This is an increase of 28% from
  ALPHA10.  Notable additions include FreeBSD 6.1, a bunch of HP
  LaserJet printers, and HP-UX 11.11.  We also got a bunch of more
  obscure submissions like Minix 3.1.2a and "Ember InSight Adapter for
  programming EM2XX-family embedded devices".  Who doesn't have a few
  of those laying around?  I'm hoping that all the obscure submissions
  mean that more of the mainstream systems are being detected out of
  the box!  Please keep those submissions (obscure or otherwise)
  coming!

4.20ALPHA10

o Integrated tons of new OS fingerprints.  The DB now contains 116
  fingerprints, which is up 63% since the previous version.  Please keep
  the submissions coming!

4.20ALPHA9

o Integrated the newly submitted OS fingerprints. The DB now contains
  71 fingerprints, up 27% from 56 in ALPHA8.  Please keep them coming!
  We still only have 4.2% as many fingerprints as the gen1 database.

o Added the --open option, which causes Nmap to show only open ports.
  Ports in the states "open|closed" and "unfiltered" might be open, so
  those are shown unless the host has an overwhelming number of them.

o Nmap gen2 OS detection used to always do 2 retries if it fails to
  find a match.  Now it normally does just 1 retry, but does 4 retries
  if conditions are good enough to warrant fingerprint submission.
  This should speed things up on average.  A new --max-os-tries option
  lets you specify a higher lower maximum number of tries.

o Added --unprivileged option, which is the opposite of --privileged.
  It tells Nmap to treat the user as lacking network raw socket and
  sniffing privileges.  This is useful for testing, debugging, or when
  the raw network functionality of your operating system is somehow
  broken.

o Fixed a confusing error message which occured when you specified a
  ping scan or list scan, but also specified -p (which is only used for
  port scans).  Thanks to Thomas Buchanan for the patch.

o Applied some small cleanup patches from Kris Katterjohn

4.20ALPHA8

o Integrated the newly submitted OS fingerprints.  The DB now contains
  56, up 33% from 42 in ALPHA7.  Please keep them coming!  We still only
  have 3.33% as many signatures as the gen1 database.

o Nmap 2nd generation OS detection now has a more sophisticated
  mechanism for guessing a target OS when there is no exact match in the
  database (see http://insecure.org/nmap/osdetect/osdetect-guess.html )

o Rewrote mswin32/nmap.rc to remove cruft and hopefully reduce some
  MFC-related compilation problems we've seen.  Thanks to KX
  (kxmail(a)gmail.com) for doing this.

o NmapFE now uses a spin button for verbosity and debugging options so
  that you can specify whatever verbosity (-v) or debugging (-d) level
  you desire.  The --randomize-hosts option was also added to NmapFE.
  Thanks to Kris Katterjohn for the patches.

o A dozen or so small patches to Nmap and NmapFE by Kris Katterjohn.

o Removed libpcap/Win32 and libpcap/msdos as Nmap doesn't use them.
  This reduces the Nmap tar.bz2 by about 50K.  Thanks to Kris Katterjohn
  for the suggestion.

4.20ALPHA7

o Did a bunch of Nmap 2nd generation fingerprint integration work.
  Thanks to everyone who sent some in, though we still need a lot more.
  Also thanks to Zhao for a bunch of help with the integration tools.
  4.20ALPHA6 had 12 fingerprints, this new version has 42.  The old DB
  (still included) has 1,684.

o Updated nmap-mac-prefixes to reflect the latest OUI DB from the IEEE
  (http://standards.ieee.org/regauth/oui/oui.txt) as of September 6, 2006.
  Also added the unregistered PearPC virtual NIC prefix, as suggested
  by Robert Millan (rmh(a)aybabtu.com).

o Applied some small internal cleanup patches by Kris Katterjohn.

4.20ALPHA6

o Fixed a bug in 2nd generation OS detection which would (usually) prevent
  fingerprints from being printed when systems don't respond to the 1st
  ICMP echo probe (the one with bogus code value of 9).  Thanks to
  Brandon Enright for reporting and helping me debug the problem.

o Fixed some problematic Nmap version detection signatures which could
  cause warning messages. Thanks to Brandon Enright for the initial patch.

4.20ALPHA5

o Worked with Zhao to improve the new OS detection system with
  better algorithms, probe changes, and bug fixes.  We're
  now ready to start growing the new database!  If Nmap gives you
  fingerprints, please submit them at the given URL.  The DB is still
  extremely small.  The new system is extensively documented at
  http://insecure.org/nmap/osdetect/ .

o Nmap now supports IP options with the new --ip-options flag.  You
  can specify any options in hex, or use "R" (record route), "T"
  (record timestamp), "U") (record route & timestamp), "S [route]"
  (strict source route), or "L [route]" (loose source route).  Specify
  --packet-trace to display IP options of responses.  For further
  information and examples, see http://insecure.org/nmap/man/ and
  http://seclists.org/nmap-dev/2006/q3/0052.html .  Thanks to Marek
  Majkowski for writing and sending the patch.

o Integrated all 2nd quarter service detection fingerprint
  submissions.  Please keep them coming!  We now have 3,671 signatures
  representing 415 protocols.   Thanks to version detection czar Doug
  Hoyte for doing this.

o Nmap now uses the (relatively) new libpcap pcap_get_selectable_fd
  API on systems which support it.  This means that we no longer need
  to hack the included Pcap to better support Linux.  So Nmap will now
  link with an existing system libpcap by default on that platform if
  one is detected.  Thanks to Doug Hoyte for the patch.

o Updated the included libpcap from 0.9.3 to 0.9.4.  The changes I
  made are in libpcap/NMAP_MODIFICATIONS .  By default, Nmap will now
  use the included libpcap unless version 0.9.4 or greater is already
  installed on the system.

o Applied some nsock bugfixes from Diman Todorov.  These don't affect
  the current version of Nmap, but are important for his Nmap
  Scripting Engine, which I hope to integrate into mainline Nmap in
  September.

o Fixed a bug which would occasionally cause Nmap to crash with the
  message "log_vwrite: write buffer not large enough".  I thought I
  conquered it in a previous release -- thanks to Doug Hoyte for finding a
  corner case which proved me wrong.

o Fixed a bug in the rDNS system which prevented us from querying
  certain authoritative DNS servers which have recursion explicitly
  disabled.  Thanks to Doug Hoyte for the patch.

o --packet-trace now reports TCP options (thanks to Zhao Lei for the
  patch).  Thanks to the --ip-options addition also found in this
  release, IP options are printed too.

o Cleaned up Nmap DNS reporting to be a little more useful and
  concise.  Thanks to Doug Hoyte for the patch.

o Applied a bunch of small internal cleanup patches by Kris Katterjohn
  (kjak(a)ispwest.com).

o Fixed the 'distclean' make target to be more comprehensive.  Thanks
  to Thomas Buchanan (Thomas.Buchanan(a)thecompassgrp.net) for the
  patch.

Nmap 4.20ALPHA4

o Nmap now provides progress statistics in the XML output in verbose
  mode.  Here are some examples of the format (etc is "estimated time
  until completion) and times are in UNIX time_t (seconds since 1970)
  format. Angle braces have been replaced by square braces:
  [taskbegin task="SYN Stealth Scan" time="1151384685" /]
  [taskprogress task="SYN Stealth Scan" time="1151384715"
                percent="13.85" remaining="187" etc="1151384902" /]
  [taskend task="SYN Stealth Scan" time="1151384776" /]
  [taskbegin task="Service scan" time="1151384776" /]
  [taskend task="Service scan" time="1151384788" /]
  Thanks to Adam Vartanian (flooey(a)gmail.com) for the patch.

o Updated the Windows installer to give an option checkbox for
  performing the Nmap performance registry changes.  The default is to
  do so.  Thanks to Adam Vartanian (flooey(a)gmail.com) for the patch.

o Applied several code cleanup patches from Marek Majkowski.

o Added --release-memory option, which causes Nmap to release all
  accessible memory buffers before quitting (rather than let the OS do
  it).  This is only useful for debugging memory leaks.

o Fixed a bug related to bogus completion time estimates when you
  request an estimate (through runtime interaction) right when Nmap is
  starting.a subsystem (such as a port scan or version detection).
  Thanks to Diman Todorov for reporting the problem and Doug Hoyte for
  writing a fix.

o Nmap no longer gets random numbers from OpenSSL when it is available
  because that turned out to be slower than Nmap's other methods
  (e.g. /dev/urandom on Linux, /dev/arandom on OpenBSD, etc.).  Thanks
  to Marek Majkowski for reporting the problem.

o Updated the Windows binary distributions (self-installer and .zip)
  to include the new 2nd generation OS detection DB (nmap-os-db).
  Thanks to Sina Bahram for reporting the problem.

o Fixed the --max-retries option, which wasn't being honored.  Thanks
  to Jon Passki (jon.passki(a)hursk.com) for the patch.

Nmap 4.20ALPHA3

o Added back Win32 support thanks to a patch by kx

o Fixed the English translation of TCP sequence difficulty reported by
  Brandon Enright, and also removed fingerprint printing for 1st
  generation fingerprints (I don't really want to deal with those
  anymore).  Thanks to Zhao Lei for writing this patch.

o Fix a problem which caused OS detection to be done in some cases
  even if the user didn't request it.  Thanks to Diman Todorov for the
  fix.

Nmap 4.20ALPHA2

o Included nmap-os-db (the new OS detection DB) within the release.
  Oops!  Thanks to Brandon Enright (bmenrigh(a)ucsd.edu) for catching
  this problem with 4.20ALPHA1.

o Added a fix for the crash in the new OS detection which would come
  with the message "Probe doesn't exist! Probe type: 1. Probe subid: 1"

Nmap 4.20ALPHA1

o Integrated initial 2nd generation OS detection patch!  The system is
  documented at http://insecure.org/nmap/osdetect/ .  Thanks to Zhao Lei
  for helping with the coding and design.

o portlist.cc was refactored to remove some code duplication.  Thanks
  to Diman Todorov for the patch.

Nmap 4.11

o Added a dozens of more detailed SSH version detection signatures, thanks
  to a SSH huge survey and integration effort by Doug Hoyte.  The
  results of his large-scale SSH scan are posted at
  http://seclists.org/nmap-dev/2006/Apr-Jun/0393.html .

o Fixed the Nmap Makefile (actually Makefile.in) to correctly handle
  include file dependencies.  So if a .h file is changed, all of the
  .cc files which depend on it will be recompiled.  Thanks to Diman
  Todorov (diman(a)xover.mud.at) for the patch.

o Fixed a compilation problem on solaris and possibly other platforms.
  The error message looked like "No rule to make target `inet_aton.o',
  needed by `libnbase.a'".  Thanks to Matt Selsky
  (selsky(a)columbia.edu) for the patch.

o Applied a patch which helps with HP-UX compilation by linking in the
  nm library (-lnm).  Thanks to Zakharov Mikhail
  (zmey20000(a)yahoo.com) for the patch.

o Added version detection probes for detecting the Nessus daemon.
  Thanks to Adam Vartanian (flooey(a)gmail.com) for sending the patch.

Nmap 4.10

o Updated nmap-mac-prefixes to reflect the latest OUI DB from the IEEE
  (http://standards.ieee.org/regauth/oui/oui.txt) as of May 31, 2006.
  Also added a couple unregistered OUI's (for QEMU and Bochs)
  suggested by Robert Millan (rmh(a)aybabtu.com).

o Fixed a bug which could cause false "open" ports when doing a UDP
  scan of localhost. This usually only happened when you scan tens of
  thousands of ports (e.g. -p- option).

o Fixed a bug in service detection which could lead to a crash when
  "--version-intensity 0" was used with a UDP scan.  Thanks to Makoto
  Shiotsuki (shio(a)st.rim.or.jp) for reporting the problem and Doug
  Hoyte for producing a patch.

o Made some AIX and HP-UX portability fixes to Libdnet and NmapFE.
  These were sent in by Peter O'Gorman
  (nmap-dev(a)mlists.thewrittenword.com).

o When you do a UDP+TCP scan, the TCP ports are now shown first (in
  numerical order), followed by the UDP ports (also in order).  This
  contrasts with the old format which showed all ports together in
  numerical order, regardless of protocol.  This was at first a "bug",
  but then I started thinking this behavior may be better.  If you
  have a preference for one format or the other, please post your
  reasons to nmap-dev.

o Changed mass_dns system to print a warning if it can't find any
  available DNS servers, but not quit like it used to.  Thanks to Doug
  Hoyte for the patch.

Nmap 4.04BETA1

o Integrated all of your submissions (about a thousand) from the first
  quarter of this year!  Please keep 'em coming!  The DB has increased
  from 3,153 signatures representing 381 protocols in 4.03 to 3,441
  signatures representing 401 protocols.  No other tool comes close!
  Many of the already existing match lines were improved too.  Thanks
  to Version Detection Czar Doug Hoyte for doing this.

o Nmap now allows multiple ignored port states.  If a 65K-port scan
  had, 64K filtered ports, 1K closed ports, and a few dozen open
  ports, Nmap used to list the dozen open ones among a thousand lines
  of closed ports.  Now Nmap will give reports like "Not shown: 64330
  filtered ports, 1000 closed ports" or "All 2051 scanned ports on
  192.168.0.69 are closed (1051) or filtered (1000)", and omit all of
  those ports from the table.  Open ports are never ignored.  XML
  output can now have multiple <extraports> directive (one for each
  ignored state).  The number of ports in a single state before it is
  consolidated defaults to 26 or more, though that number increases as
  you add -v or -d options.  With -d3 or higher, no ports will be
  consolidated.  The XML output should probably be augmented to give
  the extraports directive 'ip', 'tcp', and 'udp' attributes which
  specify the corresponding port numbers in the given state in the
  same listing format as the nmaprun.scaninfo.services attribute, but
  that part hasn't yet been implemented.  If you absoultely need the
  exact port numbers for each state in the XML, use -d3 for now.

o Nmap now ignores certain ICMP error message rate limiting (rather
  than slowing down to accomidate it) in cases such as SYN scan where
  an ICMP message and no response mean the same thing (port filtered).
  This is currently only done at timing level Aggressive (-T4) or
  higher, though we may make it the default if we don't hear problems
  with it.  In addition, the --defeat-rst-ratelimit option has been
  added, which causes Nmap not to slow down to accomidate RST rate
  limits when encountered.  For a SYN scan, this may cause closed
  ports to be labeled 'filtered' becuase Nmap refused to slow down
  enough to correspond to the rate limiting.  Learn more about this
  new option at http://www.insecure.org/nmap/man/ .  Thanks to Martin
  Macok (martin.macok(a)underground.cz) for writing the patch that
  these changes were based on.

o Moved my Nmap development environment to Visual C++ 2005 Express
  edition.  In typical "MS Upgrade Treadmill" fashion, Visual Studio
  2003 users will no longer be able to compile Nmap using the new
  solution files.  The compilation, installation, and execution
  instructions at
  http://www.insecure.org/nmap/install/inst-windows.html have been
  upgraded.

o Automated my Windows build system so that I just have to type a
  single make command in the mswin32 directory.  Thanks to Scott
  Worley (smw(a)pobox.com>, Shane & Jenny Walters
  (yfisaqt(a)waltersinamerica.com), and Alex Prinsier
  (aphexer(a)mailhaven.com) for reading my appeal in the 4.03
  CHANGELOG and assisting.

o Changed the PortList class to use much more efficient data
  structures and algorithms which take advantage of Nmap-specific
  behavior patterns.  Thanks to Marek Majkowski
  (majek(a)forest.one.pl) for the patch.

o Fixed a bug which prevented certain TCP+UDP scan commands, such as
  "nmap -sSU -p1-65535 localhost" from scanning both TCP and UDP.
  Instead they gave the error message "WARNING: UDP scan was requested,
  but no udp ports were specified.  Skipping this scan type".  Thanks to
  Doug Hoyte for the patch.

o Nmap has traditionally required you to specify -T* timing options
  before any more granular options like --max-rtt-timeout, otherwise the
  general timing option would overwrite the value from your more
  specific request.  This has now been fixed so that the more specific
  options always have precendence.  Thanks to Doug Hoyte for this patch.

o Fixed a couple possible memory leaks reported by Ted Kremenek
 (kremenek(a)cs.stanford.edu) from the Stanford University sofware
 static analysis lab ("Checker" project).

o Nmap now prints a warning when you specify a target name which
  resolves to multiple IP addresses.  Nmap proceeds to scan only the
  first of those addresses (as it always has done).  Thanks to Doug
  Hoyte for the patch.  The warning looks like this:
  Warning: Hostname google.com resolves to 3 IPs. Using 66.102.7.99.

o Disallow --host-timeout values of less than 1500ms, print a warning
  for values less than 15s.

o Changed all instances of inet_aton() into calls to inet_pton()
  instead.  This allowed us to remove inet_aton.c from nbase.  Thanks to
  KX (kxmail(a)gmail.com) for the patch.

o When debugging (-d) is specified, Nmap now prints a report on the
  timing variables in use.  Thanks to Doug Hoyte for the patch.  The
  report loos like this:
  ---------- Timing report ----------
    hostgroups: min 1, max 100000
    rtt-timeouts: init 250, min 50, max 300
    scan-delay: TCP 5, UDP 1000
    parallelism: min 0, max 0
    max-retries: 2, host-timeout 900000
  -----------------------------------

o Modified the WinPcap installer file to explicitly uninstall an
  existing WinPcap (if you select that you wish to replace it) rather
  than just overwriting the old version.  Thanks to Doug Hoyte for
  making this change.

o Added some P2P application ports to the nmap-services file.  Thanks
  to Martin Macok for the patch.

o The write buffer length increased in 4.03 was increased even further
  when the debugging or verbosity levels are more than 2 (e.g. -d3).
  Thanks to Brandon Enright (bmenrigh(a)ucsd.edu) for the patch.  The
  goal is to prevent you from ever seeing the fatal error:
  "log_vwrite: write buffer not large enough -- need to increase"

o Added a note to the Nmap configure dragon that people sick of him
  can submit their own ASCII art to [e-mail] .  If you
  are wondering WTF I am talking about, it is probably because only
  most elite Nmap users -- the ones who compile from source on UNIX --
  get to see the 'l33t ASCII Art.

Nmap 4.03

o Updated the LibPCRE build system to add the -fno-thread-jumps option
  to gcc when compiling on the new Intel-based Apple Mac OS X systems.
  Hopefully this resolves the version detection crashes that several
  people have reported on such systems.  Thanks to Kurt Grutzmacher
  (grutz(a)jingojango.net) for sending the configure.ac patch.

o Made some portability fixes to keep Nmap compiling with the newest
  Visual Studio 2005.  Thanks to KX (kxmail(a)gmail.com) for
  suggesting them.

o Service fingerprints are now provided in the XML output whenever
  they would appear in the interactive output (i.e. when a service
  response with data but is unrecognized).  They are shown in a new
  'servicefp' attribute to the 'service' tag.  Thanks to Brandon Enright
  (bmenrigh(a)ucsd.edu) for sending the patch.

o Improved the Windows build system -- mswin32/Makefile now takes care
  of packaging Nmap and creating the installers once Visual Studio (GUI)
  is done building the Release version of mswin32/nmap.sln.  If someone
  knows how to do this (build) step on the command line (using the
  Makefile), please let me know.  Or if you know how to at least make
  'Release' (rather than Debug) the default configuration, that would be
  valuable.

o WinPcap 3.1 binaries are now shipped in the Nmap tarball, along with
  a customized installer written by Doug Hoyte.  That new WinPcap
  installer is now used by the Nmap self-installer (if you request
  WinPcap installation).  Some Nmap users were uncomfortable with a
  "phone home" feature of the official WinPcap installer.  It connects
  back to CACE Technologies, ostensibly to display news and (more
  recently) advertisements.  Our new installer omits that feature, but
  should be otherwise perfectly compatible with WinPcap 3.1.

o Fixed (I hope) a problem where aggressive --min-parallelization
  option values could cause Nmap to quit with the message "box(300, 100,
  15) called (min,max,num)".  Thanks to  Richard van den Berg
  (richard.vandenberg(a)ins.com) for reporting the problem.

o Fixed a rare crash bug thanks to a report and patch from Ganga
  Bhavani (GBhavani(a)everdreamcorp.com)

o Increased a write buffer length to keep Nmap from quitting with the
  message "log_vwrite: write buffer not large enough -- need to
  increase".  Thanks to Dave (dmarcher(a)pobox.com) for reporting the
  issue.

Nmap 4.02ALPHA2

o Updated to a newer XSL stylesheet (for XML to HTML output
  transformation) by Benjamin Erb.  This new version includes IP
  address sorting, removal of javascript requirements, some new
  address, hostname, and Nmap version information, and various minor
  tweaks and fixes.

o Cleaned up the Amiga port code to use atexit() rather than the
  previous macro hack.  Thanks to Kris Katterjohn (kjak(a)ispwest.com)
  for the patch.  Applied maybe half a dozen new other code cleanup
  patches from him as well.

o Made some changes to various Nmap initialization functions which
  help ALT Linux (altlinux.org) and Owl (openwall.com) developers run
  Nmap in a chroot environment.  Thanks to Dmitry V. Levin
  (ldv(a)altlinux.org) for the patch.

o Cleaned up the code a bit by making a bunch (nearly 100) global
  symbols (mostly function calls) static.  I was also able to removed
  some unused functions and superfluous config.h.in defines.  Thanks
  to Dmitry V. Levin (ldv(a)altlinux.org) for sending a list of
  candidate symbols.

o Nmap now tests for the existence of data files using stat(2) rather
  than testing whether they can be opened for reading (with fopen).
  This is because some device files (tape drives, etc.) may react badly
  to being opened at all.  Thanks to Dmitry V. Levin
  (ldv(a)altlinux.org) for the suggestion.

o Changed Nmap to cache interface information rather than opening and
  closing it (with dnet's eth_open and eth_close functions) all the
  time.

o Applied a one-character Visual Studio 2005 compatibility patch from
  kx (kxmail(a)gmail.com).  It changed getch() into _getch() on Windows.

Nmap 4.02ALPHA1

o Added the --log-errors option, which causes most warnings and error
  messages that are printed to interactive-mode output (stdout/stderr)
  to also be printed to the normal-format output file (if you
  specified one).  This will not work for most errors related to bad
  command-line arguments, as Nmap may not have initialized its output
  files yet.  In addition, some Nmap error/warning messages use a
  different system that does not yet support this option.

o Rewrote much of the Nmap results output functions to be more
  efficient and support --log-errors.

o Fixed a flaw in the scan engine which could (in rare cases)
  lead to a deadlock situation that prevents a scan from completing.
  Thanks to Ganga Bhavani (GBhavani(a)everdreamcorp.com) for reporting
  and helping to debug the problem.

o If the pcap_open_live() call (initiates sniffing) fails, Nmap now
  tries up to two more times after waiting a little while. This is
  attempt to work around a rare bug on Windows in which the
  pcap_open_live() fails for unknown reasons.

o Fixed a flaw in the runtime interaction in which Nmap would include
  hosts currently being scanned in the number of hosts "completed"
  statistic.

o Fixed a crash in OS scan which could occur on Windows when a DHCP
  lease issue causes the system to lose its IP address.  Nmap still
  quits, but at least it gives a proper error message now.  Thanks to
  Ganga Bhavani (GBhavani(a)everdreamcorp.com) for the patch.

o Applied more than half a dozen small code cleanup patches from
  Kris Katterjohn (kjak(a)ispwest.com).

o Modified the configure script to accept CXX when specified as an
  absolute path rather than just the executable name.  Thanks to
  Daniel Roethlisberger (daniel(a)roe.ch) for this patch.

Nmap 4.01

o Fixed a bug that would cause bogus reverse-DNS resolution on
  big-endian machines.  Thanks to Doug Hoyte, Seth Miller, Tony Doan,
  and Andrew Lutomirsky for helping to debug and patch the problem.

o Fixed an important memory leak in the raw ethernet sending system.
  Thanks to Ganga Bhavani (GBhavani(a)everdreamcorp.com) for
  identifying the bug and sending a patch.

o Fixed --system-dns option so that --system_dns works too.  Error
  messages were changed to reflect the former (preferred) name.
  Thanks to Sean Swift (sean.swift(a)bradford.gov.uk) and Peter
  VanEeckhoutte (Peter.VanEeckhoutte(a)saraleefoodseurope.com) for
  reporting the problem.

o Fixed a crash which would report this message:
  "NmapOutputTable.cc:143: void NmapOutputTable::addItem(unsigned int,
  unsigned int, bool, const char*, int): Assertion `row < numRows'
  failed."  Thanks to Jake Schneider (Jake.Schneider(a)dynetics.com) for
  reporting and helping to debug the problem.

o Whenever Nmap sends packets with the SYN bit set (except for OS
  detection), it now includes the maximum segment size (MSS) tcp
  option with a value of 1460.  This makes it stand out less as almost
  all hosts set at least this option.  Thanks to Juergen Schmidt
  (ju(a)heisec.de) for the suggestion.

o Applied a patch for a Windows interface reading bug in the aDNS
  subsystem from Doug Hoyte.

o Minor changes to recognize DragonFly BSD in configure
  scripts. Thanks to Joerg Sonnenberger (joerg(a)britannica.bec.de)
  for sending the patch.

o Fixed a minor bug in an error message starting with "eth_send of ARP
  packet returned".  Thanks to J.W. Hoogervorst
  (J.W.Hoogervorst(a)uva.nl) for finding this.

Nmap 4.00

o Added the '?' command to the runtime interaction system.  It prints a
  list of accepted commands.  Thanks to Andrew Lutomirski
  (luto(a)myrealbox.com) for the patch.

o See the announcement at
  http://www.insecure.org/stf/Nmap-4.00-Release.html for high-level
  changes since 3.50.

Nmap 3.9999

o Generated a new libpcre/configure to cope with changes in LibPCRE
  6.4

o Updated nmap-mac-prefixes to reflect the latest OUI DB from the IEEE
  (http://standards.ieee.org/regauth/oui/oui.txt)

o Updated nmap-protocols with the latest IEEE internet protocols
  assignments (http://www.iana.org/assignments/protocol-numbers).

o Updated the Nmap version number and related fields that MS Visual
  Studio places in the binary.  This was done by editing
  mswin32/nmap.rc.

Nmap 3.999

o Added runtime interaction support to Windows, thanks to patches from
  Andrew Lutomirski (luto(a)myrealbox.com) and Gisle Vanem (giva(a)bgnett.no).

o Changed a couple lines of tcpip.cc (put certain IP header fields in
  host byte order rather than NBO) to (hopefully) support Mac OS X on
  Intel.  Thanks to Kurt Grutzmacher (grutz(a)jingojango.net) for the
  patch.

o Upgraded the included LibPCRE from version 6.3 to 6.4.  There was a
  report of version detection crashes on the new Intel-based MACs with
  6.3.

o Fixed an issue in which the installer would malfunction in rare
  issues when installing to a directory with spaces in it.  Thanks to
  Thierry Zoller (Thierry(a)Zoller.lu) for the report.

Nmap 3.99

o Integrated all remaining 2005 service submissions.  The DB now has
  surpassed 3,000 signatures for the first time.  There now are 3,153
  signatures for 381 service protocols.  Those protocols span the
  gamut from abc, acap, afp, and afs to zebedee, zebra, and
  zenimaging.  It even covers obscure protocols such as http, ftp,
  smtp, and ssh :).  Thanks to Version Detection Czar Doug Hoyte for
  his excellent work on this.

o Created a Windows executable installer using the open source NSIS
  (Nullsoft Scriptable Install System).  It handles Pcap installation,
  registry performance changes, and adding Nmap to your cmd.exe
  executable path.  The installer source files are in mswin32/nsis/ .
  Thanks to Google SoC student Bo Jiang (jiangbo(a)brandeis.edu) for
  creating the initial version.

o Fixed a backward compatibility bug in which Nmap didn't recognize
  the --min_rtt_timeout option (it only recognized the newly
  hyphenated --min-rtt-timeout).  Thanks to Joshua D. Abraham
  (jabra(a)ccs.neu.edu) for the bug report.

o Fixed compilation to again work with gcc-derivatives such as
  MingW. Thanks to Gisle Vanem (giva(a)bgnett.no) for sending the
  patches

Nmap 3.98BETA1

o Added run time interaction as documented at
  http://www.insecure.org/nmap/man/man-runtime-interaction.html .
  While Nmap is running, you can now press 'v' to increase verbosity,
  'd' to increase the debugging level, 'p' to enable packet tracing,
  or the capital versions (V,D,P) to do the opposite.  Any other key
  (such as enter) will print out a status message giving the estimated
  time until scan completion.  This only works on UNIX for now.  Do we
  have any volunteers to add Windows support?  You would need to
  change a handful of UNIX-specific termio calls with the Windows
  equivalents.  This feature was created by Paul Tarjan
  (ptarjan(a)stanford.edu) as part of the Google Summer of Code.

o Reverse DNS resolution is now done in parallel rather than one at a
  time.  All scans of large networks (particularly list, ping and
  just-a-few-ports scans) should benefit substantially from this
  change.  If you encounter any problems, please let us know.  The new
  --system_dns option was added so you can use the (slow) system
  resolver if you prefer that for some reason.  You can specify a
  comma separated list of DNS server IP addresses for Nmap to use with
  the new --dns_servers option.  Otherwise, Nmap looks in
  /etc/resolve.conf (UNIX) or the system registry (Windows) to obtain
  the nameservers already configured for your system.  This excellent
  patch was written by Doug Hoyte (doug(a)hcsw.org).

o Added the --badsum option, which causes Nmap to use invalid TCP or
  UDP checksums for packets sent to target hosts. Since virtually all
  host IP stacks properly drop these packets, any responses received
  are likely coming from a firewall or IDS that didn't bother to
  verify the checksum. For more details on this technique, see
  http://www.phrack.org/phrack/60/p60-0x0c.txt .  The author of that
  paper, Ed3f (ed3f(a)antifork.org), is also the author of this patch
  (which I changed it a bit).

o The 26 Nmap commands that previously included an underscore
  (--max_rtt_timeout, --send_eth, --host_timeout, etc.) have been
  renamed to use a hyphen in the preferred format
  (i.e. --max-rtt-timeout).  Underscores are still supported for
  backward compatibility.

o More excellent NmapFE patches from Priit Laes (amd(a)store20.com)
  were applied to remove all deprecated GTK API calls.  This also
  eliminates the annoying Gtk-Critical and Gtk-WARNING runtime messages.

o Changed the way the __attribute__ compiler extension is detected so
  that it works with the latest Fedora Core 4 updates (and perhaps other
  systems).  Thanks to Duilio Protti (dprotti(a)fceia.unr.edu.ar) for
  writing the patch.  The compilation error message this fixes was
  usually something like: "nmap.o(.rodata+0x17c): undefined reference
  to `__gthrw_pthread_cancel(unsigned long)"

o Added some exception handling code to mswin32/winfix.cc to prevent
  Nmap from crashing mysteriously when you have WinPcap 3.0 or earlier
  (instead of the required 3.1).  It now prints an error message instead
  asking you to upgrade, then reduces functionality to connect()-only
  mode.  I couldn't get it working with the C++ standard try/catch()
  blocks, but as soon as I used the nonstandard MS conventions
  (__try/__except(), everything worked fine. Shrug.

o Stripped the firewall API out of the libdnet included with Nmap
  because Nmap doesn't use it anyway.  This saves space and reduces the
  likelihood of compilation errors and warnings.

o Modified the previously useless --noninteractive option so that it
  deactivates runtime interaction.

Nmap 3.96BETA1

o Added --max_retries option for capping the maximum number of
  retransmissions the port scan engine will do. The value may be as low
  as 0 (no retransmits).  A low value can increase speed, though at the
  risk of losing accuracy.  The -T4 option now allows up to 6 retries,
  and -T5 allows 2.  Thanks to Martin Macok
  (martin.macok(a)underground.cz) for writing the initial patch, which I
  changed quite a bit.  I also updated the docs to reflect this neat
  new option.

o Many of the Nmap low-level timing options take a value in
  milliseconds.  You can now append an 's', 'm', or 'h' to the value
  to give it in seconds, minutes, or hours instead.  So you can specify a
  45 minute host timeout with --host_timeout 45m rather than specifying
  --host_timeout 2700000 and hoping you did the math right and have the
  correct number of zeros.  This also now works for the
  --min_rtt_timeout, --max_rtt_timeout, --initial_rtt_timeout,
  --scan_delay, and --max_scan_delay options.

o Improved the NmapFE port to GTK2 so it better-conforms to the new
  API and you don't get as many annoying messages in your terminal
  window.  GTK2 is prettier and more functional too.  Thanks to Priit
  Laes (amd(a)store20.com) for writing these
  excellent patches.

o Fixed a problem which led to the error message "Failed to determine
  dst MAC address for target" when you try to run Nmap using a
  dialup/PPP adapter on Windows rather than a real ethernet card.  Due
  to Microsoft breaking raw sockets, Nmap no longer supports dialup
  adapters, but it should now give you a clearer error message than
  the "dst MAC address" nonsense.

o Debian GNU/kFreeBSD is now supported thanks to a patch to libdnet's
  configure.in by Petr Salinger (Petr.Salinger(a)t-systems.cz).

o Tried to update to the latest autoconf only to find that there
  hasn't been a new version in more than two years :(.  I was able to
  find new config.sub and config.guess files at
  http://cvs.savannah.gnu.org/viewcvs/config/config/ , so I updated to
  those.

o Fixed a problem with the -e option when run on Windows (or UNIX with
  --send_eth) when run on an ethernet network against an external
  (routed) host.  You would get the message "NmapArpCache() can only
  take IPv4 addresses.  Sorry".  Thanks to KX (kxmail(a)gmail.com) for
  helping to track down the problem.

o Made some changes to allow source port zero scans (-g0).  Nmap used
  to refuse to do this, but now it just gives a warning that it may not
  work on all systems.  It seems to work fine on my Linux box.  Thanks
  to Bill Dale (bill_dale(a)bellsouth.net) for suggesting this feature.

o Made a change to libdnet so that Windows interfaces are listed as
  down if they are disconnected, unplugged, or otherwise unavailable.

o Ceased including foreign translations in the Nmap tarball as they
  take up too much space.  HTML versions can be found at
  http://www.insecure.org/nmap/docs.html , while XML and NROFF versions
  are available from http://www.insecure.org/nmap/data/man-xlate/ .

o Changed INSTALL and README-WIN32 files to mostly just reference the
  new Nmap Install Guide at http://www.insecure.org/nmap/install/ .

o Included docs/nmap-man.xml in the tarball distribution, which is the
  DocBook XML source for the Nmap man page.  Patches to Nmap that are
  user-visible should include patches to the man page XML source rather
  than to the generated Nroff.

o Fixed Nmap so it doesn't crash when you ask it to resume a previous
  scan, but pass in a bogus file rather than actual Nmap output.  Thanks
  to Piotr Sobolewski (piotr_sobolewski(a)o2.pl) for the fix.

Nmap 3.95

o Fixed a crash in IPID Idle scan.  Thanks to Ron
  (iago(a)valhallalegends.com>, Bakeman (bakeman(a)physics.unr.edu),
  and others for reporting the problem.

o Fixed an inefficiency in RPC scan that could slow things down and
  also sometimes resulted in the spurious warning message: "Unable to
  find listening socket in get_rpc_results"

o Fixed a 3.94ALPHA3 bug that caused UDP scan results to be listed as
  TCP ports instead.  Thanks to Justin M Cacak (jcacak(a)nebraska.edu)
  for reporting the problem.

Nmap 3.94ALPHA3

o Updated NmapFE to build with GTK2 rather than obsolete GTK1.  Thanks
  to Mike Basinger (dbasinge(a)speakeasy.net) and Meethune Bhowmick
  (meethune(a)oss-institute.org) for developing the
  patch.  I made some changes as well to prevent compilation warnings.
  The new NmapFE now seems to work, though I do get "Gtk-CRITICAL"
  assertion error messages.  If someone has time to look into this, that
  would be appreciated.

o Fixed a compilation problem on Mac OS X and perhaps other platforms
  with a one-line fix to scan_engine.cc.  Thanks to Felix Gr├╢bert
  (felix(a)groebert.org) for notifying me of the problem.

o Fixed a problem that prevented the command "nmap -sT -PT <targets>"
  from working from a non-privileged user account.  The -PT option
  doesn't change default behavior in this case, but Nmap should (and now
  does) allow it.

o Applied another VS 2005 compatibility patch from KX (kxmail(a)gmail.com).

o Define INET_ADDRSTRLEN in tcpip.h if the system doesn't define it
  for us.  This apparently aids compilation on Solaris 2.6 and 7.
  Thanks to Albert Chin (nmap-hackers(a)mlists.thewrittenword.com) for
  sending the patch..

Nmap 3.94ALPHA2

o Put Nmap on a diet, with changes to the core port scanning routine
  (ultra_scan) to substantially reduce memory consumption, particularly
  when tens of thousands of ports are scanned.

o Fixed a problem with the -S and option on Windows reporting "Failed
  to resolve/decode supposed IPv4 source address".  The -D (decoy)
  option was probably broken on that platform too.  Thanks to KX
  (kxmail(a)gmail.com) for reporting the problem and tracking down a
  potential solution.

o Better handle ICMP type 3, code 0 (network unreachable) responses to
  port scan packets.  These are rarely seen when scanning hosts that
  are actually online, but are still worth handling.

o Applied some small fixes so that Nmap compiles with Visual C++
  2005 Express, which is free from Microsoft at
  http://msdn.microsoft.com/vstudio/express/visualc/ .  Thanks to KX
  (kxmail(a)gmail.com) and Sina Bahram (sbahram(a)nc.rr.com)

o Removed foreign translations of the old man page from the
  distribution.  Included the following contributed translations
  (nroff format) of the new man page:
    Brazilian Portuguese by Lucien Raven (lucienraven(a)yahoo.com.br)
    Portuguese (Portugal) by José Domingos (jd_pt(a)yahoo.com) and
                             Andreia Gaita (shana.ufie(a)gmail.com).

o Added --thc option (undocumented)

o Modified libdnet-stripped/src/eth-bsd.c to allow for up to 128 bpf
  devices rather than 32.  This prevents errors like "Failed to open
  ethernet interface (fxp0)" when there are more than 32 interface
  aliases.  Thanks to Krok (krok(a)void.ru) for reporting the problem
  and even sending a patch.

Nmap 3.94ALPHA1

o Wrote a new man page from scratch.  It is much more comprehensive
  (more than twice as long) and (IMHO) better organized than the
  previous one.  Read it online at http://www.insecure.org/nmap/man/
  or docs/nmap.1 from the Nmap distribution.  Let me know if you have
  any ideas for improving it.

o Wrote a new "help screen", which you get when running Nmap without
  arguments.  It is also reproduced in the man page and at
  http://www.insecure.org/nmap/data/nmap.usage.txt .  I gave up trying
  to fit it within a 25-line, 80-column terminal window.  It is now 78
  lines and summarizes all but the most obscure Nmap options.

o Version detection softmatches (when Nmap determines the service
  protocol such as smtp but isn't able to determine the app name such as
  Postfix) can now parse out the normal match line fields such as
  hostname, device type, and extra info.  For example, we may not know
  what vendor created an sshd, but we can still parse out the protocol
  number.  This was a patch from  Doug Hoyte (doug(a)hcsw.org).

o Fixed a problem which caused UDP version scanning to fail to print
  the matched service.  Thanks to Martin Macok
  (martin.macok(a)underground.cz) for reporting the problem and Doug
  Hoyte (doug(a)hcsw.org) for fixing it.

o Made the version detection "ports" directive (in
  nmap-service-probes) more comprehensive.  This should speed up scans a
  bit.  The patch was done by Doug Hoyte (doug(a)hcsw.org).

o Added the --webxml option, which does the same thing as
  --stylesheet http://www.insecure.org/nmap/data/nmap.xsl , without
  requiring you to remember the exact URL or type that whole thing.

o Fixed a crash occurred when the --exclude option was used with
  netmasks on certain platforms.  Thanks to Adam
  (nmapuser(a)globalmegahost.com) for reporting the problem and to
  Greg Darke (starstuff(a)optusnet.com.au) for sending a patch (I
  modified the patch a bit to make it more efficient).

o Fixed a problem with the -S and -e options (spoof/set
  source address, and set interface by name, respectively).  The problem
  report and a partial patch were sent by Richard Birkett
  (richard(a)musicbox.net).

o Fixed a possible aliasing problem in tcpip.cc by applying a patch sent in by
  Gwenole Beauchesne (gbeauchesne(a)mandriva.com).  This problem
  shouldn't have had any effect on users since we already include the
  -fno-strict-aliasing option whenever gcc 4 is detected, but it
  brings us closer to being able to remove that option.

o Fixed a bug that caused Nmap to crash if an nmap-service-probes file
  was used which didn't contain the Exclude directive.

o Fixed a bunch of typos and misspellings throughout the Nmap source
  code (mostly in comments).  This was a 625-line patch by Saint Xavier
  (skyxav(a)skynet.be).

o Nmap now accepts target list files in Windows end-of-line format (\r\n)
  as well as standard UNIX format (\n) on all platforms.  Passing a
  Windows style file to Nmap on UNIX didn't work before unless you ran
  dos2unix first.

o Removed Identd scan support from NmapFE since Nmap no longer
  supports it.  Thanks to Jonathan Dieter (jdieter99(a)gmx.net) for the
  patch.

o Integrated all of the September version detection fingerprint
  submissions.  This was done by Version Detection Czar Doug Hoyte
  (doug(a)hcsw.org) and resulted in 86 new match lines.  Please keep
  those submissions coming!

o Fixed a divide-by-zero crash when you specify rather bogus
  command-line arguments (a TCP scan with zero tcp ports).  Thanks to
  Bart Dopheide (dopheide(a)fmf.nl) for identifying the problem and
  sending a patch.

o Fixed a minor syntax error in tcpip.h that was causing problems with
  GCC 4.1.  Thanks to Dirk Mueller (dmuell(a)gmx.net) for reporting
  the problem and sending a fix.

Nmap 3.93

o Modified Libpcap's configure.ac to compile with the
  -fno-strict-aliasing option if gcc 4.X is used.  This prevents
  crashes when said compiler is used.  This was done for Nmap in 3.90, but is
  apparently needed for pcap too.  Thanks to Craig Humphrey
  (Craig.Humphrey(a)chapmantripp.com) for the discovery.

o Patched libdnet to include sys/uio.h in src/tun-linux.c.  This is
  apparently necessary on some Glibc 2.1 systems.  Thanks to Rob Foehl
  (rwf(a)loonybin.net) for the patch.

o Fixed a crash which could occur when a ridiculously short
  --host_timeout was specified on Windows (or on UNIX if --send_eth was
  specified).  Nmap now also prints a warning if you specify a
  host_timeout of less than 1 second.  Thanks to Ole Morten Grodaas
  (grodaas(a)gmail.com) for discovering the problem.

Nmap 3.91

o Fixed a crash on Windows when you -P0 scan an unused IP on a local
  network (or a range that contains unused IPs).  This could also
  happen on UNIX if you specified the new --send_eth option.  Thanks
  to Jim Carras (JFCECL(a)engr.psu.edu) for reporting the problem.

o Fixed compilation on OpenBSD by applying a patch from Okan Demirmen
  (okan(a)demirmen.com), who maintains Nmap in the OpenBSD Ports
  collection.

o Updated nmap-mac-prefixes to include OUIs assigned by the IEEE since
  April.

o Updated the included libpcre (used for version detection) from
  version 4.3 to 6.3.  A libpcre security issue was fixed in 6.3, but
  that issue never affected Nmap.

o Updated the included libpcap from 0.8.3 to 0.9.3.  I also changed
  the directory name in the Nmap tarball from libpcap-possiblymodified
  to just libpcap.  As usual, the modifications are described in the
  NMAP_MODIFICATIONS in that directory.

Nmap 3.90

o Added the ability for Nmap to send and properly route raw ethernet
  packets containing IP datagrams rather than always sending the
  packets via raw sockets. This is particularly useful for Windows,
  since Microsoft has disabled raw socket support in XP for no good
  reason.  Nmap tries to choose the best method at runtime based on
  platform, though you can override it with the new --send_eth and
  --send_ip options.

o Added ARP scanning (-PR). Nmap can now send raw ethernet ARP requests to
  determine whether hosts on a LAN are up, rather than relying on
  higher-level IP packets (which can only be sent after a successful
  ARP request and reply anyway).  This is much faster and more
  reliable (not subject to IP-level firewalling) than IP-based probes.
  The downside is that it only works when the target machine is on the
  same LAN as the scanning machine.  It is now used automatically for
  any hosts that are detected to be on a local ethernet network,
  unless --send_ip was specified.  Example usage: nmap -sP -PR
  192.168.0.0/16 .

o Added the --spoof_mac option, which asks Nmap to use the given MAC
  address for all of the raw ethernet frames it sends.  The MAC given
  can take several formats.  If it is simply the string "0", Nmap
  chooses a completely random MAC for the session.  If the given
  string is an even number of hex digits (with the pairs optionally
  separated by a colon), Nmap will use those as the MAC.  If less than
  12 hex digits are provided, Nmap fills in the remainder of the 6
  bytes with random values.  If the argument isn't a 0 or hex string,
  Nmap looks through the nmap-mac-prefixes to find a vendor name
  containing the given string (it is case insensitive).  If a match is
  found, Nmap uses the vendor's OUI (3-byte prefix) and fills out the
  remaining 3 bytes randomly.  Valid --spoof_mac argument examples are
  "Apple", "0", "01:02:03:04:05:06", "deadbeefcafe", "0020F2", and
  "Cisco".

o Applied an enormous nmap-service-probes (version detection) update
  from SoC student Doug Hoyte (doug(a)hcsw.org).  Version 3.81 had
  1064 match lines covering 195 service protocols.  Now we have 2865
  match lines covering 359 protocols!  So the database size has nearly
  tripled!  This should make your -sV scans quicker and more
  accurate.  Thanks also go to the (literally) thousands of you who
  submitted service fingerprints.  Keep them coming!

o Applied a massive OS fingerprint update from Zhao Lei
  (zhaolei(a)gmail.com).  About 350 fingerprints were added, and many
  more were updated.  Notable additions include Mac OS X 10.4 (Tiger),
  OpenBSD 3.7, FreeBSD 5.4, Windows Server 2003 SP1, Sony AIBO (along
  with a new "robotic pet" device type category), the latest Linux 2.6
  kernels Cisco routers with IOS 12.4, a ton of VoIP devices, Tru64
  UNIX 5.1B, new Fortinet firewalls, AIX 5.3, NetBSD 2.0, Nokia IPSO
  3.8.X, and Solaris 10.  Of course there are also tons of new
  broadband routers, printers, WAPs and pretty much any other device
  you can coax an ethernet cable (or wireless card) into!

o Added 'leet ASCII art to the configurator!  ARTIST NOTE: If you think
  the ASCII art sucks, feel free to send me alternatives.  Note that
  only people compiling the UNIX source code get this. (ASCII artist
  unknown).

o Added OS, device type, and hostname detection using the service
  detection framework.  Many services print a hostname, which may be
  different than DNS.  The services often give more away as well.  If
  Nmap detects IIS, it reports an OS family of "Windows".  If it sees
  HP JetDirect telnetd, it reports a device type of "printer".  Rather
  than try to combine TCP/IP stack fingerprinting and service OS
  fingerprinting, they are both printed.  After all, they could
  legitimately be different.  An IP that gives a stack fingerprint
  match of "Linksys WRT54G broadband router" and a service fingerprint
  of Windows based on Kazaa running is likely a common NAT setup rather
  than an Nmap mistake.

o Nmap on Windows now compiles/links with the new WinPcap 3.1
  header/lib files. So please upgrade to 3.1 from
  http://www.winpcap.org before installing this version of Nmap.
  While older versions may still work, they aren't supported with Nmap.

o The official Nmap RPM files are now compiled statically for better
  compatibility with other systems.  X86_64 (AMD Athlon64/Opteron)
  binaries are now available in addition to the standard i386.  NmapFE
  RPMs are no longer distributed by Insecure.Org.

o Nmap distribution signing has changed. Release files are now signed
  with a new Nmap Project GPG key (KeyID 6B9355D0).  Fyodor has also
  generated a new key for himself (KeyID 33599B5F).  The Nmap key has
  been signed by Fyodor's new key, which has been signed by Fyodor's
  old key so that you know they are legit.  The new keys are available
  at http://www.insecure.org/nmap/data/nmap_gpgkeys.txt , as
  docs/nmap_gpgkeys.txt in the Nmap source tarball, and on the public
  keyserver network.  Here are the fingerprints:
    pub  1024D/33599B5F 2005-04-24
         Key fingerprint = BB61 D057 C0D7 DCEF E730  996C 1AF6 EC50 3359 9B5F
    uid  Fyodor <[e-mail]>
    sub  2048g/D3C2241C 2005-04-24

    pub  1024D/6B9355D0 2005-04-24
         Key fingerprint = 436D 66AB 9A79 8425 FDA0  E3F8 01AF 9F03 6B93 55D0
    uid  Nmap Project Signing Key (http://www.insecure.org/)
    sub  2048g/A50A6A94 2005-04-24

o Fixed a crash problem related to non-portable varargs (vsnprintf)
  usage. Reports of this crash came from Alan William Somers
  (somers(a)its.caltech.edu) and Christophe (chris.branch(a)gmx.de).
  This patch was prevalent on Linux boxes running an Opteron/Athlon64
  CPU in 64-bit mode.

o Fixed crash when Nmap is compiled using gcc 4.X by adding the
  -fno-strict-aliasing option when that compiler is detected.  Thanks
  to Greg Darke (starstuff(a)optusnet.com.au) for discovering that
  this option fixes (hides) the problem and to Duilio J. Protti
  (dprotti(a)flowgate.net) for writing the configure patch to detect
  gcc 4 and add the option.  A better fix is to identify and rewrite
  lines that violate C99 alias rules, and we are looking into that.

o Added "rarity" feature to Nmap version detection.  This causes
  obscure probes to be skipped when they are unlikely to help.  Each
  probe now has a "rarity" value.  Probes that detect dozens of
  services such as GenericLines and GetRequest have rarity values of
  1, while the WWWOFFLEctrlstat and mydoom probes have a rarity of 9.
  When interrogating a port, Nmap always tries probes registered to
  that port number.  So even WWWOFFLEctrlstat will be tried against
  port 8081 and mydoom will be tried against open ports between 3127
  and 3198.  If none of the registered ports find a match, Nmap tries
  probes that have a rarity less than or equal to its current
  intensity level.  The intensity level defaults to 7 (so that most of
  the probes are done).  You can set the intensity level with the new
  --version_intensity option.  Alternatively, you can just use
  --version_light or --version_all which set the intensity to 2 (only
  try the most important probes and ones registered to the port
  number) and 9 (try all probes), respectively.  --version_light is
  much faster than default version detection, but also a bit less
  likely to find a match.  This feature was designed and implemented
  by Doug Hoyte (doug(a)hcsw.org).

o Added a "fallback" feature to the nmap-service-probes database.
  This allows a probe to "inherit" match lines from other probes.  It
  is currently only used for the HTTPOptions, RTSPRequest, and
  SSLSessionReq probes to inherit all of the match lines from
  GetRequest.  Some servers don't respond to the Nmap GetRequest (for
  example because it doesn't include a Host: line) but they do respond
  to some of those other 3 probes in ways that GetRequest match lines
  are general enough to match.  The fallback construct allows us to
  benefit from these matches without repeating hundreds of signatures
  in the file.  This is another feature designed and implemented
  by Doug Hoyte (doug(a)hcsw.org).

o Fixed crash with certain --excludefile or
  --exclude arguments.  Thanks to Kurt Grutzmacher
  (grutz(a)jingojango.net) and pijn trein (ptrein(a)gmail.com) for
  reporting the problem, and to Duilio J. Protti
  (dprotti(a)flowgate.net) for debugging the issue and sending the
  patch.

o Updated random scan (ip_is_reserved()) to reflect the latest IANA
  assignments.  This patch was sent in by Felix Groebert
  (felix(a)groebert.org).

o Included new Russian man page translation by
  locco_bozi(a)Safe-mail.net

o Applied patch from Steve Martin (smartin(a)stillsecure.com) which
  standardizes many OS names and corrects typos in nmap-os-fingerprints.

o Fixed a crash found during certain UDP version scans.  The crash was
  discovered and reported by Ron (iago(a)valhallalegends.com) and fixed
  by Doug Hoyte (doug(a)hcsw.com).

o Added --iflist argument which prints a list of system interfaces and
  routes detected by Nmap.

o Fixed a protocol scan (-sO) problem which led to the error message:
  "Error compiling our pcap filter: syntax error".  Thanks to Michel
  Arboi (michel(a)arboi.fr.eu.org) for reporting the problem.

o Fixed an Nmap version detection crash on Windows which led to the
  error message "Unexpected error in NSE_TYPE_READ callback.  Error
  code: 10053 (Unknown error)".  Thanks to Srivatsan
  (srivatsanp(a)adventnet.com) for reporting the problem.

o Fixed some misspellings in docs/nmap.xml reported by Tom Sellers
  (TSellers(a)trustmark.com).

o Applied some changes from  Gisle Vanem (giva(a)bgnett.no) to make
  Nmap compile with Cygwin.

o XML "osmatch" element now has a "line" attribute giving the
  reference fingerprint line number in nmap-os-fingerprints.

o Added a distcc probes and a bunch of smtp matches from Dirk Mueller
  (mueller(a)kde.org) to nmap-service-probes.  Also added AFS version
  probe and matches from Lionel Cons (lionel.cons(a)cern.ch).  And
  even more probes and matches from Martin Macok
  (martin.macok(a)underground.cz)

o Fixed a problem where Nmap compilation would use header files from
  the libpcap included with Nmap even when it was linking to a system
  libpcap.  Thanks to Solar Designer (solar(a)openwall.com) and Okan
  Demirmen (okan(a)demirmen.com) for reporting the problem.

o Added configure option --with-libpcap=included to tell Nmap to use
  the version of libpcap it ships with rather than any that may already be
  installed on the system.  You can still use --with-libpcap=[dir] to
  specify that a system libpcap be installed rather than the shipped
  one.  By default, Nmap looks at both and decides which one is likely
  to work best.  If you are having problems on Solaris, try
  --with-libpcap=included .

o Changed the --no-stylesheet option to --no_stylesheet to be
  consistent with all of the other Nmap options.  Though I'm starting to
  like hyphens a bit better than underscores and may change all of the
  options to use hyphens instead at some point.

o Added "Exclude" directive to nmap-service-probes grammar which
  causes version detection to skip listed ports.  This is helpful for
  ports such as 9100.  Some printers simply print any data sent to
  that port, leading to pages of HTTP requests, SMB queries, X Windows
  probes, etc.  If you really want to scan all ports, specify
  --allports.  This patch came from Doug Hoyte (doug(a)hcsw.org).

o Added a stripped-down and heavily modified version of Dug Song's
  libdnet networking library (v. 1.10).  This helps with the new raw
  ethernet features.  My (extensive) changes are described in
  libdnet-stripped/NMAP_MODIFICATIONS

o Removed WinIP library (and all Windows raw sockets code) since MS
  has gone and broken raw sockets.  Maybe packet receipt via raw
  sockets will come back at some point.  As part of this removal, the
  Windows-specific --win_help, --win_list_interfaces, --win_norawsock,
  --win_forcerawsock, --win_nopcap, --win_nt4route, --win_noiphlpapi,
  and --win_trace options have been removed.

o Changed the interesting ports array from a 65K-member array of
  pointers into an STL list.  This noticeable reduces memory usage in
  some cases, and should also give a slight runtime performance
  boost. This patch was written by Paul Tarjan (ptarjan(a)gmail.com).

o Removed the BSDFIX/BSDUFIX macros.  The underlying bug in
  FreeBSD/NetBSD is still there though.  When an IP packet is sent
  through a raw socket, these platforms require the total length and
  fragmentation offset fields of an IP packet to be in host byte order
  rather than network byte order, even though all the other fields
  must be in NBO.  I believe that OpenBSD fixed this a while back.
  Other platforms, such as Linux, Solaris, Mac OS X, and Windows take
  all of the fields in network byte order.  While I removed the macro,
  I still do the munging where required so that Nmap still works on
  FreeBSD.

o Integrated many nmap-service-probes changes from Bo Jiang
  (jiangbo(a)brandeis.edu)

o Added a bunch of RPC numbers from nmap-rpc maintainer Eilon Gishri
  (eilon(a)aristo.tau.ac.il)

o Added some new RPC services to nmap-rpc thanks to a patch from
  vlad902 (vlad902(a)gmail.com).

o Fixed a bug where Nmap would quit on Windows whenever it encountered
  a raw scan of localhost (including the local ethernet interface
  address), even when that was just one address out of a whole network
  being scanned.  Now Nmap just warns that it is skipping raw scans when
  it encounters the local IP, but continues on to scan the rest of the
  network.  Raw scans do not currently work against local IP addresses
  because Winpcap doesn't support reading/writing localhost interfaces
  due to limitations of Windows.

o The OS fingerprint is now provided in XML output if debugging is
  enabled (-d) or verbosity is at least 2 (-v -v).  This patch was
  sent by Okan Demirmen (okan(a)demirmen.com)

o Fixed the way tcp connect scan (-sT) response to ICMP network
  unreachable responses (patch by Richard Moore
  (rich(a)westpoint.ltd.uk).

o Update random host scan (-iR) to support the latest IANA-allocated
  ranges, thanks to patch by Chad Loder (cloder(a)loder.us).

o Updated GNU shtool (a helper program used during 'make install' to
  version 2.0.2, which fixes a predictable temporary filename
  weakness discovered by Eric Raymond.

o Removed addport element from XML DTD, since it is no longer used
  (suggested by Lionel Cons (lionel.cons(a)cern.ch)

o Added new --privileged command-line option and NMAP_PRIVILEGED
  environmental variable.  Either of these tell Nmap to assume that
  the user has full privileges to execute raw packet scans, OS
  detection and the like.  This can be useful when Linux kernel
  capabilities or other systems are used that allow non-root users to
  perform raw packet or ethernet frame manipulation.  Without this
  flag or variable set, Nmap bails on UNIX if geteuid() is
  nonzero.

o Changed the RPM spec file so that if you define "static" to 1 (by
  passing --define "static 1" to rpmbuild), static binaries are built.

o Fixed Nmap compilation on Solaris x86 thanks to a patch from Simon
  Burr (simes(a)bpfh.net).

o ultra_scan() now sets pseudo-random ACK values (rather than 0) for
  any TCP scans in which the initial probe packet has the ACK flag set.
  This would be the ACK, Xmas, Maimon, and Window scans.

o Updated the Nmap version number, description, and similar fields
  that MS Visual Studio places in the binary.  This was done by editing
  mswin32/nmap.rc as suggested by Chris Paget (chrisp(a)ngssoftware.com)

o Fixed Nmap compilation on DragonFly BSD (and perhaps some other
  systems) by applying a short patch by Joerg Sonnenberger which omits
  the declaration of errno if it is a #define.

o Fixed an integer overflow that prevented Nmap from scanning
  2,147,483,648 hosts in one expression (e.g. 0.0.0.0/1).  Problem
  noted by Justin Cranford (jcranford(a)n-able.com).  While /1 scans
  are now possible, don't expect them to finish during your bathroom
  break.  No matter how constipated you are.

o Increased the buffer size allocated for fingerprints to prevent Nmap
  from running out and quitting (error message: "Assertion
  `servicefpalloc - servicefplen > 8' failed".  Thanks to Mike Hatz
  (mhatz(a)blackcat.com) for the report. [ Actually this was done in a
  previous version, but I forgot which one ]

o Changed from CVS to Subversion source control system (which
  rocks!). Neither repository is public (I'm paranoid because both CVS
  and SVN have had remotely exploitable security holes), so the main
  change users will see is that "Id" tags in file headers use the SVN
  format for version numbering and such.

Nmap 3.81

o Nmap now ships with and installs (in the same directory as other
  data files such as nmap-os-fingerprints) an XSL stylesheet for
  rendering the XML output as HTML.  This stylesheet was written by
  Benjamin Erb ( see http://www.benjamin-erb.de/nmap/ for examples).
  It supports tables, version detection, color-coded port states, and
  more.  The XML output has been augmented to include an
  xml-stylesheet directive pointing to nmap.xsl on the local
  filesystem.  You can point to a different XSL file by providing the
  filename or URL to the new --stylesheet argument.  Omit the
  xml-stylesheet directive entirely by specifying --no-stylesheet.
  The XML to HTML conversion can be done with an XSLT processor such
  as Saxon, Sablot, or Xalan, but modern browsers can do this on the
  fly -- simply load the XML output file in IE or Firefox.  Some
  features don't currently work with Firefox's on-the-fly rendering.
  Perhaps some Mozilla wizard can fix that in either the XSL or the
  browser itself.  I hate having things work better in IE :).  It is
  often more convenient to have the stylesheet loaded from a URL
  rather than the local filesystem, allowing the XML to be rendered on
  any machine regardless of whether/where the XSL is installed.  For
  privacy reasons (avoid loading of an external URL when you view
  results), Nmap uses the local filesystem by default.  If you would
  like the latest version of the stylesheet loaded from the web when
  rendering, specify
  --stylesheet http://www.insecure.org/nmap/data/nmap.xsl .

o Fixed fragmentation option (-f).  One -f now sets sends fragments
  with just 8 bytes after the IP header, while -ff sends 16 bytes to
  reduce the number of fragments needed.  You can specify your own
  fragmentation offset (must be a multiple of 8) with the new --mtu
  flag.  Don't also specify -f if you use --mtu.  Remember that some
  systems (such as Linux with connection tracking) will defragment in
  the kernel anyway -- so test first while sniffing with ethereal.
  These changes are from a patch by Martin Macok
  (martin.macok(a)underground.cz).

o Nmap now prints the number (and total bytes) of raw IP packets sent
  and received when it completes, if verbose mode (-v) is enabled.  The
  report looks like:
  Nmap finished: 256 IP addresses (3 hosts up) scanned in 30.632 seconds
                 Raw packets sent: 7727 (303KB) | Rcvd: 6944 (304KB)

o Fixed (I hope) an error which would cause the Windows version of
  Nmap to abort under some circumstances with the error message
  "Unexpected error in NSE_TYPE_READ callback.  Error code: 10053
  (Unknown error)".  Problem reported by "Tony Golding"
  (biz(a)tonygolding.com).

o Added new "closed|filtered" state.  This is used for Idle scan, since
  that scan method can't distinguish between those two states.  Nmap
  previously just used "closed", but this is more accurate.

o Null, FIN, Maimon, and Xmas scans now mark ports as "open|filtered"
  instead of "open" when they fail to receive any response from the
  target port.  After all, it could just as easily be filtered as open.
  This is the same change that was made to UDP scan in 3.70.  Also as
  with UDP scan, adding version detection (-sV) will change the state
  from open|filtered to open if it confirms that they really are open.

o Fixed a bug in ACK scan that could cause Nmap to crash with the
  message "Unexpected port state: 6" in some cases.  Thanks to Glyn
  Geoghegan (glyng(a)corsaire.com) for reporting the problem.

o Change IP protocol scan (-sO) so that a response from the target
  host in any protocol at all will prove that protocol is open.  As
  before, no response means "open|filtered", an ICMP protocol
  unreachable means "closed", and most other ICMP error messages mean
  "filtered".

o Patched a libpcap issue that prevented read timeouts from being
  honored on Solaris (thus slowing down Nmap substantially).  The
  problem report and patch were sent in by Ben Harris
  (bjh21(a)cam.ac.uk).

o Changed IP protocol scan (-sO) so that it sends valid ICMP, TCP, and
  UDP headers when scanning protocols 1, 6, and 17, respectively.  An
  empty IP header is still sent for all other protocols.  This should
  prevent the error messages such as "sendto in send_ip_packet:
  sendto(3, packet, 20, 0, 192.31.33.7, 16) => Operation not
  permitted" that Linux (and perhaps other systems) would give when
  they try to interpret the raw packet.  This also makes it more
  likely that these protocols will elicit a response, proving that the
  protocol is "open".

o The windows build now uses header and static library files from
  Winpcap 3.1Beta4.  It also now prints out the DLL version you are
  using when run with -d.  I would recommend upgrading to 3.1Beta4 if
  you have an older Winpcap installed.

o Nmap now prints a warning message on Windows if Winpcap is not found
  (it then reverts to raw sockets mode if available, as usual).

o Added an NTP probe and matches to the version detection database
  (nmap-service-probes) thanks to a submission from Martin
  Macok (martin.macok(a)underground.cz).

o Applied several Nmap service detection database updates sent in by
  Martin Macok (martin.macok(a)underground.cz).

o The XML nmaprun element now has a startstr attribute which gives the
  human readable calendar time format that a scan started.  Similarly
  the finished element now has a timestr attribute describing when the
  scan finished.  These are in addition to the existing nmaprun/start
  and finished/time attributes that provided the start and finish time
  in UNIX time_t notation.  This should help in development of
  XSLT stylesheets for Nmap XML output.

o Fixed a memory leak that would generally consume several hundred
  bytes per down host scanned.  While the effect for most scans is
  negligible, it was overwhelming when Scott Carlson
  (Scott.Carlson(a)schwab.com) tried to scan 16.8 million IPs
  (10.0.0.0/8).  Thanks to him for reporting the problem.  Also thanks
  to Valgrind ( http://valgrind.kde.org ) for making it easy to debug.

o Fixed a crash on Windows systems that don't include the iphlpapi
  DLL.  This affects Win95 and perhaps other variants.  Thanks to Ganga
  Bhavani (GBhavani(a)everdreamcorp.com) for reporting the problem and
  sending the patch.

o Ensured that the device type, os vendor, and os family OS
  fingerprinting classification values are scrubbed for XML compliance
  in the XML output.  Thanks to Matthieu Verbert
  (mve(a)zurich.ibm.com) for reporting the problem and sending a patch.

o Rewrote the host IP (target specification) parser for easier
  maintenance and to fix a bug found by Netris (netris(a)ok.kz)

o Changed to Nmap XML DTD to use the same xmloutputversion (1.01) as
  newer versions of Nmap.  Thanks to Laurent Estieux
  (laurent.estieux(a)free.fr) for reporting the problem.

o Fixed compilation on some HP-UX 11 boxes thanks to a patch by Petter
  Reinholdtsen (pere(a)hungry.com).

o Fixed a portability problem on some OpenBSD and FreeBSD machines
  thanks to a patch by Okan Demirmen (okan(a)demirmen.com).

o Applied Martin Macok's (martin.macok(a)underground.cz) "cosmetics
  patch", which fixes a few typos and minor problems.

Nmap 3.75

o Implemented a huge OS fingerprint database update.  The number of
  fingerprints increased more than 20% to 1,353 and many of the
  existing ones are much improved.  Notable updates include the fourth
  edition of Bell Lab's Plan9, Grandstream's BugeTone 101 IP Phone,
  and Bart's Network Boot Disk 2.7 (which runs MS-DOS).  Oh, and Linux
  kernels up to 2.6.8, dozens of new Windows fingerprints including XP
  SP2, the latest Longhorn warez, and many modified Xboxes, OpenBSD
  3.6, NetBSD up to 2.0RC4, Apple's AirPort Express WAP and OS X
  10.3.3 (Panther) release, Novell Netware 6.5, FreeBSD 5.3-BETA, a
  bunch of Linksys and D-Link consumer junk, the latest Cisco IOS 12.2
  releases, a ton of miscellaneous broadband routers and printers, and
  much more.

o Updated nmap-mac-prefixes with the latest OUIs from the IEEE.
  [ http://standards.ieee.org/regauth/oui/oui.txt ]

o Updated nmap-protocols with the latest IP protocols from IANA
  [ http://www.iana.org/assignments/protocol-numbers ]

o Added a few new Nmap version detection signatures thanks to a patch
  from Martin Macok (martin.macok(a)underground.cz).

o Fixed a crash problem in the Windows version of Nmap, thanks to a
  patch from Ganga Bhavani GBhavani(a)everdreamcorp.com).

o Fixed Windows service scan crashes that occur with the error message
  "Unexpected nsock_loop error. Error code 10022 (Unknown error)".  It
  turns out that Windows does not allow select() calls with all three
  FD sets empty.  Lame.  The Linux select() man page even suggests
  calling "select with all three sets empty, n zero, and a non-null
  timeout as a fairly portable way to sleep with subsecond precision."
  Thanks to Gisle Vanem (giva(a)bgnett.no) for debugging help.

o Added --max_scan_delay parameter.  Nmap will sometimes increase the
  delay itself when it detects many dropped packets.  For example,
  Solaris systems tend to respond with only one ICMP port unreachable
  packet per second during a UDP scan.  So Nmap will try to detect
  this and lower its rate of UDP probes to one per second.  This can
  provide more accurate results while reducing network congestion, but
  it can slow the scans down substantially.  By default (with no -T
  options specified), Nmap allows this delay to grow to one second per
  probe.  This option allows you to set a lower or higher maximum.
  The -T4 and -T5 scan modes now limit the maximum scan delay for TCP
  scans to 10 and 5 ms, respectively.

o Fixed a bug that prevented RPC scan (-sR) from working for UDP ports
  unless service detection (-sV) was used.  -sV is still usually a
  better approach than -sR, as the latter ONLY handles RPC.  Thanks to
  Stephen Bishop (sbishop(a)idsec.co.uk) for reporting the problem and
  sending a patch.

o Fixed nmap_fetchfile() to better find custom versions of data files
  such as nmap-services.  Note that the implicitly read directory
  should be ~/.nmap rather than ~/nmap .  So you may have to move any
  customized files you now have in ~/nmap .  Thanks to nnposter
  (nnposter(a)users.sourceforge.net) for reporting the problem and
  sending a patch.

o Changed XML output so that the MAC address [address] element comes
  right after the IPv4/IPv6 [address] element.  Apparently this is
  needed to comply with the DTD (
  http://www.insecure.org/nmap/data/nmap.dtd ).  Thanks to Adam Morgan
  (adam.morgan(a)Q1Labs.com) and Florian Ebner
  (Florian.Ebner(a)e-bros.de) for the problem reports.

o Fixed an error in the Nmap RPM spec file reported by Pascal Trouvin
  (pascal.trouvin(a)wanadoo.fr)

o Fixed a timing problem in which a specified large --send_delay would
  sometimes be reduced to 1 second during a scan.  Thanks to Martin
  Macok (martin.macok(a)underground.cz) for reporting the problem.

o Fixed a timing problem with sneaky and paranoid modes (-T1 and -T0)
  which would cause Nmap to continually scan the same port and never
  hit other ports when scanning certain firewalled hosts.  Thanks to
  Curtis Doty (Curtis(a)GreenKey.net) for reporting the problem.

o Fixed a bug in the build system that caused most Nmap subdirectories
  to be configured twice.  Changing the variable holding the name of
  subdirs from $subdirs to $nmap_cfg_subdirs resolved the problem --
  configure must have been using that variable name for its own internal
  operations.  Anyway, this should reduce compile time significantly.

o Made a trivial change to nsock/src/nsock_event.c to work around a "a
  bug in GCC 3.3.1 on FreeBSD/sparc64".  I found the patch by digging
  around the FreeBSD ports tree repository.  It would be nice if the
  FreeBSD Nmap port maintainers would report such things to me, rather
  than fixing it in their own Nmap tree and then applying the patch to
  every future version.  On the other hand, they deserve some sort of
  "most up-to-date" award.  I stuck Nmap 3.71-PRE1 in the dist
  directory for a few people to test, and made no announcement or
  direct link.  The FreeBSD crew found it and upgraded anyway :).  The
  gcc-workaround patch was apparently submitted to the FreeBSD folks
  by Marius Strobl (marius(a)alchemy.franken.de).

o Fixed (I hope) an OS detection timing issue which would in some
  cases lead to the warning that "insufficient responses for TCP
  sequencing (3), OS detection may be less accurate."  Thanks to Adam
  Kerrison (adam(a)tideway.com) for reporting the problem.

o Modified the warning given when files such as nmap-services exist in
  both the compiled in NMAPDATADIR and the current working directory.
  That message should now only appear once and is more clear.

o Fixed ping scan subsystem to work a little bit better when
  --scan_delay (or some of the slower -T templates which include a scan
  delay) is specified.  Thanks to Shahid Khan (khan(a)asia.apple.com)
  for suggestions.

o Taught connect() scan to properly interpret ICMP protocol
  unreachable messages.  Thanks to Alan Bishoff
  (abishoff(a)arc.nasa.gov) for the report.

o Improved the nmapfe.desktop file to better comply with standards.
  Thanks to Stephane Loeuillet (stephane.loeuillet(a)tiscali.fr) for
  sending the patch.

Nmap 3.70

o Rewrote core port scanning engine, which is now named ultra_scan().
  Improved algorithms make this faster (often dramatically so) in
  almost all cases.  Not only is it superior against single hosts, but
  ultra_scan() can scan many hosts (sometimes hundreds) in parallel.
  This offers many efficiency/speed advantages.  For example, hosts
  often limit the ICMP port unreachable packets used by UDP scans to
  1/second.  That made those scans extraordinarily slow in previous
  versions of Nmap.  But if you are scanning 100 hosts at once,
  suddenly you can receive 100 responses per second.  Spreading the
  scan amongst hosts is also gentler toward the target hosts.  Nmap
  can still scan many ports at the same time, as well.  If you find
  cases where ultra_scan is slower or less accurate, please send a
  report (including exact command-lines, versions used, and output, if
  possible) to Fyodor.

o Added --max_hostgroup option which specifies the maximum number of
  hosts that Nmap is allowed to scan in parallel.

o Added --min_hostgroup option which specifies the minimum number of
  hosts that Nmap should scan in parallel (there are some exceptions
  where Nmap will still scan smaller groups -- see man page).  Of
  course, Nmap will try to choose efficient values even if you don't
  specify hostgroup restrictions explicitly.

o Rewrote TCP SYN, ACK, Window, and Connect() scans to use
  ultra_scan() framework, rather than the old pos_scan().

o Rewrote FIN, Xmas, NULL, Maimon, UDP, and IP Protocol scans to use
  ultra_scan(), rather than the old super_scan().

o Overhauled UDP scan.  Ports that don't respond are now classified as
  "open|filtered" (open or filtered) rather than "open".  The (somewhat
  rare) ports that actually respond with a UDP packet to the empty
  probe are considered open.  If version detection is requested, it
  will be performed on open|filtered ports.  Any that respond to any of
  the UDP probes will have their status changed to open.  This avoids a
  the false-positive problem where filtered UDP ports appear to be
  open, leading to terrified newbies thinking their machine is
  infected by back orifice.

o Nmap now estimates completion times for almost all port scan types
  (any that use ultra_scan()) as well as service scan (version
  detection).  These are only shown in verbose mode (-v).  On scans
  that take more than a minute or two, you will see occasional updates
  like:
  SYN Stealth Scan Timing: About 30.01% done; ETC: 16:04 (0:01:09 remaining)
  New updates are given if the estimates change significantly.

o Added --exclude option, which lets you specify a comma-separated
  list of targets (hosts, ranges, netblocks) that should be excluded
  from the scan.  This is useful to keep from scanning yourself, your
  ISP, particularly sensitive hosts, etc.  The new --excludefile reads
  the list (newline-delimited) from a given file.  All the work was
  done by Mark-David McLaughlin (mdmcl(a)cisco.com> and William McVey
  ( wam(a)cisco.com ), who sent me a well-designed and well-tested
  patch.

o Nmap now has a "port scan ping" system.  If it has received at least
  one response from any port on the host, but has not received
  responses lately (usually due to filtering), Nmap will "ping" that
  known-good port occasionally to detect latency, packet drop rate,
  etc.

o Service/version detection now handles multiple hosts at once for
  more efficient and less-intrusive operation.

o Nmap now wishes itself a happy birthday when run on September 1 in
  verbose mode!  The first public release was on that date in 1997.

o The port randomizer now has a bias toward putting
  commonly-accessible ports (80, 22, etc.) near the beginning of the
  list.  Getting a response early helps Nmap calculate response times and
  detect packet loss, so the scan goes faster.

o Host timeout system (--host_timeout) overhauled to support host
  parallelization.  Hosts times are tracked separately, so a host that
  finishes a SYN scan quickly is not penalized for an exceptionally
  slow host being scanned at the same time.

o When Nmap has not received any responses from a host, it can now
  use certain timing values from other hosts from the same scan
  group.  This way Nmap doesn't have to use absolute-worst-case
  (300bps SLIP link to Uzbekistan) round trip timeouts and such.

o Enabled MAC address reporting when using the Windows version
  of Nmap.  Thanks to Andy Lutomirski (luto(a)stanford.edu) for
  writing and sending the patch.

o Workaround crippled raw sockets on Microsoft Windows XP SP2 scans.
  I applied a patch by Andy Lutomirski (luto(a)stanford.edu) which
  causes Nmap to default to WinPcap sends instead.  The WinPcap send
  functionality was already there for versions of Windows such as NT and
  Win98 that never supported Raw Sockets in the first place.

o Changed how Nmap sends ARP requests on Windows to use the iphlpapi
  SendARP() function rather than creating it raw and reading the
  response from the Windows ARP cache.  This works around a
  (reasonable) feature of Windows Firewall which ignored such
  unsolicited responses.  The firewall is turned on by default as of
  Windows XP SP2.  This change was implemented by Dana Epp
  (dana(a)vulscan.com).

o Fixed some Windows portability issues discovered by Gisle Vanem
  (giva(a)bgnett.no).

o Upgraded libpcap from version 0.7.2 to 0.8.3.  This was an attempt
  to fix an annoying bug, which I then found was actually in my code
  rather than libpcap :).

o Removed Ident scan (-I).  It was rarely useful, and the
  implementation would have to be rewritten for the new ultra_scan()
  system.  If there is significant demand, perhaps I'll put it back in
  sometime.

o Documented the --osscan_limit option, which saves time by skipping
  OS detection if at least one open and one closed port are not found on
  the remote hosts.  OS detection is much less reliable against such
  hosts anyway, and skipping it can save some time.

o Updated nmapfe.desktop file to provide better NmapFE desktop support
  under Fedora Core and other systems.  Thanks to Mephisto
  (mephisto(a)mephisto.ma.cx) for sending the patch.

o Further nmapfe.desktop changes to better fit the freedesktop
  standard.  The patch came from Murphy (m3rf(a)swimmingnoodle.com).

o Fixed capitalization (with a Perl script) of many over-capitalized
  vendor names in nmap-mac-prefixes.

o Ensured that MAC address vendor names are always escaped in XML
  output if they contain illegal characters (particularly '&').  Thanks
  to Matthieu Verbert (mve(a)zurich.ibm.com) for the report and a patch.

o Changed xmloutputversion in XML output from 1.0 to 1.01 to note that
  there was a slight change (which was actually the MAC stuff in 3.55).
  Thanks to Lionel CONS (lionel.cons(a)cern.ch) for the suggestion.

o Many Windows portability fix and bug fixes, thanks to patch from
  Gisle Vanem (giva(a)bgnett.no).  With these changes, he was able to
  compile Nmap on Windows using MingW + gcc 3.4 C++ rather than MS
  Visual Studio.

o Removed (addport) tags from XML output.  They used to provide open
  ports as they were discovered, but don't work now that the port
  scanners scan many hosts at once.  They did not specify an IP
  address.  Of course the appropriate (port) tags are still printed
  once scanning of a target is complete.

o Configure script now detects GNU/k*BSD systems (whatever those are),
  thanks to patch from Robert Millan (rmh(a)debian.org)

o Fixed various crashes and assertion failures related to the new
  ultra_scan() system, that were found by Arturo "Buanzo" Busleiman
  (buanzo(a)buanzo.com.ar), Eric (catastrophe.net), and Bill Petersen
  (bill.petersen(a)alcatel.com).

o Fixed some minor memory leaks relating to ping and list scanning as
  well as the Nmap output table.  These were found with Valgrind (
  http://valgrind.kde.org/ ).

o Provide limited --packet_trace support for TCP connect() (-sT)
  scans.

o Fixed compilation on certain Solaris machines thanks to a patch by
  Tom Duffy (tduffy(a)sun.com)

o Fixed some warnings that crop up when compiling Nbase C files with a
  C++ compiler.  Thanks to Gisle Vanem (giva(a)bgnett.no) for sending
  the patch.

o Tweaked the License blurb on source files and in the man page.  It
  clarifies some issues and includes a new GPL exception that
  explicitly allows linking with the OpenSSL library.  Some people
  believe that the GPL and OpenSSL licenses are incompatible without
  this special exception.

o Fixed some serious runtime portability issues on *BSD systems.
  Thanks to Eric (catastrophe.net) for reporting the problem.

o Changed the argument parser to better detect bogus arguments to the
  -iR option.

o Removed a spurious warning message relating to the Windows ARP cache
  being empty.  Patch by Gisle Vanem (giva(a)bgnett.no).

o Removed some C++-style line comments (//) from nbase, because some C
  compilers (particularly on Solaris) barf on those.  Problem reported
  by Raju Alluri <Raju.Alluri(a)Sun.COM>

Nmap 3.55

o Added MAC address printing.  If Nmap receives packet from a target
  machine which is on an Ethernet segment directly connected to the
  scanning machine, Nmap will print out the target MAC address.  Nmap
  also now contains a database (derived from the official IEEE
  version) which it uses to determine the vendor name of the target
  ethernet interface.  The Windows version of Nmap does not yet have
  this capability.  If any Windows developer types are interesting in
  adding it, you just need to implement IPisDirectlyConnected() in
  tcpip.cc and then please send me the patch.  Here are examples from
  normal and XML output (angle brackets replaced with [] for HTML
  changelog compatibility):
  MAC Address: 08:00:20:8F:6B:2F (SUN Microsystems)
  [address addr="00:A0:CC:63:85:4B" vendor="Lite-on Communications" addrtype="mac" /]

o Updated the XML DTD to support the newly printed MAC addresses.
  Thanks to Thorsten Holz (thorsten.holz(a)mmweg.rwth-aachen.de) for
  sending this patch.

o Added a bunch of new and fixed service fingerprints for version
  detection.  These are from Martin Macok
  (martin.macok(a)underground.cz).

o Normalized many of the OS names in nmap-os-fingerprints (fixed
  capitalization, typos, etc.).  Thanks to Royce Williams
  (royce(a)alaska.net) and Ping Huang (pshuang(a)alum.mit.edu) for
  sending patches.

o Modified the mswine32/nmap_performance.reg Windows registry file to
  use an older and more compatible version.  It also now includes the
  value "StrictTimeWaitSeqCheck"=dword:00000001 , as suggested by Jim
  Harrison (jmharr(a)microsoft.com).  Without that latter value, the
  TcpTimedWaitDelay value apparently isn't checked.  Windows users
  should apply the new registry changes by clicking on the .reg file.
  Or do it manually as described in README-WIN32.  This file is also
  now available in the data directory at
  http://www.insecure.org/nmap/data/nmap_performance.reg

o Applied patch from Gisle Vanem (giva(a)bgnett.no) which allows the
  Windows version of Nmap to work with WinPCAP 3.1BETA (and probably
  future releases).  The Winpcap folks apparently changed the encoding
  of adapter names in this release.

o Fixed a ping scanning bug that would cause this error message: "nmap:
  targets.cc:196: int hostupdate (Target **, Target *, int, int, int,
  timeout_info *, timeval *, timeval *, pingtune *, tcpqueryinfo *,
  pingstyle): Assertion `pt->down_this_block > 0' failed."  Thanks to
  Beirne Konarski (beirne(a)neo.rr.com) for reporting the problem.

o If a user attempts -PO (the letter O), print an error suggesting
  that they probably mean -P0 (Zero) to disable ping scanning.

o Applied a couple patches (with minor changes) from Oliver Eikemeier
  (eikemeier(a)fillmore-labs.com) which fix an edge case relating to
  decoy scanning IP ranges that must be sent through different
  interfaces, and improves the Nmap response to certain error codes
  returned by the FreeBSD firewall system.  The patches are from
  http://cvsweb.freebsd.org/ports/security/nmap/files/ .

o Many people have reported this error: "checking for type of 6th
  argument to recvfrom()... configure: error: Cannot find type for 6th
  argument to recvfrom()".  In most cases, the cause was a missing or
  broken C++ compiler.  That should now be detected earlier with a
  clearer message.

o Fixed the FTP bounce scan to better detect filtered ports on the
  target network.

o Fixed some minor bugs related to the new MAC address printing
  feature.

o Fixed a problem with UDP-scanning port 0, which was reported by
  Sebastian Wolfgarten (sebastian(a)wolfgarten.com).

o Applied patch from Ruediger Rissmann (RRI(a)zurich.ibm.com), which
  helps Nmap understand an EACCESS error, which can happen at least
  during IPv6 scans from certain platforms to some firewalled targets.

o Renamed ACK ping scan option from -PT to -PA in the documentation.
  Nmap has accepted both names for years and will continue to do
  so.

o Removed the notice that Nmap is reading target specifications from a
  file or stdin when you specify the -iL option.  It was sometimes
  printed to stdout even when you wanted to redirect XML or grepable
  output there, because it was printed during options processing before
  output files were handled.  This change was suggested by Anders Thulin
  (ath(a)algonet.se).

o Added --source_port as a longer, but hopefully easier to remember,
  alias for -g.  In other words, it tries to use the constant source
  port number you specify for probes.  This can help against poorly
  configured firewalls that trust source port 20, 53, and the like.

o Removed undocumented (and useless) -N option.

o Fixed a version detection crash reported in excellent detail by
  Jedi/Sector One (j(a)pureftpd.org).

o Applied patch from Matt Selsky (selsky(a)columbia.edu) which helps
  Nmap build with OpenSSL.

o Modified the configure/build system to fix library ordering problems
  that prevented Nmap from building on certain platforms.  Thanks to
  Greg A. Woods (woods(a)weird.com) and Saravanan
  (saravanan_kovai(a)HotPop.com) for the suggestions.

o Applied a patch to Makefile.in from Scott Mansfield
  (thephantom(a)mac.com) which enables the use of a DESTDIR variable
  to install the whole Nmap directory structure under a different root
  directory.  The configure --prefix option would do the same thing in
  this case, but DESTDIR is apparently a standard that package
  maintainers like Scott are used to.  An example usage is
  "make DESTDIR=/tmp/packageroot".

o Removed unnecessary banner printing in the non-root connect() ping
  scan.  Thanks to Tom Rune Flo (tom(a)x86.no) for the suggestion and
  a patch.

o Updated the headers at the top of each source file (mostly to
  advance the copyright year to 2004 and note that Nmap is a registered
  trademark).

o The SInfo line of submitted fingerprints now provides the target's
  OUI (first three bytes of the MAC address) if available.  Example:
  "M=00A0CC".  To save a couple bytes, the "Time" field in SInfo has
  been renamed to "Tm".  The OUI helps identify the device vendor, and
  is only available when the source and target machines are on the
  same ethernet network.

Nmap 3.50

o Integrated a ton of service fingerprints, increasing the number of
  signatures more than 50%.  It has now exceeded 1,000 for the first
  time, and represents 180 unique service protocols from acap, afp,
  and aim to xml-rpc, zebedee, and zebra.

o Implemented a huge OS fingerprint update.  The number of
  fingerprints has increased more than 13% to 1,121.  This is the first
  time it has exceeded 1000.  Notable updates include Linux 2.6.0, Mac
  OS X up to 10.3.2 (Panther), OpenBSD 3.4 (normal and pf "scrub all"),
  FreeBSD 5.2, the latest Windows Longhorn warez, and Cisco PIX 6.3.3.
  As usual, there are a ton of new consumer devices from ubiquitous
  D-Link, Linksys, and Netgear broadband routers to a number of new IP
  phones including the Cisco devices commonly used by Vonage.  Linksys
  has apparently gone special-purpose with some of their devices, such
  as their WGA54G "Wireless Game Adapter" and WPS54GU2 wireless print
  server.  A cute little MP3 player called the Rio Karma was submitted
  multiple times and I also received and integrated fingerprints for the
  Handspring Treo 600 (PalmOS).

o Applied some man page fixes from Eric S. Raymond
  (esr(a)snark.thyrsus.com).

o Added version scan information to grepable output between the last
  two '/' delimiters (that space was previously unused).  So the format
  is now "portnum/state/protocol/owner/servicename/rpcinfo/versioninfo"
  as in "53/open/tcp//domain//ISC Bind 9.2.1/" and
  "22/open/tcp//ssh//OpenSSH 3.5p1 (protocol 1.99)/".  Thanks to
  MadHat (madhat(a)unspecific.com) for sending a patch (although I did
  it differently).  Note that any '/' characters in the
  version (or owner) field are replaced with '|' to keep awk/cut
  parsing simple.  The service name field has been updated so that it
  is the same as in normal output (except for the same sort of
  escaping discussed above).

o Integrated an Oracle TNS service probe and match lines contributed
  by Frank Berger (fm.berger(a)gmx.de).  New probe contributions are
  always appreciated!

o Fixed a crash that could happen during SSL version detection due to
  SSL session ID cache reference counting issues.

o Applied patch from Rob Foehl (rwf(a)loonybin.net) which fixes the
  --with_openssl=DIR configure argument.

o Applied patch to nmap XML dtd (nmap.dtd) from Mario Manno
  (mm(a)koeln.ccc.de).  This accounts for the new version scanning
  functionality.

o Updated the Windows build system so that you don't have to manually
  copy nmap-service-probes to the output directory.  I also updated
  the README-WIN32 to elaborate further on the build process.

o Added configure option --with-libpcre=included which causes Nmap to
  build with its included version of libpcre even if an acceptable
  version is available on the system.

o Upgraded to Autoconf 2.59 (from 2.57).  This should help HP-UX
  compilation problems reported by Petter Reinholdtsen
  (pere(a)hungry.com) and may have other benefits as well.

o Applied patch from Przemek Galczewski (sako(a)avet.com.pl) which
  adds spaces to the XML output in places that apparently help certain
  older XML parsers.

o Made Ident-scan (-I) limits on the length and type of responses
  stricter so that rogue servers can't flood your screen with 1024
  characters.  The new length limit is 32.  Thanks to Tom Rune Flo
  (tom(a)x86.no) for the suggestion and a patch.

o Fingerprints for unrecognized services can now be a bit longer to
  avoid truncating as much useful response information.  While the
  fingerprints can be longer now, I hope they will be less frequent
  because of all the newly recognized services in this version.

o The nmap-service-probes "match" directive can now take a service
  name like "ssl/vmware-auth".  The service will then be reported as
  vmware-auth (or whatever follows "ssl/") tunneled by SSL, yet Nmap
  won't actually bother initiating an SSL connection.  This is useful
  for SSL services which can be fully recognized without the overhead
  of making an SSL connection.

o Version scan now chops commas and whitespace from the end of
  vendorproductname, version, and info fields.  This makes it easier to
  write templates incorporating lists.  For example, the tcpmux service
  (TCP port 1) gives a list of supported services separated by CRLF.
  Nmap uses this new feature to print them comma separated without
  having an annoying trailing comma as so (linewrapped):
  match tcpmux m|^(sgi_[-.\w]+\r\n([-.\w]+\r\n)*)$|
        v/SGI IRIX tcpmux//Available services: $SUBST(1, "\r\n", ",")/

Nmap 3.48

o Integrated an enormous number of version detection service
  submissions.  The database has almost doubled in size to 663
  signatures representing the following 130 services:
    3dm-http afp apcnisd arkstats bittorent chargen citrix-ica
    cvspserver cvsup dantzretrospect daytime dict directconnect domain
    echo eggdrop exec finger flexlm font-service ftp ftp-proxy gnats
    gnutella-http hddtemp hp-gsg http http-proxy hylafax icecast ident
    imap imaps imsp ipp irc ircbot irc-proxy issrealsecure jabber
    kazaa-http kerberos-sec landesk-rc ldap linuxconf lmtp lotusnotes
    lpd lucent-fwadm meetingmaker melange microsoft-ds microsoft-rdp
    mldonkey msactivesync msdtc msrpc ms-sql-m mstask mud mysql
    napster ncacn_http ncp netbios-ns netbios-ssn netrek netsaint
    netstat netwareip networkaudio nntp nsclient nsunicast ntop-http
    omniback oracle-mts oracle-tns pcanywheredata pksd pmud pop2 pop3
    pop3s poppass postgresql powerchute printer qotd redcarpet
    rendezvous rlogind rpc rsync rtsp sdmsvc sftp shell shivahose
    sieve slimp3 smtp smux snpp sourceoffice spamd ssc-agent ssh ssl
    svrloc symantec-av symantec-esm systat telnet time tinyfw upnp
    uucp veritasnetbackup vnc vnc-http vtun webster whois wins
    winshell wms X11 xfce zebra

o Added the ability to execute "helper functions" in version
  templates, to help clean up/manipulate data captured from a server
  response.  The first defined function is P() which includes only
  printable characters in a captured string.  The main impetus for
  this is to deal with Unicode strings like
  "W\0O\0R\0K\0G\0R\0O\0U\0P\0" that many MS protocols send.  Nmap can
  now decode that into "WORKGROUP".

o Added SUBST() helper function, which replaces strings in matched
  appname/version/extrainfo strings with something else.  For example,
  VanDyke Vshell gives a banner that includes
  "SSH-2\.0-VShell_2_2_0_528".  A substring match is used to pick out
  the string "2_2_0_528", and then SUB21ST(1,"_",".") is called on that
  match to form the version number 2.2.0.528.

o If responses to a probe fail to match any of the registered match
  strings for that probe, Nmap will now try against the registered "null
  probe" match strings.  This helps in the case that the NULL probe
  initially times out (perhaps because of initial DNS lookup) but the
  banner appears in later responses.

o Applied some portability fixes (particularly for OpenBSD) from Chad
  Loder (cloder(a)loder.us), who is also now the OpenBSD Nmap port
  maintainer.

o Applied some portability fixes from Marius Strobl
  (marius(a)alchemy.franken.de).

o The tarball distribution of Nmap now strips the binary at install
  time thanks to a patch from Marius Strobl
  (marius(a)alchemy.franken.de).

o Fixed a problem related to building Nmap on systems that lack PCRE
  libs (and thus have to use the ones included by Nmap).  Thanks to Remi
  Denis-Courmont (deniscr6(a)cti.ecp.fr) for the report and patch.

o Alphabetized the service names in each Probe section in
  nmap-service-probes (makes them easier to find and add to).

o Fixed the problem several people reported where Nmap would quit with
  a "broken pipe" error during service scanning.  Thanks to Jari Ruusu
  (jari.ruusu(a)pp.inet.fi) for sending a patch.  The actual error
  message was "Unexpected error in NSE_TYPE_READ callback.  Error
  code: 32 (Broken pipe)"

o Fixed protocol scan (-sO), which I had broken when adding the new
  output table format.  It would complain "NmapOutputTable.cc:128:
  failed assertion `row < numRows'".  Thanks to Matt Burnett
  (marukka(a)mac.com) for notifying me of the problem.

o Upgraded Libpcap to the latest tcpdump.org version (0.7.2) from
  0.7.1

o Applied a patch from Peter Marschall (peter(a)adpm.de) which adds
  version detection support to nmapfe.

o Fixed a problem with XML output being invalid when service detection
  was done on SSL-tunneled ports.  Thanks to the several people who
  reported this - it means that folks are actually using the XML
  output :).

o Fixed (I hope) some Solaris Sun ONE compiler compilation problems
  reported (w/patches) by Mikael Mannstrom (candyman(a)penti.org)

o Fixed the --with-openssl configure option for people who have
  OpenSSL installed in a path not automatically found by their
  compilers.  Thanks to  Marius Strobl (marius(a)alchemy.franken.de) for
  the patch.

o Made some portability changes for HP-UX and possibly other types of
  machines, thanks to a patch from Petter Reinholdtsen (pere(a)hungry.com)

o Applied a patch from Matt Selsky (selsky(a)columbia.edu) which fixes
  compilation on some Solaris boxes, and maybe others.  The error said
  "cannot compute sizeof (char)"

o Applied some patches from the NetBSD ports tree that Hubert Feyrer
  (hubert.feyrer(a)informatik.fh-regensburg.de) sent me.  The NetBSD
  Nmap ports page is at http://www.NetBSD.org/packages/net/nmap/ .

o Applied some Makefile patches from the FreeBSD ports tree that I
  found at http://www.freebsd.org/cgi/cvsweb.cgi/ports/security/nmap/files/

Nmap 3.45

o Integrated more service signatures from MadHat
  (madhat(a)unspecific.com), Brian Hatch (bri(a)ifokr.org), Niels
  Heinen (zillion(a)safemode.org), Solar Designer
  (solar(a)openwall.com), Seth Master
  (smaster(a)stanford.edu), and Curt Wilson
  (netw3_security(a)hushmail.com).  We now have 378 signatures
  recognizing 86 unique service protocols.

o Added new HTTPOptions and RTSPRequest probes suggested by MadHat
  (madhat(a)unspecific.com)

o Changed the .spec file to compile Nmap RPMs without SSL support to
  improve compatibility (Some users might not have OpenSSL, and even
  those who do might not have the right version (libopenssl.so.2 vs
  libopenssl.so.4, etc).

o Applied a patch from Solar Eclipse (solareclipse(a)phreedom.org)
  which increases the allowed size of the 'extrainfo' version field from
  80 characters to 128.  The main benefit is to allow longer apache module
  version strings.

o Fixed Windows compilation and improved the Windows port slightly (no
  more macro to redefine read().

o Applied some updates to README-WIN32 sent in by Kirby Kuehl
  (kkuehl(a)cisco.com).  He improved the list of suggested registry
  changes and also fixed a typo or two.  He also attached a .reg file
  automate the Nmap connect() scan performance enhancing registry
  changes.  I am now including that with the Nmap Windows binary .zip
  distribution (and in mswin32/ of the source distro).

o Applied a one-line patch from Dmitry V. Levin (ldv(a)altlinux.org)
  which fixes a test Nmap does during compilation to see if an existing
  libpcap installation is recent enough.

Nmap 3.40PVT17

o Wrote and posted a new paper on version scanning to
  http://www.insecure.org/nmap/versionscan.html .  Updated
  nmap-service-probes and the Nmap man page to simply refer to this
  URL.

o Integrated more service signatures from my own scanning as well as
  contributions from Brian Hatch (bri(a)ifokr.org), MadHat
  (madhat(a)unspecific.com), Max Vision (vision(a)whitehats.com), HD
  Moore (hdm(a)digitaloffense.net), Seth Master
  (smaster(a)stanford.edu), and Niels Heinen (zillion(a)safemode.org).
  MadHat also contributed a new probe for Windows Media Service.  Many
  people set a LOT of signatures, which has allowed
  nmap-service-probes to grow from 295 to 356 signatures representing
  85 service protocols!

o Applied a patch (with slight changes) from Brian Hatch
  (bri(a)ifokr.org) which enables caching of SSL sessions so that
  negotiation doesn't have to be repeated when Nmap reconnects to the same
  between probes.

o Applied a patch from Brian Hatch (bri(a)ifokr.org) which optimizes the
  requested SSL ciphers for speed rather than security.  The list was
  based on empirical evidence from substantial benchmarking he did with
  tests that resemble nmap-service-scanning.

o Updated the Nmap man page to discuss the new version scanning
  options (-sV, -A).

o I now include nmap-version/aclocal.m4 in the distribution as this is
  required to rebuild the configure script ( thanks to Dmitry V. Levin
  (ldv(a)altlinux.org) for notifying me of the problem.

o Applied a patch from Dmitry V. Levin (ldv(a)altlinux.org) which
  detects whether the PCRE include file is <pcre.h> or <pcre

o Applied a patch from Dmitry V. Levin (ldv(a)altlinux.org) which
  fixes typos in some error messages.  The patch apparently came from
  the highly-secure and stable Owl and Alt Linux distributions.  Check
  them out at http://www.openwall.com/Owl/ and
  http://www.altlinux.com/

o Fixed compilation on Mac OS X - thanks to Brian Hatch
  (bri(a)ifokr.org> and Ryan Lowe (rlowe(a)pablowe.net) for giving me
  access to Mac OS X boxes.

o Stripped down libpcre build system to remove libtool dependency and
  other cruft that Nmap doesn't need. (this was mostly a response to
  libtool-related issues on Mac OS X).

o Added a new --version_trace option which causes Nmap to print out extensive
  debugging info about what version scanning is doing (this is a subset
  of what you would get with --packet_trace).  You should usually use
  this in combination with at least one -d option.

o Fixed a port number printing bug that would cause Nmap service
  fingerprints to give a negative port number when the actual port was
  above 32K.  Thanks to Seth Master (smaster(a)stanford.edu) for finding
  this.

o Updated all the header text again to clarify our interpretation of
  "derived works" after some suggestions from Brian Hatch
  (bri(a)ifokr.org)

o Updated the Nsock config.sub/config.guess to the same newer versions
  that Nmap uses (for Mac OS X compilation).

Nmap 3.40PVT16

o Fixed a compilation problem on systems w/o OpenSSL that was
  discovered by Solar Designer.  I also fixed some compilation
  problems on non-IPv6 systems.  It now compiles and runs on my
  Solaris and ancient OpenBSD systems.

o Integrated more services thanks to submissions from Niels Heinen
  (zillion(a)safemode.org).

o Canonicalized the headers at the top of each Nmap/Nsock header source
  file.  This included clarifying our interpretation of derived works,
  updating the copyright date to 2003, making the header a bit wider,
  and a few other light changes.  I've been putting this off for a
  while, because it required editing about a hundred !#$# files!

Nmap 3.40PVT15

o Fixed a major bug in the Nsock time caching system.  This could
  cause service detection to inexplicably fail against certain ports in
  the second or later machines scanned.  Thanks to Solar Designer and HD
  Moore for helping me track this down.

o Fixed some *BSD compilation bugs found by
  Zillion (zillion(a)safemode.org).

o Integrated more services thanks to submissions from Fyodor Yarochkin
  (fygrave(a)tigerteam.net), and Niels Heinen
  (zillion(a)safemode.org), and some of my own exploring.  There are
  now 295 signatures.

o Fixed a compilation bug found by Solar Designer on machines that
  don't have struct sockaddr_storage.  Nsock now just uses "struct
  sockaddr *" like connect() does.

o Fixed a bug found by Solar Designer which would cause the Nmap
  portscan table to be truncated in -oN output files if the results are
  very long.

o Changed a bunch of large stack arrays (e.g. int portlookup[65536])
  into dynamically allocated heap pointers.  The large stack variables
  apparently caused problems on some architectures.  This issue was
  reported by osamah abuoun (osamah_abuoun(a)hotmail.com).

Nmap 3.40PVT14

o Added IPv6 support for service scan.

o Added an 'sslports' directive to nmap-service-probes.  This tells
  Nmap which service checks to try first for SSL-wrapped ports.  The
  syntax is the same as the normal 'ports' directive for non-ssl ports.
  For example, the HTTP probe has an 'sslports 443' line and
  SMTP-detecting probes have and 'sslports 465' line.

o Integrated more services thanks to submissions from MadHat
  (madhat(a)unspecific.com), Solar Designer (solar(a)openwall.com), Dug
  Song (dugsong(a)monkey.org), pope(a)undersec.com, and Brian Hatch
  (bri(a)ifokr.org).  There are now 288 signatures, matching these 65
  service protocols:
    chargen cvspserver daytime domain echo exec finger font-service
    ftp ftp-proxy http http-proxy hylafax ident ident imap imaps ipp
    ircbot ircd irc-proxy issrealsecure landesk-rc ldap meetingmaker
    microsoft-ds msrpc mud mysql ncacn_http ncp netbios-ns netbios-ssn
    netsaint netwareip nntp nsclient oracle-tns pcanywheredata pop3
    pop3s postgres printer qotd redcarpet rlogind rpc rsync rtsp shell
    smtp snpp spamd ssc-agent ssh ssl telnet time upnp uucp vnc
    vnc-http webster whois winshell X11

o Added a Lotus Notes probe from Fyodor Yarochkin
  (fygrave(a)tigerteam.net).

o Dug Song wins the "award" for most obscure service fingerprint
  submission.  Nmap now detects Dave Curry's Webster dictionary server
  from 1986 :).

o Service fingerprints now include a 'T=SSL' attribute when SSL
  tunneling was used.

o More portability enhancements thanks to Solar Designer and his Linux
  2.0 libc5 boxes.

o Applied a patch from Gisle Vanem (giva(a)bgnett.no) which improves
  Windows emulation of the UNIX mmap() and munmap() memory mapping calls.

Nmap 3.40PVT13

o Added SSL-scan-through support.  If service detection finds a port to be
  SSL, it will transparently connect to the port using OpenSSL and use
  version detection to determine what service lies beneath.  This
  feature is only enabled if OpenSSL is available at build time.  A
  new --with-openssl=DIR configure option is available if OpenSSL is
  not in your default compiler paths.  You can use --without-openssl
  to disable this functionality.  Thanks to Brian Hatch
  (bri(a)ifokr.org) for sample code and other assistance.  Make sure
  you use a version without known exploitable overflows.  In
  particular, versions up to and including OpenSSL 0.9.6d and
  0.9.7-beta2 contained serious vulnerabilities described at
  http://www.openssl.org/news/secadv_20020730.txt .  Note that these
  vulnerabilities are well over a year old at the time of this
  writing.

o Integrated many more services thanks to submissions from Brian
  Hatch, HellNBack ( hellnbak(a)nmrc.org ), MadHat, Solar Designer,
  Simple Nomad, and Shawn Wallis (swallis(a)ku.edu).  The number of
  signatures has grown from 242 to 271.  Thanks!

o Integrated Novell Netware NCP and MS Terminal Server probes from
  Simple Nomad (thegnome(a)nmrc.org).

o Fixed a segfault found by Solar Designer that could occur when
  scanning certain "evil" services.

o Fixed a problem reported by Solar Designer and MadHat (
  madhat(a)unspecific.com ) where Nmap would bail when certain Apache
  version/info responses were particularly long.  It could happen in
  other cases as well.  Now Nmap just prints a warning.

o Fixed some portability issues reported by Solar Designer
  ( solar(a)openwall.com )

Nmap 3.40PVT12

o I added probes for SSL (session startup request) and microsoft-ds
  (SMB Negotiate Protocol request).

o I changed the default read timeout for a service probe from 7.5s to 5s.

o Fixed a one-character bug that broke many scans when -sV was NOT
  given.  Thanks to Blue Boar (BlueBoar(a)thievco.com) for the report.

Nmap 3.40PVT11

o Integrated many more services thanks to submissions from Simple
  Nomad, Solar Designer, jerickson(a)inphonic.com, Curt Wilson, and
  Marco Ivaldi.  Thanks!  The match line count has risen from 201 to 242.

o Implemented a service classification scheme to separate the
  vendor/product name from the version number and any extra info that
  is provided.  Instead of v/[big version string]/, the new match
  lines include v/[vendor/productname]/[version]/[extrainfo]/ .  See
  the docs at the top of nmap-service-probes for more info.  This
  doesn't change the normal output (which lumps them together anyway),
  but they are separate in the XML so that higher-level programs can
  easily match against just a product name.  Here are a few examples
  of the improved service element:
  <service name="ssh" product="OpenSSH" version="3.1p1"
     extrainfo="protocol 1.99" method="probed" conf="10" />
  <service name="domain" product="ISC Bind" version="9.2.1"
     method="probed" conf="10" />
  <state state="open" /><service name="rpcbind" version="2"
     extrainfo="rpc #100000" method="probed" conf="10" />
  <service name="rndc" method="table" conf="3" />

o I went through nmap-service-probes and added the vendor name to more
  entries.  I also added the service name where the product name
  itself didn't make that completely obvious.

o SCO Corporation of Lindon, Utah (formerly Caldera) has lately taken
  to an extortion campaign of demanding license fees from Linux users
  for code that they themselves knowingly distributed under the terms
  of the GNU GPL.  They have also refused to accept the GPL, claiming
  that some preposterous theory of theirs makes it invalid.  Meanwhile
  they have distributed GPL-licensed Nmap in (at least) their
  "Supplemental Open Source CD".  In response to these blatant
  violations, and in accordance with section 4 of the GPL, we hereby
  terminate SCO's rights to redistribute any versions of Nmap in any
  of their products, including (without limitation) OpenLinux,
  Skunkware, OpenServer, and UNIXWare.

Nmap 3.40PVT10

o Added "soft matches".  These are similar to normal match lines in
  that they provide a regex for recognizing a service (but no version).
  But instead of stopping at softmatch service recognition, the scan
  continues looking for more info.  It only launches probes that are
  known-capable of matching the softmatched service.  If no version
  number is found, at least the determined service is printed.  A
  service print for submission is also provided in that case.  So this
  provides more informative results and improves efficiency.

o Cleaned up the Windows support a bit and did more testing and
  fixing.  Windows service detection seems to be working fine for me
  now, although my testing is still pretty limited.  This release
  includes a Windows binary distribution and the README-WIN32 has been
  updated to reflect new compilation instructions.

o More service fingerprints!  Thanks to Solar Designer, Max Vision,
  Frank Denis (Jedi/Sector One) for the submissions.  I also added a
  bunch from my own testing. The number of match lines went from 179
  to 201.

o Updated XML output to handle new version and service detection
  information.  Here are a few examples of the new output:
  <port protocol="tcp" portid="22"><state state="open" /><service
    name="ssh" version="OpenSSH 3.1p1 (protocol 1.99)" method="probed"
    conf="10" /></port>
  <port protocol="tcp" portid="111"><state state="open" /><service
    name="rpcbind" version="2 (rpc #100000)" method="probed" conf="10" /></port>
  <port protocol="tcp" portid="953"><state state="open" /><service
    name="rndc" method="table" conf="3" /></port>

o Fixed issue where Nmap would quit when ECONNREFUSED was returned
  when we try to read from an already-connected TCP socket.  FreeBSD
  does this for some reason instead of giving ECONNRESET.  Thanks to
  Will Saxon (WillS(a)housing.ufl.edu) for the report.

o Removed the SERVICEMATCH_STATIC match type from
  nmap-service-probes.  There wasn't much benefit of this over regular
  expressions, so it isn't worth maintaining the extra code.

Nmap 3.40PVT9

o Added/fixed numerous service fingerprints thanks to submissions from
  Max Vision, MadHat, Seth Master.  Match lines went
  from 164 to 179.

o The Winpcap libraries used in the Windows build process have been
  upgraded to version 3.0.

o Most of the Windows port is complete.  It compiles and service scan
  works (I didn't test very deeply) on my WinXP box with VS.Net 2003.
  I try to work out remaining kinks and do some cleanup for the next
  version.  The Windows code was restructured and improved quite a bit,
  but much more work remains to be done in that area.  I'll probably
  do a Windows binary .zip release of the next version.

o Various minor fixes

Nmap 3.40PVT8

o Service scan is now OFF by default.  You can activate it with -sV.
  Or use the snazzy new -A (for "All recommended features" or
  "Aggressive") option which turns on both OS detection and service
  detection.

o Fixed compilation on my ancient OpenBSD 2.3 machine (a Pentium 60 :)

o Added/fixed numerous service fingerprints thanks to submissions from
  Brian Hatch, HD Moore, Anand R., and some of my own testing.  The
  number of match lines in this version grows from 137 to 164!  Please
  keep 'em coming!

o Various important and not-so-important fixes for bugs I encountered
  while test scanning.

o The RPC grinder no longer prints a startup message if it has no
  RPC-detected ports to scan.

o Some of the service fingerprint length limitations are relaxed a bit
  if you enable debugging (-d).

Nmap 3.40PVT7

o Added a whole bunch of services submitted by Brian Hatch
  (bri(a)ifokr.org).  I also added a few Windows-related probes.
  Nmap-service-probes has gone from 101 match strings to 137.  Please
  keep the submissions coming.

o The question mark now only appears for ports in the OPEN state and
  when service detection was requested.

o I now print a separator bar between service fingerprints when Nmap
  prints more than one for a given host so that users understand to
  submit them individually (suggested by Brian Hatch (bri(a)ifokr.org))

o Fixed a bug that would cause Nmap to print "empty" service
  fingerprints consisting of just a semi-colon.  Thanks to Brian Hatch
  (bri(a)ifokr.org) for reporting this.

Nmap 3.40PVT6

o Banner-scanned hundreds of thousands of machines for ports
  21,23,25,110,3306 to collect default banners.  Where the banner made
  the service name/version obvious, I integrated them into
  nmap-service-probes.  This increased the number of 'match' lines from
  27 to more than 100.

o Created the service fingerprint submission page at
  http://www.insecure.org/cgi-bin/servicefp-submit.cgi

o Changed the service fingerprint format slightly for easier
  processing by scripts.

o Applied a large portability patch from Albert Chin-A-Young
  (china(a)thewrittenword.com).  This cleans up a number of things,
  particularly for IRIX, Tru64, and Solaris.

o Applied NmapFE patch from Peter Marschall (peter(a)adpm.de) which
  "makes sure changes in the relay host and scanned port entry fields
  are displayed immediately, and also keeps the fields editable after
  de- and reactivating them."

Nmap 3.40PVT4

o Limited the size of service fingerprints to roughly 1024 bytes.
  This was suggested by Niels Heinen (niels(a)heinen.ws), because the previous
  limit was excessive.  The number of fingerprints printed is also now
  limited to 10.

o Fixed a segmentation fault that could occur when ping-scanning large
  networks.

o Fixed service scan to gracefully handle host_timeout occurrences when
  they happen during a service scan.

o Fixed a service_scan bug that would cause an error when hosts send
  data and then close() during the NULL probe (when we haven't sent
  anything).

o Applied a patch from Solar Designer (solar(a)openwall.com) which
  corrects some errors in the Russian man page translation and also a
  couple typos in the regular man page.  Then I spell-checked the man
  page to reduce future instances of foreigners sending in diffs to
  correct my English :).

Nmap 3.40PVT3

o Nmap now prints a "service fingerprint" for services that it is
  unable to match despite returning data.  The web submission page it
  references is not yet available.

o Service detection now does RPC grinding on ports it detects to be
  running RPC.

o Fixed a bug that would cause Nmap to quit with an Nsock error when
  --host_timeout was used (or when -T5 was used, which sets it
  implicitly).

o Fixed a bug that would cause Nmap to fail to print the OS
  fingerprint in certain cases.  Thanks to Ste Jones
  (root(a)networkpenetration.com) for the problem report.

Nmap 3.40PVT2

o Nmap now has a simple VERSION detection scheme.  The 'match' lines in
  nmap-service-probes can specify a template version string
  (referencing subexpression matches from the regex in a Perl-like
  manner) so that the version is determined at the same time as the
  service.  This handles many common services in a highly efficient
  manner.  A more complex form of version detection (that initiates
  further communication w/the target service) may be necessary
  eventually to handle services that aren't as forthcoming with
  version details.

o The Nmap port state table now wastes less whitespace due to using a new
  and stingy NmapOutputTable class.  This makes it easier to read, and
  also leaves more room for version info and possibly other enhancements.

o Added 's' option to match lines in nmap-service-probes.  Just as
  with the Perl 's' option, this one causes '.' in the regular
  expression to match any character INCLUDING newline.

o The WinPcap header timestamp is no longer used on Windows as it
  sometimes can be a couple seconds different than gettimeofday() (which
  is really _ftime() on Windows) for some reason.  Thanks to Scott
  Egbert (scott.egbert(a)citigroup.com) for the report.

o Applied a patch by Matt Selsky (selsky(a)columbia.edu) which fixes
  configure.in in such a way that the annoying header file "present but
  cannot be compiled" warning for Solaris.

o Applied another patch from Matt that (we hope) fixes the "present
  but cannot be compiled" warning -- this time for Mac OS X.

o Port table header names are now capitalized ("SERVICE", "PORT", etc)

Nmap 3.40PVT1

o Initial implementation of service detection.  Nmap will now probe
  ports to determine what is listening, rather than guessing based on
  the nmap-services table lookup.  This can be very useful for
  services on unidentified ports and for UDP services where it is not
  always clear (without these probes) whether the port is really open
  or just firewalled.  It is also handy for when services are run on
  the well-known-port of another protocol -- this is happening more
  and more as users try to circumvent increasingly strict firewall
  policies.

o Nmap now uses the excellent libpcre (Perl Compatible Regular
  Expressions) library from http://www.pcre.org/ .  Many systems
  already have this, otherwise Nmap will use the copy it now includes.
  If your libpcre is hidden away in some nonstandard place, give
  ./configure the new --with-libpcre=DIR directive.

o Nmap now uses the C++ Standard Template Library (STL).  This makes
  programming easier, but if it causes major portability or bloat
  problems, I'll reluctantly remove it.

o Applied a patch from Javier Kohen (jkohen(a)coresecurity.com) which
  normalizes the names of many Microsoft entries in the
  nmap-os-fingerprints file.

o Applied a patch by Florin Andrei (florin(a)sgi.com) to the Nmap RPM
  spec file.  This uses the 'Epoch' flag to prevent the Redhat Network
  tool from marking my RPMs as "obsolete" and "upgrading" to earlier
  Redhat-built versions.  A compilation flag problem is also fixed.

Nmap 3.30

o Implemented the largest-ever OS fingerprint update!  Roughly 300
  fingerprints were added/modified.  These massive changes span the
  gamut from AIX 5.1 to the ZyXEL Prestige broadband router line.
  Notable updates include OpenBSD 3.3, FreeBSD 5.1, Mac OS X 10.2.6,
  Windows 2003 server, and more WAPs and broadband routers than you
  can shake a stick at.  Someone even submitted a fingerprint for
  Debian Linux running on the Microsoft Xbox.  You have to love that
  irony :).  Thanks to everyone who submitted fingerprints using the
  URL Nmap gives you when it gets a clean reading but is stumped.  The
  fingerprint DB now contains almost 1000 fingerprints.

o Went through every one of the fingerprints to normalize the
  descriptions a bit.  I also looked up what all of the devices are
  (thanks E*Bay and Google!).  Results like "Nexland ISB Pro800 Turbo"
  and "Siemens 300E Release 6.5" are much more useful when you add the
  words "cable modem" and "business phone system"

o Added a new classification system to nmap-os-fingerprints.  In
  addition to the standard text description, each entry is now
  classified by vendor name (e.g. Sun), underlying OS (e.g. Solaris),
  OS generation (e.g. 7), and device type ("general purpose", router,
  switch, game console, etc).  This can be useful if you want to (say)
  locate and eliminate the SCO systems on a network, or find the
  wireless access points (WAPs) by scanning from the wired side.

o Classification system described above is now used to print out a
  "device type" line and OS categories for matches.  The free-form
  English details are still printed as well.  Nmap can sometimes
  provide classifications even where it used to provide nothing
  because of "too many matches".  These have been added to XML output
  as well.  They are not printed for the "grepable output", as I
  consider that format deprecated.

o Nmap will now sometimes guess in the "no exact matches" case, even
  if you don't use the secret --osscan_guess or -fuzzy options.

o Applied another huge NmapFE patch from Peter Marschall
  (peter(a)adpm.de).  This revamps the interface to use a tabbed
  format that allows for many more Nmap options to be used.  It also
  cleans up some crufty parts of the code.  Let me and Peter know what
  you think (and if you encounter any problems).

o Windows and Amiga ports now use packet receive times from libpcap.
  Let me know if you get any "time computation problem" errors.

o Updated version of the Russian man page translation from Alex Volkov
  (alex(a)cherepovets-city.ru).

Nmap 3.28

o Fixed (I hope) an issue that would cause Nmap to print "Serious time
  computation problem in adjust_timeout ..." and quit.  The ultimate
  cause was demonstrated by this --packet_trace snippet that Russel
  Miller (rmiller(a)duskglow.com) sent me:
  SENT (0.0500s) ICMP 0.0.0.0 > 127.0.0.1 Echo request (type=8/code=0) ...
  RCVD (0.0450s) ICMP 127.0.0.1 > 127.0.0.1 Echo reply (type=0/code=0) ...
  As you can see, the ping reply appears to come BEFORE the request
  was sent(!).  This sort of thing happens on at least Linux and
  Windows.  The send time is obtained from gettimeofday(timeval, NULL),
  while receive time libpcap packet header.  If anyone knows why this
  occurs, or (even better) knows a good way to fix it, let me know.
  For now, I am allowing the response to come up to .05s "before" the
  request.  That is gross.

o For years, Nmap has added -I/usr/local/include and -L/usr/local/lib
  to the compiler line to grab local libraries.  I have removed this
  behavior by default, and added a '--with-localdirs' configure option
  that adds it back.  If Nmap fails to compile now without the above
  option, please let me know.  I can change the default back if this
  change causes more problems than it solves.  People (such as certain
  ports tree packagers) who know they don't want /usr/local should
  specify --without-localdirs rather than relying on that always being
  the default.

o Fixed (I hope) a problem that led to the error message "Assertion
  `tqi->sockets[probe_port_num][seq] == -1' failed".

o Fixed a problem that would cause Nmap on Windows to send ICMP ping
  packets from 0.0.0.0 instead of the appropriate source IP.  Thanks
  to Yeti (boxed(a)blueyonder.co.uk) for the report.

o Applied some changes from Solar Designer (solar(a)openwall.com)
  which fix some typos and also suggest safer /tmp/ behavior in the
  HACKING file and Lithuanian man page.  These changes are for the
  Nmap package of his Openwall GNU/*/Linux (Owl) distribution.
  [ http://www.openwall.com/Owl/ ]

o For Solaris, I now define NET_SIZE_T to size_t rather than socklen_t
  in nmap.h.  Isn't that exciting?!!!  Hopefully this will help
  compilation on Solaris 2.6 (and perhaps earlier).  If any Solaris
  users notice new compilation problems, please let me know.  Thanks to
  Al Smith (Al.Smith(a)aeschi.ch.eu.org) for reporting the issue.

o Removed an errant getopt() prototype in nbase/getopt.h which should
  hopefully improve compilation on certain Solaris boxes and BSD
  variants.

o SCO operating systems are no longer supported due to their recent
  (and absurd) attacks against Linux and IBM.  Bug reports relating to
  UnixWare will be ignored, or possibly even laughed at derisively.
  Note that I have no reason to believe anyone has ever used Nmap on
  SCO systems.  UnixWare and OpenServer suck.

o Fixed a problem with small --max_parallelism values when non-root ping
  scanning that would cause Nmap to say "sendconnecttcpquery: Could
  not scavenge a free socket!" and quit.  Problem was reported by
  Justin A (justin(a)bouncybouncy.net) as Debian Bug #195463.

o Applied (with a few modifications) a large NmapFE patch from Peter
  Marschall (peter(a)adpm.de).  This patch adds a bunch more scan/ping
  options and cleans up some redundant NmapFE code.

o Included new Russian man page translation by Alex Volkov
  (alex(a)cherepovets-city.ru)

o Changed many single-quotes (') into double quotes (") in the man
  page due to a disagreement over whether to represent them as (') or
  (\') in nroff.

o Included --packet_trace support for Explicit Congestion Notification
  (RFC 2481/3168) flags thanks to a patch sent in by Maik Pfeil
  (root(a)bundesspionageministerium.de)

o Included --packet_trace support for a few (unusual) ICMP types in
  case Nmap receives them.  The patch was also sent by Maik Pfeil.

o Fixed a problem with redirecting XML/Grep/Machine output to stdout
  on Windows (e.g. -oX - ).  Problem was reported by Wei Jiang
  (Wei.Jiang(a)bindview.com)

o Made "-g -Wall" compiler flags dependent on availability of gcc/g++
  sine some other compilers do not support them.

o I spam-protected the email addresses in this file.  I fervently hope
  that within 5 years we will be able to defeat this scourge through
  technology and laws, so that we may again list our email addresses
  openly without fear of abuse by criminal spammers.  Oh, and it would
  be a shame if the spiders went through this whole page and only
  found [e-mail], [e-mail], [e-mail], [e-mail],
  [e-mail], [e-mail], and [e-mail].

Nmap 3.27

o Nmap now compiles under Amiga thanks to patches sent by Diego
  Casorran (dcr8520(a)amiga.org).

o Fixed a backwards WIN32 ifdef that broke UDP and small-fragment
  scans for some operating systems other than Linux and Windows.
  Thanks to Guido van Rooij (guido(a)gvr.org) for reporting the problem
  and sending a patch.

o Applied patch from Marius Strobl (marius(a)alchemy.franken.de) which improves
  the definition of NET_SIZE_T on FreeBSD so that it compiles on
  64-bit platforms.

Nmap 3.26

o Fixed Mac OS X Compilation (at least on most of the machines
  tested).  You will probably need to type
  "./configure CPP=/usr/bin/cpp" instead of simply "./configure".  If
  you still have trouble, drop me an email.  Thanks to everyone who
  provided or offered shell accounts!

o Fixed a segmentation fault several people reported that was
  introduced in 3.25.  This problem manifests itself intermittently
  in many normal situations involving large-network scanning.  So all
  3.25 users are urged to upgrade.  Pre-3.25 users should upgrade too,
  since 3.25 included so many improvements :).

Nmap 3.25

o I added UDP-based "ping" scanning.  The -PU option can take an
  optional portlist like the TCP "ping" options (-PS, -PA), but it sends
  a UDP packet to the targets and expects hosts that are up to reply
  with a port unreachable (or possibly a UDP response if the port is
  open).  This one is likely to work best against closed ports, since
  many open ports don't respond to empty requests.

o Fixed (I hope) problem where Nmap would abort, complaining that
  "Assertion `pt->down_this_block > 0' failed".  Thanks to
  ray(a)24hoursecurity.org and mugz(a)x-mafia.com for reporting and
  helping me debug this problem.

o Fixed a GCC dependency reported by Ayamura Kikuchi
  (ayamura(a)keio.net)

o Fixed an "assertion failure" which would cause Nmap to exit when you
  specify a --max_rtt_timeout below 3000.  Thanks to Tammy Rathbun
  (rathbun2(a)llnl.gov) and Jan Roger Wilkens (jrw(a)proseq.net) for
  reporting this.

o Packet receive times are now obtained from libpcap rather than
  simply using the time the packets are passed to Nmap.  This should
  improve performance slightly.  I was not able to get this to work
  properly on Windows (either pcap or raw) -- join the nmap-dev list
  if you have ideas.

o Fixed bug that caused Nmap to ignore certain RST responses when you
  do both -PS and -PA.

o Modified ping scan to work better when many instances of Nmap are
  executed concurrently.

o I'm now linking directly to the gzip compressed version of Nmap on
  the homepage as well as the .bz2.

o Fixed a portability problem that caused BSD Make to bail out.

o Fixed a divide by zero error caused when non-root users (on UNIX)
  explicitly request ICMP pings (which require root privileges).  Now it
  prints a warning and uses the normal non-root TCP connect() ping.
  Jaroslav Sladek (jup(a)matfyz.cz) found the bug and provided the patch.

o Made Nmap more tolerant of corrupt nmap-services and nmap-protocols
  files thanks to report & patch sent by Phix (phix(a)hush.com)

o Added some more port numbers sent in by Seth Master
  (smaster(a)stanford.edu).  He has been a frequent nmap-services
  contributor in the last couple months.

o Added --packet_trace support to Windows

o Removed superfluous "addport" line in the XML output (patch from Max
  Schubert (nmap(a)webwizarddesign.com)).

o Merged wintcpip.cc into tcpip.cc to avoid the headache of
  maintaining many nearly-identical functions.

o Fixed an assertion failure crash related to combining port 0 scans
  and OS scan.  Thanks to A.Jones(a)mvv.de for reporting this.

o Fixed some compilation problems on systems without IPv6 support --
  patch sent by Jochen Erwied (Jochen.Erwied(a)mbs-software.info)

o Applied patch from Jochen Erwied (Jochen.Erwied(a)mbs-software.info)
  which fixes the format strings used for printing certain timestamps.

o Upgraded to autoconf 2.57, including the latest config.guess/config.sub

o Renamed configure.ac files to configure.in as recommended by the
  latest autoconf documentation.

o Changed the wording of NmapFE Gnome entries to better-comply with
  Gnome's Human Interface Guidelines (HIG).  Suggested by Axel Krauth
  (krauth(a)fmi.uni-passau.de)

Nmap 3.20

o The random IP input option (-iR) now takes an argument specifying
  how many IPs you want to scan (e.g. -iR 1000).  Specify 0 for the old
  never-ending scan behavior.

o Fixed a tricky memory leak discovered by Mugz (mugz(a)x-mafia.com).

o Fixed output truncation problem noted by Lionel CONS (lionel.cons(a)cern.ch)

o Fixed a bug that would cause certain incoming ICMP error messages to
  be improperly ignored.

Nmap 3.15BETA3

o Made numerous improvements to the timing behavior of "-T Aggressive"
  (same as -T4) scans.  It is now recommended for regular use by
  impatient people with a fast connection.  "-T Insane" mode has also
  been updated, but we only recommend that for, well, insane people.

o Made substantial changes to the SYN/connect()/Window scanning
  algorithms for improved speeds, especially against heavily filtered
  hosts.  If you notice any timing problems (misidentified ports,
  etc.), please send me the details (including full Nmap output and a
  description of what is wrong).  Reports of any timing problems with
  -T4 would be helpful as well.

o Changed Nmap such that ALL syn scan packets are sent from the port
  you specify with -g.  Retransmissions used to utilize successively
  higher ports.  This change has a downside in that some operating
  systems (such as Linux) often won't reply to the retransmissions
  because they reuse the same connection specifier quad
  (srcip:srcport:dstip:dstport).  Overall I think this is a win.

o Added timestamps to "Starting nmap" line and each host port scan in
  verbose (-v) mode.  These are in ISO 8601 standard format because
  unlike President Bush, we actually care about International
  consensus :).

o Nmap now comes by default in .tar.bz2 format, which compresses about
  20% further.  You can still find .tgz in the dist directory at
  http://download.insecure.org/nmap/dist/?M=D .

o Various other minor bug fixes, new services, fingerprints, etc.

Nmap 3.15BETA2

o I added support for a brand new "port" that many of you may have
  never scanned before!  UDP & TCP "port 0" (and IP protocol 0) are now
  permitted if you specify 0 explicitly.  An argument like "-p -40"
  would still scan ports 1-40.  Unlike ports, protocol 0 IS now scanned
  by default.  This now works for ping probes too (e.g., -PS, -PA).

o Applied patch by Martin Kluge (martin(a)elxsi.info) which adds --ttl
  option, which sets the outgoing IPv4 TTL field in packets sent via
  all raw scan types (including ping scans and OS detection).  The
  patch "should work" on Windows, but hasn't been tested.  A TTL of 0
  is supported, and even tends to work on a LAN:
    14:17:19.474293 192.168.0.42.60214 > 192.168.0.40.135: S 326:326(0) [ttl 0]
    14:17:19.474456 192.168.0.40.135 > 192.168.0.42.60214: S 280:280(0) ack 326 (ttl 128)

o Applied patch by Gabriel L. Somlo ( somlo(a)acns.colostate.edu ) which
  extends the multi-ping-port functionality to nonroot and IPv6
  connect() users.

o I added a new --datadir command line option which allows you to
  specify the highest priority directory for Nmap data files
  nmap-services, nmap-os-fingerprints, and nmap-rpc.  Any files which
  aren't in the given dir, will be searched for in the $NMAPDIR
  environmental variable, ~/nmap/, a compiled in data directory
  (e.g. /usr/share/nmap), and finally the current directory.

o Fixed Windows (VC++ 6) compilation, thanks to patches from Kevin
  Davis (computerguy(a)cfl.rr.com) and Andy Lutomirski
  (luto(a)stanford.edu)

o Included new Latvian man page translation by
  "miscelerious options" (misc(a)inbox.lv)

o Fixed Solaris compilation when Sun make is used rather than GNU
  make.  Thanks to Tom Duffy (tduffy(a)sun.com) for assistance.

o Applied patch from Stephen Bishop (sbishop(a)idsec.co.uk) which
  prevents certain false-positive responses when Nmap raw TCP ping scans
  are being run in parallel.

o To emphasize the highly professional nature of Nmap, I changed all
  instances of "fucked up" in error message text into "b0rked".

o Fixed a problem with nmap-frontend RPMs that would cause a bogus
  /bin/xnmap link to be created (it should only create
  /usr/bin/xnmap).  Thanks to Juho Schultz
  (juho.schultz(a)astro.helsinki.fi) for reporting the problem.

o I made the maximum number of allowed routes and interfaces allowed
  on the scanning machine dynamic rather than hardcoded #defines of 1024
  and 128.  You never know -- some wacko probably has that many :).

Nmap 3.15BETA1

o Integrated the largest OS fingerprint DB updates ever! Thanks to
  everyone who contributed signatures!  New or substantially modified
  fingerprints included the latest Windows 2K/XP changes, Cisco IOS
  12.2-based routers and PIX 6.3 firewalls, FreeBSD 5.0, AIX 5.1,
  OpenBSD 3.2, Tru64 5.1A, IBM OS/400 V5R1M0, dozens of wireless APs,
  VOIP devices, firewalls, printers, print servers, cable modems,
  webcams, etc.  We've even got some mod-chipped Xbox fingerprints
  now!

o Applied NetBSD portability patch by Darren Reed
  (darrenr(a)reed.wattle.id.au)

o Updated Makefile to better-detect if it can't make nmapfe and
  provide a clearer error message.  Also fixed a couple compiler
  warnings on some *BSD platforms.

o Applied patch from "Max" (nmap(a)webwizarddesign.com) which adds the
  port owner to the "addport" XML output lines which are printed (only
  in verbose mode, I think) as each open port is discovered.

o I killed the annoying whitespace that is normally appended after the
  service name.  Now it is only there when an owner was found via -sI
  (in which case there is a fourth column and so "service" must be
  exactly 24 characters).

Nmap 3.10ALPHA9

o Reworked the "ping scan" algorithm (used for any scan except -P0 or
  -sL) to be more robust in the face of low-bandwidth and congested
  connections.  This also improves reliability in the multi-port and
  multi-type ping cases described below.

o "Ping types" are no longer exclusive -- you can now do combinations
  such as "-PS22,53,80 -PT113 -PN -PE" in order to increase your odds of
  passing through strict filters.  The "PB" flag is now deprecated
  since you can achieve the same result via "PE" and "PT" options.

o Applied patch (with modest changes) by Gabriel L. Somlo
  (somlo(a)acns.colostate.edu), which allows multiple TCP probe ports in
  raw (root) mode.  See the previous item for an example.

o Fixed a libpcap compilation issue noted by Josef 'Jupp' Schugt
  (deusxmachina(a)webmail.co.za) which relates to the definition (or
  lack thereof) of ARPHRD_HDLC (used for Cisco HDLC frames).

o Tweaked the version number (-V) output slightly.

Nmap 3.10ALPHA7

o Upgraded libpcap from version 0.6.2 to 0.7.1.  Updated the
  libpcap-possiblymodified/NMAP_MODIFICATIONS file to give a much
  more extensive list (including diffs) of the changes included
  in the Nmap bundled version of Libpcap.

o Applied patch to fix a libpcap alignment bug found by Tom Duffy
  (tduffy(a)sun.com).

o Fixed Windows compilation.

o Applied patch by Chad Loder (cloder(a)loder.us) of Rapid7 which
  fixes OpenBSD compilation.  I believe Chad is now the official
  OpenBSD Nmap "port" maintainer.  His patch also adjusted
  random-scan (-iR) to include the recently allocated 82.0.0.0/8
  space.

o Fixed (I hope) a few compilation problems on
  non-IPv6-enabled machines which were noted by Josef 'Jupp'
  Schugt (jupp(a)gmx.de)

o Included some man page translations which were inadvertently
  missed in previous tarballs.

o Applied patch from Matthieu Verbert (mve(a)zurich.ibm.com) which
  places the Nmap man pages under ${prefix}/share/man rather than
  ${prefix}/man when installed via RPM.  Maybe the tarball
  install should do this too?  Opinions?

o Applied patch from R Anderson (listbox(a)pole-position.org) which
  improves the way ICMP port unreachables from intermediate hosts
  are handled during UDP scans.

o Added note to man page related to Nmap US export control.  I
  believe Nmap falls under ECCN 5D992, which has no special
  restrictions beyond the standard export denial to a handful of
  rogue nations such as Iraq and North Korea.

o Added a warning that some hosts may be skipped and/or repeated
  when someone tries to --resume a --randomize_hosts scan.  This
  was suggested by Crayden Mantelium (crayden(a)sensewave.com)

o Fixed a minor memory leak noted by Michael Davis
  (mike(a)datanerds.net).

Nmap 3.10ALPHA4

o Applied patch by Max Schubert (nmap(a)webwizarddesign.com) which adds
  an add-port XML tag whenever a new port is found open when Nmap is
  running in verbose mode.  The new tag looks like:
  [addport state="open" portid="22" protocol="tcp"/]
  I also updated docs/nmap.dtd to recognize this new tag.

o Added German translation of Nmap man page by Marc Ruef
  (marc.ruef(a)computec.ch).  It is also available at
  http://www.insecure.org/nmap/data/nmap_manpage-de.html

o Includes a brand new French translation of the man page by Sebastien
  Blanchet.  You could probably guess that it is available at
  http://www.insecure.org/nmap/data/nmap_manpage-fr.html

o Applied some patches from Chad Loder (cloder(a)loder.us) which update
  the random IP allocation pool and improve OpenBSD support.  Some
  were from the OBSD Nmap patchlist.

o Fixed a compile problem on machines without PF_INET6.  Thanks to
  Josef 'Jupp' Schugt (deusxmachina(a)webmail.co.za) for noting this.

Nmap 3.10ALPHA3

o Added --min_parallelism option, which makes scans more aggressive
  and MUCH faster in certain situations -- especially against
  firewalled hosts.  It is basically the opposite of --max_parallelism
  (-M).  Note that reliability can be lost if you push it too far.

o Added --packet_trace option, which tells Nmap to display all of the
  packets it sends and receives in a format similar to tcpdump.  I
  mostly added this for debugging purposes, but people wishing to learn
  how Nmap works or for experts wanting to ensure Nmap is doing
  exactly what they expect.  If you want this feature supported under
  Windows, please send me a patch :).

o Fixed a segmentation fault in Idlescan (-sI).

o Made Idlescan timing more conservative when -P0 is specified to
  improve accuracy.

o Fixed an infinite-loop condition that could occur during certain
  dropped-packet scenarios in an Idle scan.

o Nmap now reports execution times to millisecond precision (rather
  than rounding to the nearest second).

o Fixed an infinite loop caused by invalid port arguments.  Problem
  noted by fejed (fejed(a)uddf.net).

Nmap 3.10ALPHA2

o Fixed compilation and IPv6 support on FreeBSD (tested on
  4.6-STABLE).  Thanks to Niels Heinen (niels.heinen(a)ubizen.com) for
  suggestions.

o Made some portability changes based on suggestions by Josef 'Jupp'
  Schugt (jupp(a)gmx.de)

o Fixed compilation and IPv6 support on Solaris 9 (haven't tested
  earlier versions).

Nmap 3.10ALPHA1

o IPv6 is now supported for TCP scan (-sT), connect()-style ping
  scan (-sP), and list scan (-sL)!  Just specify the -6 option and the
  IPv6 numbers or DNS names.  Netmask notation is not currently
  supported -- I'm not sure how useful it is for IPv6, where even petty
  end users may be allocated trillions of addresses (/80).  If you
  need one of the scan types that hasn't been ported yet, give
  Sebastien Peterson's patch a try at http://nmap6.sourceforge.net/ .
  If there is demand, I may integrate more of that into Nmap.

o Major code restructuring, which included conversion to C++ -- so
  you'll need g++ or another C++ compiler.  I accidentally let a C++
  requirement slip in a while back and found that almost everyone has
  such a compiler.  Windows (VC++) users: see the README-WIN32 for new
  compilation instructions.

o Applied patch from Axel Nennker (Axel.Nennker(a)t-systems.com) which
  adds a --without-nmapfe option to the configure script.  This is
  useful if your system doesn't have the proper libraries (e.g. GTK) or
  if you think GUIs are for sissies :).

o Removed arbitrary max_parallelism (-M) limitations, as suggested by
  William McVey ( wam(a)cisco.com ).

o Added DEC OSF to the platforms that require the BSDFIX() macro due
  to taking IP length and offset fields in host rather than network byte
  order.  Suggested by Dean Bennett (deanb(a)gbtn.net)

o Fixed an debug statement C ambiguity discovered by Kronos
  (kronos(a)kronoz.cjb.net)

Nmap 3.00

o Woohoo! :)

Nmap 2.99RC2

o Fixed an important memory initialization bug which was causing
  crashes on Mac OS X (and possibly other platforms).  The problem was
  located by Pieter ten Pierick (P.tenPierick(a)chello.nl)

o Various minor bugfixes/cleanup

Nmap 2.99RC1

o Implemented the biggest OS fingerprint update since December 1999!
  More than 200 fingerprints were added/modified.  This includes
  OpenBSD 3.1, Solaris 9, Mac OS 10.1.5, OS/400, FreeBSD 4.6, The
  latest MS WinXP changes, new CISCO equiptment, and loads of network
  devices such as VoIP phones, switches, printers, WAPs, etc.

o Updated build system to work on MacOS X.

o I removed "credit" lines from the nmap-os-fingerprints file out of
  concern that evil spammers might harvest the 602 addresses.  Plus
  those took up 28K and the size of nmap-os-fingerprints has already
  caused trouble for some handheld devices.  If anyone actually cares
  about the "fame" of being listed, let me know and I'll put you back
  in.  I still appreciate everyone who submits fingerprints!  I just
  don't want you to be spammed when the fingerprint file goes online.

o Minor usage screen (nmap -h) fix suggested by Martin Kluge
  ( martin(a)elxsi.info )

o Insured that the initial pound (#) in C preprocessor directives is
  always in column 1 (portability fix).  Problem noted by Shamsher
  Sran (ssran(a)bechtel.com)

Nmap 2.54BETA37

o Made SYN scan the default for privileged (root) users.  This offers
  far better performance for Windows users due to their broken
  connect() call, and is usually even preferred on UNIX because it is
  more stealthy and less likely to crash applications listening on the
  target host.

o Fixed a problem noted by Ping Huang (pshuang(a)alum.mit.edu) relating
  to -PI scans of a machine's own non-localhost interfaces (eg
  scanning your ethernet address).

o Applied patch from Patrice Goetghebeur (pgoetghebeur(a)mac.com) which
  fixes PPP/SLIP support on Mac OS X.

o Applied dozens of nmap-services portnumber mapping updates
  researched and sent by palante(a)subterrain.net

o Updated nmap-rpc to the latest version from Eilon Gishri
  (eilon(a)aristo.tau.ac.il)

o Fixed --resume option to better detect all of the previously scanned
  hosts in an -oN file (bug report from Adam.Scott(a)predictive.com )

o Adjusted random IP generator (for -iR) to account for newly
  allocated ip space from
  http://www.iana.org/assignments/ipv4-address-space as noted by Chad
  Loder (cloder(a)acm.org)

o Updated config.sub and config.guess to the versions in
  automake-1.6.2 .

o Applied patch from Markus A. Nonym (g17m0(a)lycos.com) which checks
  for a recent version of GTK+ in ./configure before even trying to
  build NmapFE (avoids the previous ugly compiler errors).

o Applied patch from benkj(a)gmx.it which fixes misbehavior when Nmap
  would receive EOF (including ^D) in interactive mode.

o Fixed format string bugs (not the security-related kind) found by
  Takehiro YONEKURA (yonekura(a)obliguard.com) and Kuk-hyeon Lee
  (errai(a)inzen.com)

o Applied patch from Greg Steuck (greg-nmap-dev(a)nest.cx) which fixes
  an alignment problem in charpool.c that could cause bus errors on
  64-bit platforms.

o Applied portability fix patch from Matt Christian (mattc(a)visi.com)

Nmap 2.54BETA36

o Fixed major connect scan problem introduced in BETA35

o Changed NmapFE to use the version number 2.54BETA36 rather than
  0.2.54BETA36.  I had to do this because RedHat took the liberty of
  releasing a so-called "2.54BETA31" version of nmap-frontend in their
  7.3 distribution.  Thus my upgrades were failing to install on such
  systems because a "later" version is already installed.

Nmap 2.54BETA35

o Fixed an issue that could cause the abort message "Serious time
  computation problem in adjust_timeout ...".  If you still see this,
  please let me know.

o Fixed Windows compilation (and I really mean it this time -- tested
  myself).

o Applied configure script patch to recognize Solaris 2.10 when it
  eventually becomes available (from James Carlson
  (james.d.carlson(a)east.sun.com)

o Applied some portability fixes from Albert Chin
  (china(a)thewrittenword.com)

o Applied libpcap aclocal.m4 patch to enable debugging (-g) when
  compiling libpcap with gcc.  Patch from Ping Huang
  (pshuang(a)alum.mit.edu)

o Restructured "TCP probe port" output message a bit as suggested by
  Ping Huang (pshuang(a)alum.mit.edu)

Nmap 2.54BETA34

o Windows compilation fixed thanks to new VC++ project file (nmap.dsp) sent
  by Evan Sparks (gmplague(a)sdf.lonestar.org) (I had forgotten to include
  the new main.c).

o Various nmap-services updates

o Fixed a bunch of typos and capitalization issues in
  nmap-os-fingerprints by applying patch sent in by Royce Williams
  (royce(a)alaska.net).

Nmap 2.54BETA33

o Tons of OS fingerprint updates.  More than 100 fingerprints added or
  changed, including OpenBSD 3, FreeBSD 4.5, Solaris 9 pre-release,
  Commodor 64 (with the TFE Ethernet Card and uIP stack), Compaq iPAQ,
  Cisco IOS 12.2(8), AIX 5.1, IRIX 6.5.15, various
  Redback/Racal/Juniper/BigIP/HP/Siemens/Brocade/Quantum devices,
  numerous printers/switches, KRONOS network clock, WTI Network Power
  Switch, Windows XP, and many more.  Thanks to everyone who
  contributed!

o Applied fix for an important RPC scanning bug sent in by Pasi Eronen
  (pasi.eronen(a)nixu.com)

o Applied fix for nasty OS fingerprinting bug found by William
  Robertson (wkr(a)cs.ucsb.edu)

o Do not show uptime when obviously spoofed (eg OpenBSD 3.0)

o Slightly changed (I hope improved) the whitespace in Nmap output so
  that messages relating to the same host are kept together (and
  different hosts different separated by newlines).

o Moved main() function into a new file, cleverly named main.c.

Nmap 2.54BETA32

o Applied Windows pinging fix and from Andy Lutomirski
  (Luto(a)myrealbox.com)

o Applied a few more Windows fixes from Andy.

o Fixed a flaw in several error-checking statements noted by Giacomo
  Cariello (jwk(a)bug.it)

o Applied Win32 compilation fixes sent by Kirby Kuehl (kkuehl(a)cisco.com)
  and jens.vogt(a)bluewin.ch

Nmap 2.54BETA31

o Added ICMP Timestamp and Netmask ping types (-PP and -PM).  These
  (especially timestamp) can be useful against some hosts that do not
  respond to normal ping (-PI) packets.

o Documented the --data_length option and made it work with all the
  ICMP ping types (echo request, netmask, and timestamp).

o Added check for strings.h before including it in portlist.c .  This
  fixes a compilation problem on some versions of Windows.  Problem
  first noted by Michael Vorin (mvorin(a)hotmail.com)

o Applied patch from Andy Lutomirski (Luto(a)myrealbox.com) which fixes
  a crash on some Windows platforms when timeouts occur.

o Fixed "grepable output" (-oG) so that it prints IPID sequence class
  rather than printing the TCP ISN sequence index twice.  Problem
  noted by Russell Fulton (r.fulton(a)auckland.ac.nz)

o Added mysterious, undocumented --scanflags option.

o Applied patch from Andy Lutomirski (Luto(a)myrealbox.com) which fixes
  some important Windows bugs.  Apparently this can cause a dramatic
  speedup in some circumstances.  The patch had other misc. changes
  too.

o Fix bug noted by Chris V (iselldrugstokidsonline(a)yahoo.com) in which
  Nmap could segmentation fault with the (bogus) command: './nmap -sO
  -p 1-65535 hostname' (protocol only can go up to 255).  That being
  said, Nmap should never segfault just because of bogus options.

o Fixed problem noted by Maximiliano (emax25(a)arnet.com.ar) where Nmap
  would get stuck in a (nearly) infinite loop when you try to "resume"
  a random host (-iR) scan.

o Included a number of fingerprint updates, but I still have many more
  web submissions to go through.  Also made some nmap-services
  portlist updates.

o Included a bunch of fixes (mostly to prevent compiler warnings) from
  William McVey (wam(a)cisco.com)

Nmap 2.54BETA30

o Added a Document Type Definition (DTD) for the Nmap XML output
  format (-oX) to the docs directory.  This allows validating parsers
  to check nmap XML output files for correctness.  It is also useful
  for application programmers to understand the XML output structure.
  The DTD was written by William McVey (wam(a)cisco.com) of Cisco Secure
  Consulting Services ( http://www.cisco.com/go/securityconsulting ).

o Merged in a number of Windows fixes/updates from Andy Lutomirski
  (Luto(a)myrealbox.com)

o Merged in fixes/updates (mostly to the Windows functionality) from
  Matt Hargett (matt(a)use.net)

o Applied patch by Colin Phipps (cph(a)netcraft.com) which correctly
  encodes special characters in the XML output.

o Applied patch by William McVey (wam(a)cisco.com) which adds the uptime
  information printed with -O to the XML output format.

o Fixed byte-order bug in Windows packet matching code which caused
  -PS and -PT to fail.  Bug found and patch sent by Tim Adam
  (tma(a)osa.com.au)

o Fixed segfault problem with "-sU -F".  Nobody reported this until I
  noticed it :(.  Anytime you see "Segmentation Fault" in the latest
  version of Nmap, it is probably a bug -- please mail me the command
  you used, the OS/platform you are running on, and whether it is
  reproducable.

o Added a convenience option "-oA (basefilename)".  This tells Nmap to
  log in ALL the major formats (normal, grepable, and XML).  You give
  a base for the filename, and the output files will be base.nmap,
  base.gnmap, and base.xml.

o Documented the --append_output option which tells Nmap to append
  scan results to any output files you have specified rather than
  overwriting the files.

o Integrate TIMEVAL_SEC_SUBTRACT() fix by Scott Renfro (scott(a)renfro.org)
  which improves timing accuracy.

Nmap 2.54BETA29

o Integrated William McVey's multi-portlist patch.  This allows you to
  specify different port numbers when scanning both TCP & UDP.  For
  example, if you want to UDP for 53,111 and 137 while TCP scanning
  for 21-25,80,139,515,6000,8080 you could do: nmap -sSU -p
  U:53,111,137,T:21-25,80,139,515,6000,8080 target.com .  Prior to
  this patch, you had to either use different Nmap executions or scan
  both UDP & TCP of each port.  See the man page for more usage info.

o Added/updated a bunch of fingerprints, including Windows XP release
  candidates #1 & #2, OpenBSD 2.9, various home gateways/cable modem,
  MacOS X 10.0.4, Linux 2.4.7, Guantlet Firewall 4.0a, a few Cisco
  routers, and, most importantly, the Alcatel Advanced Reflexes IP
  Phone :).  Many other fingerprints were updated as well.

o Found and fixed some relatively major memory leaks based on reports
  sent in by H D Moore (hdm(a)secureaustin.com), mugz
  (mugz(a)x-mafia.org), and Steven Van Acker (deepstar(a)ulyssis.org)

o Applied patch from Chad Loder (chad_loder(a)rapid7.com) which improves
  random target host selection (-iR) by excluding more undesirable
  addresses.

o Fixed portscan timing bug found by H D Moore (hdm(a)secureaustin.com).
  This bug can occur when you specify a --max_rtt_timeout but not
  --initial_rtt_timeout and then scan certain firewalled hosts.

o Fixed port number printing bug found by "Stephen Leavitt"
  (stephen_j_leavitt(a)hotmail.com)

o The Nmap source tarball now extracts with more lenient permissions
  (sometimes world-readable or world-executable, but never
  world-writable).  If you don't want this, set your umask to 077
  (which is what I do).  Suggested by Line Printer (lps(a)rahul.net)

Nmap 2.54BETA28

o I hope that I have fixed the Libpcap "Unknown datalink type" problem that
  many people reported.  If you still receive this error, please send
  me the following info:
  1) Full output of Nmap including the command you typed
  2) What OS/OS version you are using
  3) What type of interface is the scan going through (PPP, ISDN, ethernet,
     PPPoE, etc)
  4) Whether you compiled from source or used the RPM version

o Hopefully fixed Libpcap lex/yacc generated file problem that
  plagued a few folks.

o Various minor fixes/changes/updates

Nmap 2.54BETA27

o Fixed bug that caused "adding open port" messages to be printed even
  when verbose mode was not specified. (patch sent by Doug Hoyte (
  dugely(a)yahoo.com ).

o Fixed bug in zombie:port option parsing in Idlescan as well a few
  other bugs in patch sent by Germano Caronni (gec(a)acm.org)

o Fixed Windows compilation (I broke it when I added Idlescan).

o Fixed a (Win32 only) port identification bug which would cause some
  ports to be listed as "unknown" even when Nmap should know their
  name.  This was found at patched by David Griffiths
  (davidg(a)intrinsica.co.uk).

o Fixed more nmap-os-fingerprints syntax/grammar violations found by
  Raymond Mercier of VIGILANTe

o Fixed a memory leak in Nbase str*casecmp() functions by applying
  patch sent by Matt (matt(a)use.net).  I plan to kill this whole
  strcasecmp.c file as soon as possible (it is a mess).

Nmap 2.54BETA26

o Added Idlescan (IPID blind scan).  The usage syntax is
  "-sI [zombie]".

o Fixed a bunch of fingerprints that were corrupt due to violations of
  the fingerprint syntax/grammar (problems were found by Raymond
  Mercier of VIGILANTe )

o Fixed command-line option parsing bug found
  by "m r rao" (mrrao(a)del3.vsnl.net.in )

o Fixed an OS fingerprinting bug that caused many extra packets to be
  sent if you request a lot of decoys.

o Added some debug code to help diagnose the "Unknown datalink type"
  error.  If Nmap is giving you this error, please send the following
  info to [e-mail] : 1) The full output from Nmap
  (including the command arguments) 2) What OS and OS version are you
  using 3) What type of adaptor are you using (modem, ethernet, FDDI,
  etc)

o Added a bunch of IDS sensor/console/agent port numbers from
  Patrick Mueller (pmueller(a)neohapsis.com)

Nmap 2.54BETA25

o Added a whole bunch of new OS fingerprints (and adjustments) ranging
  from big important ones (Linux 2.4.X, OpenBSD 2.9, FreeBSD 4.3,
  Cisco 12.2.1, MacOS X, etc) to some that are more obscure ( such as
  Apple Color LaserWriter 12/660 PS and VirtualAccess LinxpeedPro 120 )

o Upgraded Libpcap to the latest version (0.6.2) from tcpdump.org.  I
  modified the build system slightly by shipping pre-generated
  scanner.c/grammer.c (instead of using lex/yacc) and I also upgraded
  to the newest config.sub/config.guess .

o Fixed some issues with the new Libpcap under Linux (patches will be
  sent to the developers).

o Added "All zeros" IP.ID sequence classification to account for the
  new Linux 2.4 scheme which seems to use 0 whenever the DF bit is set
  (probably a good idea).

o Tweaked TCP Timestamp and IP.ID sequence classification algorithms

Nmap 2.54BETA24

o Fixed compilation problems on MacOS X publis release.  Thanks to
  Nicolas Dawson (nizcolas(a)myrealbox.com) for securing an account for
  me.

o On the suggestion of the ever-helpful LaMont Jones (lamont(a)hp.com),
  I obtained the newest config.guess/config.sub from
  http://subversions.gnu.org/cgi-bin/cvsweb/config and made
  libpcap/nbase use symlinks rather than copeis of the file

o Applied patch from LaMont Jones (lamont(a)hp.com) which makes Nmap
  compatible with gcc 3.0 (apparently printf() is a macro in that
  version)

o Applied patch from Colin Phipps (cph(a)netcraft.com) which fixes a
  problem that kept UDP RPC scanning from working unless you were also
  doing a TCP scan.

o Applied a patch from Chris Eagle (cseagle(a)redshift.com) which fixes
  Windows compilation (I broke it with a recent change).

o Updated Lithuanian translation of man page based on a newer version sent
  by Aurimas Mikalauskas (inner(a)crazy.lt)

o Killed carriage returns in nmap.c and nmapfe.c, which caused
  problems for some (SGI) compilers.  Problem noted by Artur
  Niederstebruch (artur(a)sgi.com)

o Updated to latest version of rpc program number list, maintained by
  Eilon Gishri (eilon(a)aristo.tau.ac.il)

o Fixed a quoting bug in the Nmap man page found by
  Rasmus Andersson (rasmus(a)pole-position.org)

o Applied RPM spec file changes from "Benjamin Reed"
  (ranger(a)befunk.com) which allows you to avoid building the frontend
  by adding "--define frontend 0" to the build command (eg --rebuild,
  --ba, etc).

Nmap 2.54BETA22

o Eliminated usage of u_int32_t (was causing compilation errors on
  some Sun and HP boxes).  Problem first noted by Nick Munger
  (nmunger(a)Oswego.EDU) and Ralf Hildebrandt
  (Ralf.Hildebrandt(a)innominate.com) and Antonin Sprinzl
  (Antonin.Sprinzl(a)tuwien.ac.at)

o Defined integer-width typedefs such as u32/s32/u16/etc. in Nbase.
  Went through much of the Nmap code and substituted these in where
  correct lengths are important (port numbers, IP addresses, etc).

Nmap 2.54BETA21

o Cleaned up a few build/distribution issues that were reported by
  LaMont Jones (lamont(a)hp.com)

o Fixed compiler warning noted by Gabor Z. Papp (gzp(a)papp.hu) )

Nmap 2.54BETA20

o Added TCP Timestamp sequence checking for OS detection and
  Netcraft-style uptime tests.

o Found and fixed (I hope) byte alignment problem which was causing
  bus errors on SPARC64 ( reported by H D Moore
  (hdm(a)secureaustin.com) and Matthew Franz (mfranz(a)cisco.com) )

o Apple Darwin (Mac OS X) 1.2 portability patch from Rob Braun
  (bbraun(a)synack.net)

o Added IPID sequence number predictability report (also now used in
  OS detection).

o Show actual IPID, TCP ISN, and TCP timestamp values in XML format
  output rather than just the cooked results.

o Suppress IPID and TCP ISN predictability report unless you use -v
  (you need -O as well).

o Applied Solaris 8 compilation fixes from Germano Caronni (
  gec(a)acm.org )

o Applied configure.in variable name typo fixes from Christian
  Weisgerber (naddy(a)openbsd.org)

o Applied some more changes from Andy Lutomirski
  (Luto(a)mailandnews.com) which provides better detection and
  reporting from some heinous errors.

o Added -n and -R (always/never DNS resolve) options to the man page.

Nmap 2.54BETA19

o I ported NmapFE to Windows so that Win32 users can use the graphical
  interface.  It generally works, although I haven't tested much.
  Patches welcome!

o Various little fixes and cleanups, especially to the Windows port.

o Applied patch from Andy Lutomirski (Luto(a)mailandnews.com) which
  enhances some of the Win* error messages and adds the --win_trace
  debugging option.

o Applied some patches from Jay Freeman (saurik(a)saurik.com)
  o New --data_length option adds indicated number of random data
    bytes to send with scan packet and tcp ping packet (does not
    currently work with ICMP ping packet).  Does not affect OS
    detection, RPC, or connect() scan packets.
    o Windows portability fixes
    o Various other little fixes.

o Renamed rpc.h and error.h because they conflict with Windows include
  files.  By the way, this was a pain to figure out because VC++ is
  such a crappy compiler!  It basically just says problem in
  "foobar.h" without giving you any idea how foobar.h got included!
  gcc gives you a nice message tracing the chain of include files!

Nmap 2.54BETA16

o Upgraded to latest version of Winpcap ( 2.1-beta )

o Merged in Windows port code from Ryan Permeh ( ryan(a)eeye.com) and
  Andy Lutomirski ( Luto(a)mailandnews.com ).

o Took out C++ compiler test from nbase configure script.  It was
  inserted accidently, but I found it interesting that only 2 people
  complained about this causing them problems.  I guess most everyone
  already has C++ compilers.

o Applied patch from Steve Bleazard (steve(a)bleazard.com) which fixed
  bug in internal Smoothed Round Trim Time calculations.

o Fixed CFLAGS computation error in configure.  Problem discovered and
  patched by Fredrik Lundholm (exce7(a)ce.chalmers.se)

o Added more debugging code for "Unknown datalink type" error -- if
  you get this, please send me the full error msg including hex
  values.

o Added Portuguese man page translations from Antonio Pires de Castro
  Junior (apcastro(a)ic.unicamp.br).

o Capitalized all references to God in error messages.

Nmap 2.54BETA7

o Applied patch from Hubert Feyrer
  (hubert.feyrer(a)informatik.fh-regensburg.de) which adds support for
  the new NetBSD DLT_PPP_* types.

o Updated to Eilon Gishri's (eilon(a)aristo.tau.ac.il) newest version
  of nmap-rpc at ftp://ftp.tau.ac.il/pub/users/eilon/rpc/rpc

o Moved a bunch of the scanning engine related functions to new files
  (scan_engine.c and scan_engine.h ).  Timing functions were moved to
  the new timing.c/timing.h .  Other stuff was shifted to
  tcpip.c/tcpip.h.  At some point, nmap.c will only contain the Nmap
  command line UI.

o Updated Russian version of man page from Alex Volkov (topcat(a)nm.ru)

Nmap 2.54BETA6

o Added XML output (-oX).  Hopefully this will help those of you
  writing Nmap front ends and other tools that utilize Nmap.  The
  "machine-readable" output has been renamed "grepable" (-oG) to
  emphasize that XML is now the preferred machine-readable output
  format.  But don't worry if your tool uses -oM , that format (and
  the deprecated -oM flag) won't go away any time soon (if ever).
  Thanks to Stou Sandalski (tangui(a)cell2000.net) and Fredrick Paul
  Eisele (phreed(a)gmail.com) for sending proposals that inspired the
  format used.

o Applied patch from Stefan Rapp (s.rapp(a)hrz.uni-dortmund.de) which
  fixes a variable argument integer promotion problem in the new
  snprintf compatibility file.  This is important for Redhat 7
  systems.

o Reorganized output-related routines so that they now reside in
  output.c & output.h.  Let me know if I accidently screwed up the
  behavior of any scan types in the process.

Nmap 2.54BETA5

o Revamped the 'compatibility libraries' subsystem.  Moved all of that
  to a new library called 'libnbase' and changed Nmap and NmapFE to
  use that.  I included a better version of *snprintf and some other
  compatibility files.  Obviously I cannot test these changes on every
  whacked OS that needs this compatibility cruft, so please let me
  know if you run into compilation problems.

o Fixed a problem found by Martyn Tovey (martyn(a)netcraft.com) when
  using Nmap on platforms that dislike division by zero.

o Removed 128.210.*.* addresses from Nmap man page due to complaints
  from Purdue security staff.

o Fixed FreeBSD (some versions) compilation problem found by Martyn
  Tovey (martyn(a)netcraft.com)

Nmap 2.54BETA4

o Upgraded to the very latest Libpcap version ( the 9/3/00 CVS
  snapshot ).  This version is from the tcpdump.org group rather than
  the Lawrence Livermore crew.  The most important advantage is Linux
  Socket Filter support (so you won't have that annoying syslog
  message about Nmap using the obsolete SOCK_PACKET interface).

o I tried to install Nmap on yet another machine without lex/yacc or
  flex/bison.  That was the last straw!  I am now shipping the
  generated C files, which eliminates the lex/yacc requirement.

o Applied patch by Jay Freeman (saurik) (saurik(a)saurik.com) to make
  Nmap C++-clean (this was lot of tedious work!  Thanks!).  Note that
  Nmap still uses a normal C compiler by default, but Nmap derivatives
  may appreciate C++ compatibility.  Note that this only applies to
  "Nmap proper", not libpcap.

o Added a HACKING file for people who want to help with Nmap
  development.  It describes preferred patch formats, development
  resources, and offers a number of useful changes that would likely
  be accepted into the main tree.

o Fixed a configure.in error found by Vacuum
  (vacuum(a)technotronic.com) which could cause compilation errors.

o Fingerprint file adjustments for better Win* detection

o Ensure libpcap is not configured and/or installed if you already
  have a "new enough" version (0.4a6+) installed.

o Included Italian translation of Nmap man page from Giorgio Zoppi
  (deneb(a)supereva.it) .

o Fixed a SYN scan problem that could cause a major slowdown on some
  busy networks.

o Fixed a crash problem in NmapFE reported by sverre ( sverre(a)gmx.net )

o Added an "SInfo" line to most printed fingerprints.  It looks
  similar to this:
  SInfo(V=2.54BETA4%P=i686-pc-linux-gnu%D=9/4%Time=9681031%O=7%C=1)
  and contains information useful when fingerprints are reported (Nmap
  version/platform, scan date, and open/closed ports used)

o Fixed RPCGrind (-sR) scan.  It has been almost completely broken
  since 2.54BETA2 (which has been out for two weeks) and nobody
  reported it!  I noticed the problem myself during testing of
  something else.  I am disappointed that nobody bothered to even let
  me know that this was broken.  Does anyone even use RPC Scan?

o Various other small fixes/improvements

Nmap 2.54BETA3

o Went through and added/adjusted a bunch of fingerprints.  A lot of
  people submitted Windows Millenium Edition (WinME) beta
  fingerprints, but nobody submitted IPs for them.  So please let me
  know if this version detects your WinME boxes.

o Applied NmapFE patch from Michael Fischer v. Mollard (mfvm(a)gmx.de)
  which made did the following:
  o Added delete event so that NmapFE always quits when you kill it
    with your window manager
  o added the menubar to the vbox instead to the fixed widget

o Various small fixes/improvements

Nmap 2.54BETA2

o Added a shortcut which can make single port SYN scans of a network
  much faster.  For example, if a new sendmail vulnerability is found,
  this reduces the time it takes to scan your whole network for port
  25.  This shortcut takes effect when you do "-PS[port] -sS
  -p[port]".  For example 'nmap -n -sS -p25 -PS25 24.0.0.0/8".  This
  optimization doubled the scan speed in a 30,000 IP test I performed.

o Added -sL (List scan).  Just as ping scan (-sP) allows you to short
  circuit the scan right after pinging, -sL allows you to short
  circuit the scan right after target selection.  This allows you to
  see what hosts WOULD be scanned without actually doing it.  The
  hosts will be resolved unles you use -n.  Primary uses:
  1) Get all the IPs in a network (like A.B.C.D/16) and take out
     machines that are too fragile to be scanned safely before
     calling Nmap with the new list (using -iL).
  2) Test that a complex spec like 128.4,5,7-9.*.7 does what you
     expect before actual scanning.
  3) When all you want to do is resolve a bunch of IPs.
  4) You just want results of a zone transfer (if it is implemented).

o Added some new fingerprints and adjusted some others based on
  submissions to the DB (I still have a lot more to go through so
  don't worry if your submission is still not detected).

o Added a warning when you scan 0 hosts (eg "nmap -v").  There are
  various other output tweaks as well.

o Ensured that 0.0.0.0 can be scanned by nmap (although on some OSs,
  like Linux, it won't work due to what seem to be kernel bugs).  Oh
  well.  I'll look into it later.

Nmap 2.54BETA1

o Added an extremely cool scan type by Gerhard Rieger ( rieger at
  iue.tuwien.ac.at ) -- IP Protocol scanning.  Basically it sends a
  bunch of IP headers (no data) with different "protocol" fields to
  the host.  The host then (usually) sends back a protocol unreachable
  for those that it does not support.  By exclusion, nmap can make a
  list of those that are supported.  This is similar in concept to
  (and is implemented using most of the same scanning routines as) UDP
  scanning.  Note that some hosts do not send back protocol
  unreachables -- in that case all protocols will appear "open".

o Fixed an uninitialized variable problem in NmapFE (found by Alvin
  Starr (alvin at iplink.net )

o Fixed a packaging problem that lead to the Nmap man page being
  included twice in the .tgz .

o Fixed dangling nroff include in xnmap man page (noted by Debian Nmap
  package maintainer LaMont Jones (lamont(a)security.hp.com)

o Give a warning when no targets at all are specified

o Updated 'make uninstall' so that it deletes all relevant files

o Included latest nmap-rpc from Eilon Gishri (eilon at aristo.tau.ac.il)

o Eliminated -I. from Nmap's and NmapFE's makefiles (suggested by "Jay
  Freeman (saurik)" (saurik at saurik.com)

o Added Russian documentation by Alex Volkov

o Added Lithuanian documentation from Aurimas Mikalauskas (inner(a)dammit.lt)

Nmap 2.53

o Fixed a commenting issue that could cause trouble for non-GNU
  compilers (first found by Jan-Frode Myklebust (janfrode at
  parallab.uib.no))

o A few new services to nmap-services

Nmap 2.52

o Added very simple man pages for xnmap/nmapfe (lack of man pages for
  these was noticed by LaMont Jones (lamont(a)hp.com), the Debian Nmap
  package maintainer, based on bug report by Adrian Bunk
  (bunk(a)fs.tum.de ).

o Fixed a "Status: Down" machine name output problem in machine
  parseable logs found by Alek O. Komarnitsky (alek(a)ast.lmco.com)

o Took some wierd files out of the doc directory (cd, grep , vi, and
  .swp)

o Fixed some typos found by Thomas Klausner (wiz(a)danbala.ifoer.tuwien.ac.at)

o Updated nmap-rpc with new entries found in the latest version of
  Eilon Gishri's rpc list.

Nmap 2.51

o Fixed target parsing bug found by Steve Horsburgh (shorsburgh(a)horsburgh.com).

o Changed makefile/rpm to store fingerprint, rpc, and services file in
  $prefix/share/nmap rather than $prefix/lib/nmap , since these files
  are architecture independent.  You should now use ./configure
  --datadir instead of ./configure --libdir to change the default
  location.  Suggested by Thomas Klausner
  (wiz(a)danbala.ifoer.tuwien.ac.at).

o I am now including Eilon Gishri's (eilon(a)aristo.tau.ac.il) rpc
  number list (which he recently merged with the Nmap 2.50 rpc list).

o Included Spanish and French HTML versions of the Nmap man page (may
  not always be up to date).

Nmap 2.50

o Fixed an IP calculation error which could occur in some cases where
  you scan machines on different devices (like lo and eth0).  This
  problem was discoved by Jonathan Fine (jfine(a)psu.edu).

o Fixed a problem that could, in rare cases, cause a SYN scan scan to
  crash (the error message was "attempt to add port number X with
  illegal state 0").  This problem was reported by Erik Benner
  (erik(a)xyzzy.net)

o Changed the .spec file so that RPM versions create a xnmap link to
  nmapfe ( the normal make install has done this for a long time ).

Nmap 2.3BETA21

o A number of people reported problems with nmapfe in various
  environments (specifically gdk errors, hangs, and crashes).  I think
  that is now fixed.  Let me know if you still have the problem (make
  sure the title bar says BETA21).

o Added a bunch of OS fingerprints based on all the contributions in
  the last month or so.

o Fixed a bug that completely broke RPC scanning in BETA19.

o Added list of ports scanned near the top of each machine log WHEN
  -v was specified.  Here is an example of the format:
  # Ports scanned: TCP(13;1-10,22,25) UDP(0;)
  The "13" above is the number of TCP ports being scanned.

o Got rid of a snprintf() from nmapfe sine some systems don't have it
  :( and I'm to lazy to integrate in the snprintf that comes with nmap
  right now.

o Fixed important target IP range parsing bug found by Jean-Yves Simon
  ( lethalwp(a)linuxbe.org ).

o Applied patch by albert chin (china at thewrittenword.com) which
  adds --with-libpcap[=DIR] option to configure and and adds an
  elegant approach for -lnsl and -lsocket checking to configure .

o Fixed a bug which could cause Nmap to mark a port filtered based on
  ICMP dest. unreachable packets relating to a different host than the
  one being scanned.

o Fixed output problem relating to ident scan noted by Peter Marschall
  ( peter.marschall at mayn.de )

o Applied patch to services.c by Andrew Brown (atatat(a)atatdot.net)
  which prevents some useless debugging (-d) output when reading some
  kindss of /etc/services files.

o Added "Host: [machinename] (ip) Status: Down" to machine logs when
  the verbose option is given (just like down hosts are reported to
  stdout when verbose is given).  Suggested by Alek Komarnitsky.

o Applied NetBSD compatibility patch provided by Mipam (reinoud at
  ibbnet.org) which changes an autoconf macro to check for
  getopt_long_only instead of getopt_long.

o Nmap used to print an inaccuracy warning when no open TCP ports were
  found on the target machine.  Due to a bug, this was not always
  being printed.  Problem found by Matt (matt at use.net) and Ajay
  Gupta2 (Ajay.Gupta2 at ey.com).

o Added the number of ports in the ignored state right after the state
  name in machine parseable logs.  It used to looke like: "Ignored
  State: closed" whereas now it looks like: "Ignored State: closed
  (1508)" Meaning that 1508 ports were closed and thus are not
  specifically enumerated.

o Changed all nmapfe calls to gdk_font_load into gdk_fontset_load .
  Bennett Feitell (bfeitell at panix.com) suggested that this fixed
  some nmapfe font problems.

Nmap 2.3BETA20

o Applied patch sent in by s.rapp(a)hrz.uni-dortmund.de which fixes a
  memory alignment bug in osscan.c which could cause core dumps on
  machines which require aligned access (like SPARC).

o Fixed a compilation problem on machines that do not have MAP_FAILED
  defined (as a return value to mmap).  Problem noted by Phil
  Stracchino (alaric(a)babcom.com).

Nmap 2.3BETA19

o Tweaked the output so that it now tells how many ports are not shown
  and what state the ignored ports are in.  This info could be
  inferred before by people who had studied the manpage, but now the
  info is explicitly available.  I cleaned up a bunch of stuff
  internally to make this happen.  I hope I didn't break anything!

o Changed NmapFE so that it always kills any running Nmap process when
  you press exit. Problem noted by Marc Renner
  (mrenner(a)ci.marysville.wa.us)

o Apparently some Linux (glibc) systems now come with a "strcasestr"
  function.  So I have made autoconf look for this and use the native
  version if supported. (problem noted by Sami Farin
  (sfarin(a)ratol.fi)).

o Added a new attribute "Ignored State: xxx" to the machine parseable
  logs, where xxx is the state (closed, filtered, or UNfiltered) that
  is being ignored.  Ports in that state are not listed (they weren't
  listed in earlier versions either).  Perhaps I should list ALL ports
  for machine parseable output.  Opinions?

o Merged in a patch sent in by Mipam (reinoud(a)ibbnet.org) which is
  apparently part of the OpenBSD Nmap "port".  Although Nmap seems to
  work fine for me on my OpenBSD 2.4 box, a couple OpenBSD users have
  complained of problems.  Hopefully this will help. (it adds DLT_LOOP
  and DLT_ENC offset cases when reading from libpcap).

o A few really minor bugfixes.

Nmap 2.3BETA18

o Fixed a very important bug that occurred when SYN scanning
  localhost.  Many thanks to Dries Schellekens (
  gwyllion(a)ace.ulyssis.student.kuleuven.ac.be ) for first reporting
  the problem.

o Uros Prestor from TurboLinux informed us that the latest versions of
  Nmap work with Linux on the upcoming Intel Merced/Itanium IA-64
  processors.  He also said that the TurboLinux distribution includes
  Nmap.  Kudos to them!  As well as the other distros that support
  Nmap (Debian, Red Hat, Suse, Trinux) and of course FreeBSD, NetBSD,
  & OpenBSD.  Does anyone know if Nmap ships with the latest from
  Mandrake or Corel?  The latest Solaris includes some Free software.
  If anyone can get them to ship Nmap, I will buy you a case of beer
  :).

o Added a #define to change vsnprintf to vsprintf on machines which do
  not support the former (mostly Solaris 2.5.1 and earlier).  This
  function is less safe.  For people who care about security, we
  recommend an upgrade to Solaris 8 (or Linux/*BSD).

o Changed the NmapFE version to 0.[nmap_version] rather than always
  leaving it at 0.9.5 (which was confusing).  Thanks to J.D.K. Chipps
  (jdkc(a)woptura.com) for noticing this.

o Added support for "-vv" (means the same as "-v -v").  Older versions
  of Nmap supported it (noted by George Kurtz).

Nmap 2.3BETA17

o Added ACK scanning.  This scan technique (which van Houser and
  others have been bugging me to add for years :), is great for
  testing firewall rulesets.  It can NOT find open ports, but it can
  distinguish between filtered/unfilterd by sending an ACK packet to
  each port and waiting for a RST to come back.  Filtered ports will
  not send back a RST (or will send ICMP unreachables).  This scan
  type is activated with -sA .

o Documented the Window scan (-sW) which Lamont Granquist added in
  September 99.

o Added a whole bunch of OS fingerprints that people have submitted.

o "Protocol" field in output eliminated.  It is now printed right next
  to the number (/etc/services style).  Like "22/tcp".  I wonder what
  I should put in the extra white space this leaves on the report :).

o Added --resume option to continue a large network scan where you
  left off.  This is useful for recovering from errors (modem drops
  carrier, network outage, etc).  It also allows you to start and stop
  for policy reasons (like if a client only wants you to scan on
  weekends or at night) or if you want to run the scan on a different
  host.  Usage is 'nmap --resume logfile' where logfile can be either
  normal (-oN) or machine parseable (-oM) logfile from the scan that
  was aborted.  No other options can be given (the options in the
  logfile from the original scan will be used).  Nmap will start off
  with the host after the last one successfully scanned in the log
  file.

o Added --append_output option which causes -oN/-oM/-oS to APPEND to
  the output file you specify rather than overwriting it.

o Various internal code cleanup, makefile fixes, etc.

o Changed version number from 2.3BETA* to 2.30BETA* to appease various
  packaging systems that thought 2.3BETA was < 2.12 .

o Nmap output to files now correctly flushes output after scanning for
  each host is finished.

o Fixed compiler -L flags error found by Ralf Hildebrandt
  (R.Hildebrandt(a)tu-bs.de)

o Fixed configure scripts so that options you give to the Nmap
  configure (like --prefix ) are also passed to the nmapfe configure
  script.  This problem was noted by Ralf Hildebrandt
  (R.Hildebrandt(a)tu-bs.de).  While I was at it, I added some other
  cleanups to the system.

o Added --noninteractive option for when nmap is called from scripts
  (where stuff like prompting users for info is unacceptable).  It
  does not currently do anything (Nmap never prompts) and script
  writers should probably wait until at least May '2000 so their
  scripts still work with earlier versions of Nmap.

o Updated to the latest config.guess and config.sub from Autoconf 2.13

o Applied patch by Sven (s.carstens(a)gmx.de> which fixes a
  segmentation fault problem in Nmapfe colored mode as well as some
  output niceties.

o Changed some C++ comments to C-style for portability (noticed by
  "Sergei V. Rousakov" (sergei(a)cas.Vanderbilt.Edu) )

Nmap 2.3BETA14

o Peter Kosinar (goober(a)gjh.sk) performed some cleanup of the output
  routines and as a bonus he added skript kiddie output mode!!!  Try
  it out by adding "-oS - " to your nmap command line.  Note that
  using '-' to represent stdout instead of a filename is something you
  can do with any of the output modes.

o Ensured that Nmap always gives up on ident scan after the first port
  attempt finds it to be closed (problem noticed by Matt
  (matt(a)use.net))

o Changed strsep's in nmapfe to more portable strtok's (should
  especially help Nmapfe compiles on Solaris)

o Changed permutation algorithm to make port order and host order
  shuffling more random.

o Various minor changes and internal code cleanup.

o Fixed integer overflow that was limiting the max --host_timeout
  value to about 2,000,000 milliseconds (~1/2 hour).  The limit is now
  about 4,000,000,000 milliseconds (~1 month).  I really hope you
  don't need more than that :).

Nmap 2.3BETA13

o I made Nmap smarter about detecting filtering during UDP, Xmas,
  NULL, and FIN scans.

o Updated Nmapfe to 0.9.5 (+ a patch from NmapFE author Zach Smith)

o Fixed a problem where NmapFE would fail to honor $PATH (Noticed by
  K. Scott Rowe (kscott(a)nmt.edu)

o Added a couple ICMP unreachable messages Nmap was missing (found by
  Bifrost (bifrost(a)minions.com)).

o Internal cleanup that improves the way some port lists are stored.

o Added some more RPC numbers from (mmmorris(a)netscape.net)

o Relaxed the dependency requirements of nmapfe rpm (now will accept
  any version of Nmap).

Nmap 2.3BETA12

o Added interactive mode which adds convenience for managing nmap
  sessions and also enhances privacy.  Get to it with --interactive
  and then type 'h' for help.

o Added/modified many fingerprints including the latest 2.3.X Linux
  releases, the latest Win2000 builds, the Apple Airport Wireless
  device, and several dozen more.

o Migrated to RPM .spec file sent in by Tim Powers
  (timp(a)redhat.com).  That is the file they will be using to package
  Nmap with the power tools CD in the next Redhat release.  The most
  important changes are that Nmap (only the RPM version) now installs
  in /usr/* instead of /usr/local/* and the frontend is now
  dynamically linked with GTK and comes in a separate rpm.

o The -i (input from list) option has been deprecated.  From now on
  you should use -iL [filename] to read from a list or -iR to have
  Nmap generate random IPs to scan.  This -iR option is new.

o The -o and -m options have been deprecated.  From now on, you should
  use -oN for normal (human readable) output and -oM for machine
  parseable output.  At some point I might add -oH (HTML output) or
  -oSK (sKr|pt kiDdi3 0uTPut).

o Added --randomize_hosts option, which causes hosts be be scanned in
  non-sequential order.  This makes scans less conspicuous.  For
  efficiency reasons, the hosts are chopped into groups of 2048 and
  then each group is internally shuffled (the groups still go in
  order).

o Rearranged the help ('nmap -h' or 'nmap' or 'nmap --help') screen to
  be shorter (37 -> 23 lines!) and include some of the new features of
  this release.  The man page was updated as well.

o Fixed longstanding bug where nmap -sS mylocalnetwork/24 would not
  successfully scan the host running nmap.

o Internal improvements to make scanning faster with -i (input list)
  or when you specify multiple machines on the command line.

o Uses faster GCD algorithm and fixed several typos (sent in by Peter
  Kosinar).

o Provide more information in machine/human readable output files
  (start time, end time, RPC program name, Nmap version number)

o Killed the -A option (if you don't know what that is then you won't
  miss it.  In fact, even if you do know what it is you won't miss
  it.)

Nmap 2.3BETA10

o Added about 70 new OS fingerprints so that Nmap can detect more
  systems.  The most important new fingerprints are probably:
  * The new SP5+ NT boxes -- After all these years MS FINALLY made
    sequence prediction harder (on NT anyway).
  * Solaris 8 Pre-Release
  * Sega Dreamcast (Hack that!)
  * Latest Windows 2000 builds
  * OpenBSD 2.6

Nmap 2.3BETA9

o Applied patch by Mark Abene (Phiber Optik) to fix several type
  length issues so that it works on Linux/Alpha.

o Applied patch by Matthieu Verbert (mve(a)zurich.ibm.com) to speed up OSScan

Nmap 2.3Beta8

o Added "firewall mode" timing optimizations which can decrease the
  ammount of time neccessary to SYN or connect scan some heavily
  filtered hosts.

o Added min_rtt_timeout timing option (see man page for details)

o Changed "TCP Ping" to use a random ACK value rather than 0 (an IDS
  called Snort was using this to detect Nmap TCP Pings).

o Some changes for better Alpha/Linux support based on investigation
  by Bill Beers (wbeers(a)carolina.rr.com)

o Applied changes for FDDI support by Tobias J. Nijweide (tobias(a)mesa.nl)

o Applied a socket binding patch from LaMont Jones
  (lamont(a)security.hp.com) which can be useful when using -S to
  specify one of multiple interfaces on a machine.

o Made OS detection smart enough to first check scan results for a
  known closed port instead of immediately resorting to a random one.
  This improves OS detection against some machines behind packet
  filters. (suggested by van Hauser)

o Applied a shortcut suggestion by Thomas Reinke which can lead to a
  tremendous speedup against some firewalled hosts.

o Added some ports commonly used for RPC to nmap-services

o Fixed a problem with the timing of an RPC scan (could come before
  the UDP scans they rely on)

o Added a number of new ports to nmap-services

Nmap 2.3Beta6

o Added sophisticated timing controls to give the user much more
  control over Nmap's speed.  This allows you to make Nmap much more
  aggressive to scan hosts faster, or you can make Nmap more "polite"
  -- slower but less likely to wreak havoc on your Network.  You can
  even enforce large delays between sending packets to sneak under IDS
  thresholds and prevent detection.  See the new "Timing Options"
  section of the Nmap man page for more information on using this.

o Applied Lamont Granquist's (lamontg(a)u.washington.edu) Window scan
  patch (I changed the name from ACK scan to Window scan since I may
  add another scan that uses ACK packets and I don't want them to be
  confused).  -sW activates this scan type.  It is mostly effective
  against BSD, AIX, Digital UNIX, and various older HP/UX, SunOS, and
  VAX. (See nmap-hackers mailing list archives for an extensive list).

o Added various long options people expect to see like --version ,
  --help , --usage , etc.  Some of the new timing options are also long.
  I had to add getopt_long C files since most non-Linux boxes don't
  support getopt_long in libc.

o Human readable (-o) output changed to include the time/date of the
  scan.  Suggested by van Hauser.

Nmap 2.3-Beta5

o Changed RPC output based on suggestions by David O'Brien
  (obrien(a)NUXI.com) and Lance Spitzner (lance(a)spitzner.net).  I
  got rid of the "(Non-RPC)" unnecessary clutter which appeared after
  each non RPC port and the "(untested)" that appeard after each
  "filtered" port.

o Added a ton of new OS fingerprints people submitted.  I had about
  400 in my inbox.  Of course, almost 100 of them were submissions for
  www.windows2000test.com :).

o Changed the machine parseable output of RPC information to include
  the version information.  If we figured out the RPC info, it is now
  provided as "program-num*lowversion-highversion".  If we didn't get
  the number, but we think the port is RPC, the field simply contains
  "R".  If we believe the port is NOT RPC, then the field contains
  "N".  If the field is empty, we did not RPC scan the port.  Thanks
  to H D Moore (nlog(a)ings.com) for making me aware how much the
  earlier machine parseable RPC logging sucked :).

Nmap 2.3-Beta4

o Added direct (non-portmapper) RPC scanning to determine what RPC
  program is listening on a particular port.  This works for UDP and
  TCP ports and is currently implemented using sockets (which means
  you can't use decoys, but on the other hand you don't have to be
  root).  Thanks go to ga (ga(a)capyork.com) for writing sample code
  to demonstrate the technique.  The RPC services list included with
  nmap was compiled by Vik Bajaj (vbajaj(a)sas.upenn.edu) with help
  from various members of the nmap-hackers list.

o Fixed a problem that could cause freezes when you scan machines on
  at least two different types of interfaces as part of the same
  command.

o Identified and found workaround for Linux kernel bug which allows
  connect() to sometimes succeed inapropriately when scanning closed
  ports on localhost.

o Fixed problems relating to people who specify the same port more
  than once on the command line.  While the right answer is "well,
  don't do that!", I decided to fix nmap to handle this gracefully.

o Tweaked UDP scanning to be more effective against Solaris ICMP error
  limiting.

o Fixed strtol() integer overflow problem found by Renaud Deraison
  (deraison(a)cvs.nessus.org)

o The HTML translation of the Man page at
  http://www.insecure.org/nmap/nmap_manpage.html should now be
  complete (man2html was dropping lines before).

o Added a note in the man page that Nmap 2.0+ is believed to be
  COMPLETELY Y2K COMPLIANT!  I've been getting a lot of letters from
  laywers about that recently.  You should still be able to port scan
  on Jan 1st (well ... as long as you have electricity and gangs of
  looting thugs haven't stolen your computers :)

Nmap 2.2-Beta4

o Integrated nmapfe code from Zach Smith to allow the nmapfe output
  window to resize when you resize the nmapfe window.

o Integrated patch sent in by Stefan Erben (stefan(a)erben.com) which
  allows nmap to recognize and ignore null interfaces.  If you were
  getting a bogus error like "eth0 not found in /proc/net/route" then
  this should solve your problem.

o Applied patch from Alexander Savelyev (fano(a)ham.kiev.ua) which
  gives nmap the parameters necessary to support SLIP and PPP on BSDI
  systems.

o Upgraded to a new version of shtool (1.2.3)

Nmap 2.2-Beta3

o Adopted Ralf S. Engelschall's excellent shtool script for
  simplifying the nmap makefile and making it more portable

o Various other minor changes to nmapfe.

Nmap 2.2-Beta2

o Cleaned up build environment more, fixed up RPM and Makefile.in,
  eliminated the automake stuff.

o Added nmapfe feature to show nmap command as you change options

o Changed nmapfe to use a global MyWidgets struct rather than tons of
  global vars all over the place.

o Made nmapfe much smarter about rejecting stupid option attempts.  It
  now tries to correct things when you specify illegal options.

o GTK+ 1.0 compatibility fixes

o Integrated nmapfe changes from Zach

Nmap 2.2-BETA1

o Integrated in nmapfe -- a cool front end wrottem by Zach Smith (matrxweb(a)hotmail.com)

Nmap 2.12

o Changed the way tcp connect() scan determines the results of a
  connect() call.  Hopefully this will make nmap a little more
  portable.

o Got rid of the security warning message for people who are missing
  /dev/random and /dev/urandom due to complaints about the warning.
  This only silences the warnings -- it still uses relatively weak
  random number generation under Solaris and other systems that lack
  this functionality.

o Eliminated pow() calls on Linux boxes.  I think some sort of glibc
  bug was causing nmap to sigsegv in some cases inside of pow().  Most
  people weren't affected, but those who were would almost always
  SIGSEGV with -O.

o Fixed an rpm problem noted by Mark Smith (marks(a)senet.com.au)

Nmap 2.11

o Many new fingerprints added.  I received more than 300 submissions
  between this release and the last one.

o Fixed IRIX problems which prevented OS scanning from working on that
  platform.  The problem was researched and solution found by Lamont
  Granquist (lamontg(a)u.washington.edu).  You can also thank him for
  porting nmap to almost every UNIX around.

o Added support for '-m -' to redirect machine readable logs to stdout
  for shell pipelining, etc.  I also changed machine readable output
  to show service names now that we use a nmap specific services file
  rather than /etc/services.  These features were suggested by Dan
  Farmer.  You can also thank him for SATAN (the auditing tool).

o Fixed a link-list bug that could cause hangs in UDP,FIN,NULL, and
  XMAS scans.  Also fixed a ptr problem that could cause SIGSEGV.
  These problem were discovered and tracked down by Ben Laurie
  (ben(a)algroup.co.uk).  You can also thank him for Apache, OpenSSL,
  and Apache-SSL.

o Fixed installation problem for people without a /usr/local/man/man1
  directory.  Found by Jeffrey Robertson (a-jeffro(a)microsoft.com).
  I guess you can thank him for Win98 ;).

o Several other little fixes to the installation script and minor
  scanner tweaks.

Nmap 2.10

o Private test release

Nmap 2.09

o Private test release

Nmap 2.08

o Bugfix for problem that can cause nmap to appear to "freeze up" for
  long periods of time when run on some busy networks. (found by
  Lamont Granquist)

Nmap 2.07

o Fixed a lockup on Solaris (and perhaps other proprietary UNIX
  systems) caused by a lack of /dev/random & /dev/urandom and a rand()
  that only returns values up to 65535.  Users of Free operating
  systems like Linux, FreeBSD, or OpenBSD probably shouldn't bother
  upgrading.

Nmap 2.06

o Fixed compile problems on machines which lack snprintf() (found by
  Ken Williams (jkwilli2(a)unity.ncsu.edu))

o Added the squid proxy to nmap-services (suggested by Holger Heimann)

o Fixed a problem where the new memory allocation system was handing
  out misaligned pointers.

o Fixed another memory allocation bug which probably doesn't cause any
  real-life problems.

o Made nmap look in more places for nmap-os-fingerprints

Nmap 2.05

o Tons of new fingerprints.  The number has grown by more than 25%.
  In particular, Charles M. Hannum (root(a)ihack.net) fixed several
  problems with NetBSD that made it easy to fingerprint and he sent me
  a huge new batch of fingerprints for various NetBSD releases down to
  1.2.  Other people sent NetBSD fingerprints down to 1.0.  I finally
  got some early Linux fingerprints in (down to 1.09).

o Nmap now comes with its own nmap-services which I created by merging
  the /etc/services from a bunch of OS' and then adding Netbus, Back
  Orifice, etc.

o Random number generation now takes advantage of the /dev/urandom or
  /dev/random that most Free operating systems offer.

o Increased the maximum number of OS guesses nmap will make, told nmap
  never to give you two matches where the OS names are byte-to-byte
  equivalent.  Fixed nmap to differentiate between "no OS matches
  found" and "too many OS matches to list".

o Fixed an information leak in the packet TTL values (found by HD
  Moore (hdmoore(a)usa.net))

o Fixed the problem noted by Savva Uspensky about offsets used for
  various operating systems' PPP/SLIP headers.  Due to lack of
  responses regarding other operating systems, I have made assumptions
  about what works for BSDI, NetBSD, and SOLARIS.  If this version no
  longer works on your modem, please let me know (and tell me whether
  you are using SLIP/PPP and what OS you are running).

o Machine parseable logs are now more machine parseable (I now use a
  tab to seperate test result fields rather than the more ambiguous
  spaces.  This may break a few things which rely on the old format.
  Sorry.  They should be easy to fix.

o Added my nmap-fingerprintinting-article.txt to the distribution in
  the docs directory.

o Fixed problem where nmap -sS (my_ethernet_or_ppp_ip_address) would
  not correctly scan localhost (due to the kernel rerouting the
  traffic through localhost).  Nmap should now detect and work around
  this behavior.

o Applied patch sent to my by Bill Fenner (fenner(a)parc.xerox.com)
  which fixes various SunOS compatibility problems.

o Changed the makefile 'all' target to use install-sh rather than
  mkdir -p (doesn't work on some systems)

o Documentation updated and clarified slightly.

o Added this CHANGELOG file to the distribution.