Партнёры:
Хостинг:
Archive-name: uucp-internals
Version: $Revision: 1.105 $
Last-modified: $Date: 1995/07/16 16:09:52 $
UUCP Protocol Internals
***********************
Recent changes:
* Conversion to Texinfo format.
* Description of the `E' command.
* Description of optional number following `-N' and `ROKN' in UUCP
protocol startup.
* Detailed description of the `y' protocol.
* Mention the name uuxqt uses for lock files.
This article was written by Ian Lance Taylor `<ian@airs.com>' and I may
even update it periodically. Please send me mail about suggestions or
inaccuracies.
This article describes how the various UUCP protocols work, and
discusses some other internal UUCP issues. It does not describe how to
configure UUCP, nor how to solve UUCP connection problems, nor how to
deal with UUCP mail. I do not know of any FAQ postings on these topics.
There are some documents on the net describing UUCP configuration, but I
can not keep an up to date list here; try using archie.
If you haven't read the `news.announce.newusers' articles, read them.
This article is in digest format. Some newsreaders will be able to
break it apart into separate articles. Please don't ask me how to do
this, though.
This article covers the following topics. If questions about one of
these topics is posted to `comp.mail.uucp', please send mail to the
poster referring her or him to this FAQ. There is no reason to post a
followup, as most of us know the answer already.
UUCP Protocol Sources
Alarm in Debugging Output
UUCP Grades
UUCP Lock Files
Execution File Format
UUCP Protocol
UUCP `g' Protocol
UUCP `f' Protocol
UUCP `t' Protocol
UUCP `e' Protocol
UUCP `G' Protocol
UUCP `i' Protocol
UUCP `j' Protocol
UUCP `x' Protocol
UUCP `y' Protocol
UUCP `d' Protocol
UUCP `h' Protocol
UUCP `v' Protocol
Thanks
----------------------------------------------------------------------
From: UUCP Protocol Sources
Subject: UUCP Protocol Sources
UUCP Protocol Sources
=====================
"Unix-to-Unix Copy Program," said PDP-1. "You will never find a
more wretched hive of bugs and flamers. We must be cautious."
--DECWars
I took a lot of the information from Jamie E. Hanrahan's paper in the
Fall 1990 DECUS Symposium, and from `Managing UUCP and Usenet' by Tim
O'Reilly and Grace Todino (with contributions by several other people).
The latter includes most of the former, and is published by
O'Reilly & Associates, Inc.
103 Morris Street, Suite A
Sebastopol, CA 95472
It is currently in its tenth edition. The ISBN number is
`0-937175-93-5'.
Some information is originally due to a Usenet article by Chuck Wegrzyn.
The information on execution files comes partially from Peter Honeyman.
The information on the `g' protocol comes partially from a paper by
G.L. Chesson of Bell Laboratories, partially from Jamie E. Hanrahan's
paper, and partially from source code by John Gilmore. The information
on the `f' protocol comes from the source code by Piet Berteema. The
information on the `t' protocol comes from the source code by Rick
Adams. The information on the `e' protocol comes from a Usenet article
by Matthias Urlichs. The information on the `d' protocol comes from
Jonathan Clark, who also supplied information about QFT. The UUPlus
information comes straight from Christopher J. Ambler, of UUPlus
Development; it applies to version 1.52 and up of the shareware version
of UUPlus Utilities, called FSUUCP 1.52, but referred to in this
article as UUPlus.
Although there are few books about UUCP, there are many about networks
and protocols in general. I recommend two non-technical books which
describe the sorts of things that are available on the network: `The
Whole Internet', by Ed Krol, and `Zen and the Art of the Internet', by
Brendan P. Kehoe. Good technical discussions of networking issues can
be found in `Internetworking with TCP/IP', by Douglas E. Comer and
David L. Stevens and in `Design and Validation of Computer Protocols'
by Gerard J. Holzmann.
------------------------------
From: Alarm in Debugging Output
Subject: Alarm in Debugging Output
Alarm in Debugging Output
=========================
The debugging output of many versions of UUCP will include messages like
`alarm 1' or `pkcget: alarm 1'. Taylor UUCP does not use the word
`alarm', but will instead log messages like `Timed out waiting for
packet'.
These types of messages mean that the UUCP package has timed out while
waiting for some sort of response from the remote system. If it happens
consistently when trying to transfer a particular file, then the most
likely problem is that one of the modems will not transmit the XON or
XOFF characters. Several UUCP protocols require an eight bit clean
connection, which means that the modems must treat XON or XOFF as normal
data characters, not as flow control signals. This should always be
checked first.
Other possible problems are that the modems have simply dropped their
connection, or perhaps on one side or the other the serial buffer is
overflowing and dropping characters. Another possibility is that the
UUCP packages disagree about some aspect of the UUCP protocol, which is
uncommon but does happen occasionally.
Using the information in the following sections, you should be able to
figure out what type of data your UUCP was expecting to receive. This
may give some indication as to exactly what the problem is. It is
difficult to be more specific, since there are many possiblities.
------------------------------
From: UUCP Grades
Subject: UUCP Grades
UUCP Grades
===========
Modern UUCP packages support a priority grade for each command. The
grades generally range from `A' (the highest) to `Z' followed by `a' to
`z'. Some UUCP packages (including Taylor UUCP) also support `0' to
`9' before `A'. Some UUCP packages may permit any ASCII character as a
grade.
On Unix, these grades are encoded in the name of the command file
created by `uucp' or `uux'. A command file name generally has the form
`C.nnnngssss' where `nnnn' is the remote system name for which the
command is queued, `g' is a single character grade, and `ssss' is a
four character sequence number. For example, a command file created
for the system `airs' at grade `Z' might be named `C.airsZ2551'.
The remote system name will be truncated to seven characters, to ensure
that the command file name will fit in the 14 character file name limit
of the traditional Unix file system. UUCP packages which have no other
means of distinguishing which command files are intended for which
systems thus require all systems they connect to to have names that are
unique in the first seven characters. Some UUCP packages use a variant
of this format which truncates the system name to six characters. HDB
and Taylor UUCP use a different spool directory format, which allows up
to fourteen characters to be used for each system name.
The sequence number in the command file name may be a decimal integer,
or it may be a hexadecimal integer, or it may contain any alphanumeric
character. Different UUCP packages are different.
UUPlus Utilities (as FSUUCP, a shareware DOS based UUCP and news
package) uses up to 8 characters for file names in the spool (this is a
DOS file system limitation; actually, with the extension, 11 characters
are available, but FSUUCP reserves that for future use). FSUUCP
defaults mail to grade `D', and news to grade `N', except that when the
grade of incoming mail can be determined, that grade is preserved if
the mail is forwarded to another system. The default grades may be
changed by editing the `LIB/MAILRC' file for mail, or the `UUPLUS.CFG'
file for news.
UUPC/extended for DOS, OS/2 and Windows NT handles mail at grade `C',
news at grade `d', and file transfers at grade `n'. The UUPC/extended
`UUCP' and `RMAIL' commands accept grades to override the default, the
others do not.
I do not know how command grades are handled in other non-Unix UUCP
packages.
Modern UUCP packages allow you to restrict file transfer by grade
depending on the time of day. Typically this is done with a line in
the `Systems' (or `L.sys') file like this:
airs Any/Z,Any2305-0855 ...
This allows grades `Z' and above to be transferred at any time. Lower
grades may only be transferred at night. I believe that this grade
restriction applies to local commands as well as to remote commands,
but I am not sure. It may only apply if the UUCP package places the
call, not if it is called by the remote system.
Taylor UUCP can use the `timegrade' and `call-timegrade' commands to
achieve the same effect. It supports the above format when reading
`Systems' or `L.sys'.
UUPC/extended provides the `symmetricgrades' option to announce the
current grade in effect when calling the remote system.
UUPlus allows specification of the highest grade accepted on a per-call
basis with the `-g' option in `UUCICO'.
This sort of grade restriction is most useful if you know what grades
are being used at the remote site. The default grades used depend on
the UUCP package. Generally `uucp' and `uux' have different defaults.
A particular grade can be specified with the `-g' option to `uucp' or
`uux'. For example, to request execution of `rnews' on `airs' with
grade `d', you might use something like
uux -gd - airs!rnews < article
Uunet queues up mail at grade `C', but increases the grade based on the
size. News is queued at grade `d', and file transfers at grade `n'.
The example above would allow mail (below some large size) to be
received at any time, but would only permit news to be transferred at
night.
------------------------------
From: UUCP Lock Files
Subject: UUCP Lock Files
UUCP Lock Files
===============
This discussion applies only to Unix. I have no idea how UUCP locks
ports on other systems.
UUCP creates files to lock serial ports and systems. On most, if not
all, systems, these same lock files are also used by `cu' to coordinate
access to serial ports. On some systems `getty' also uses these lock
files, often under the name `uugetty'.
The lock file normally contains the process ID of the locking process.
This makes it easy to determine whether a lock is still valid. The
algorithm is to create a temporary file and then link it to the name
that must be locked. If the link fails because a file with that name
already exists, the existing file is read to get the process ID. If the
process still exists, the lock attempt fails. Otherwise the lock file
is deleted and the locking algorithm is retried.
Older UUCP packages put the lock files in the main UUCP spool directory,
`/usr/spool/uucp'. HDB UUCP generally puts the lock files in a
directory of their own, usually `/usr/spool/locks' or `/etc/locks'.
The original UUCP lock file format encodes the process ID as a four byte
binary number. The order of the bytes is host-dependent. HDB UUCP
stores the process ID as a ten byte ASCII decimal number, with a
trailing newline. For example, if process 1570 holds a lock file, it
would contain the eleven characters space, space, space, space, space,
space, one, five, seven, zero, newline. Some versions of UUCP add a
second line indicating which program created the lock (`uucp', `cu', or
`getty/uugetty'). I have also seen a third type of UUCP lock file
which does not contain the process ID at all.
The name of the lock file is traditionally `LCK..' followed by the base
name of the device. For example, to lock `/dev/ttyd0' the file
`LCK..ttyd0' would be created. On SCO Unix, the lock file name is
always forced to lower case even if the device name has upper case
letters.
System V Release 4 UUCP names the lock file using the major and minor
device numbers rather than the device name. The file is named
`LK.XXX.YYY.ZZZ', where XXX, YYY and ZZZ are all three digit decimal
numbers. XXX is the major device number of the device holding the
directory holding the device file (e.g., `/dev'). YYY is the major
device number of the device file itself. ZZZ is the minor device
number of the device file itself. If `s' holds the result of passing
the device to the stat system call (e.g., `stat ("/dev/ttyd0", &s)'),
the following line of C code will print out the corresponding lock file
name:
printf ("LK.%03d.%03d.%03d", major (s.st_dev),
major (s.st_rdev), minor (s.st_rdev));
The advantage of this system is that even if there are several links to
the same device, they will all use the same lock file name.
When two or more instances of `uuxqt' are executing, some sort of
locking is needed to ensure that a single execution job is only started
once. I don't know how most UUCP packages deal with this. Taylor UUCP
uses a lock file for each execution job. The name of the lock file is
the same as the name of the `X.*' file, except that the initial `X' is
changed to an `L'. The lock file holds the process ID as described
above.
------------------------------
From: Execution File Format
Subject: Execution File Format
Execution File Format
=====================
UUCP `X.*' files control program execution. They are created by `uux'.
They are transferred between systems just like any other file. The
`uuxqt' daemon reads them to figure out how to execute the job
requested by `uux'.
An `X.*' file is simply a text file. The first character of each line
is a command, and the remainder of the line supplies arguments. The
following commands are defined:
`C command'
This gives the command to execute, including the program and all
arguments. For example, `rmail ian@airs.com'.
`U user system'
This names the user who requested the command, and the system from
which the request came.
`I standard-input'
This names the file from which standard input is taken. If no
standard input file is given, the standard input will probably be
attached to `/dev/null'. If the standard input file is not from
the system on which the execution is to occur, it will also appear
in an `F' command.
`O standard-output [system]'
This names the standard output file. The optional second argument
names the system to which the file should be sent. If there is no
second argument, the file should be created on the executing
system.
`F required-file [filename-to-use]'
The `F' command can appear multiple times. Each `F' command names
a file which must exist before the execution can proceed. This
will usually be a file which is transferred from the system on
which `uux' was executed, but it can also be a file from the local
system or some other system. If the file is not from the local
system, then the command will usually name a file in the spool
directory. If the optional second argument appears, then the file
should be copied to the execution directory under that name. This
is necessary for any file other than the standard input file. If
the standard input file is not from the local system, it will
appear in both an `F' command and an `I' command.
`R requestor-address'
This is the address to which mail about the job should be sent.
It is relative to the system named in the `U' command. If the `R'
command does not appear, then mail is sent to the user named in the
`U' command.
`Z'
This command takes no arguments. It means that a mail message
should be sent if the command failed. This is the default
behaviour for most modern UUCP packages, and for them the `Z'
command does not actually do anything.
`N'
This command takes no arguments. It means that no mail message
should be sent, even if the command failed.
`n'
This command takes no arguments. It means that a mail message
should be sent if the command succeeded. Normally a message is
sent only if the command failed.
`B'
This command takes no arguments. It means that the standard input
should be returned with any error message. This can be useful in
cases where the input would otherwise be lost.
`e'
This command takes no arguments. It means that the command should
be processed with `/bin/sh'. For some packages this is the default
anyhow. Most packages will refuse to execute complex commands or
commands containing wildcards, because of the security holes this
opens.
`E'
This command takes no arguments. It means that the command should
be processed with the `execve' system call. For some packages
this is the default anyhow.
`M status-file'
This command means that instead of mailing a message, the message
should be copied to the named file on the system named by the `U'
command.
`# comment'
This command is ignored, as is any other unrecognized command.
Here is an example. Given the following command executed on system
test1
uux - test2!cat - test2!~ian/bar !qux '>~/gorp'
(this is only an example, as most UUCP systems will not permit the cat
command to be executed) Taylor UUCP will produce something like the
following `X.' file:
U ian test1
F D.test1N003r qux
O /usr/spool/uucppublic test1
F D.test1N003s
I D.test1N003s
C cat - ~ian/bar qux
The standard input will be read into a file and then transferred to the
file `D.test1N003s' on system `test2'. The file `qux' will be
transferred to `D.test1N003r' on system `test2'. When the command is
executed, the latter file will be copied to the execution directory
under the name `qux'. Note that since the file `~ian/bar' is already
on the execution system, no action need be taken for it. The standard
output will be collected in a file, then copied to the directory
`/usr/spool/uucppublic' on the system `test1'.
------------------------------
From: UUCP Protocol
Subject: UUCP Protocol
UUCP Protocol
=============
The UUCP protocol is a conversation between two UUCP packages. A UUCP
conversation consists of three parts: an initial handshake, a series of
file transfer requests, and a final handshake.
The Initial Handshake
---------------------
Before the initial handshake, the caller will usually have logged in the
called machine and somehow started the UUCP package there. On Unix this
is normally done by setting the shell of the login name used to
`/usr/lib/uucp/uucico'.
All messages in the initial handshake begin with a `^P' (a byte with
the octal value `\020') and end with a null byte (`\000'). A few
systems end these messages with a line feed character (`\012') instead
of a null byte; the examples below assume a null byte is being used.
Some options below are supported by QFT, which stands for Queued File
Transfer, and is (or was) an internal Bell Labs version of UUCP.
Taylor UUCP size negotiation was introduced by Taylor UUCP, and is also
supported by DOS based UUPlus and Amiga based wUUCP and UUCP-1.17.
The initial handshake goes as follows. It is begun by the called
machine.
called: `\020Shere=hostname\000'
The hostname is the UUCP name of the called machine. Older UUCP
packages do not output it, and simply send `\020Shere\000'.
caller: `\020Shostname options\000'
The hostname is the UUCP name of the calling machine. The
following options may appear (or there may be none):
`-QSEQ'
Report sequence number for this conversation. The sequence
number is stored at both sites, and incremented after each
call. If there is a sequence number mismatch, something has
gone wrong (somebody may have broken security by pretending
to be one of the machines) and the call is denied. If the
sequence number changes on one of the machines, perhaps
because of an attempted breakin or because a disk backup was
restored, the sequence numbers on the two machines must be
reconciled manually.
`-xLEVEL'
Requests the called system to set its debugging level to the
specified value. This is not supported by all systems.
`-pGRADE'
`-vgrade=GRADE'
Requests the called system to only transfer files of the
specified grade or higher. This is not supported by all
systems. Some systems support `-p', some support `-vgrade='.
UUPlus allows either `-p' or `-v' to be specified on a
per-system basis in the `SYSTEMS' file (`gradechar' option).
`-R'
Indicates that the calling UUCP understands how to restart
failed file transmissions. Supported only by System V
Release 4 UUCP, QFT, and Taylor UUCP.
`-ULIMIT'
Reports the ulimit value of the calling UUCP. The limit is
specified as a base 16 number in C notation (e.g.,
`-U0x1000000'). This number is the number of 512 byte blocks
in the largest file which the calling UUCP can create. The
called UUCP may not transfer a file larger than this.
Supported only by System V Release 4 UUCP, QFT and UUPlus.
UUPlus reports the lesser of the available disk space on the
spool directory drive and the ulimit variable in
`UUPLUS.CFG'. Taylor UUCP understands this option, but does
not generate it.
`-N[NUMBER]'
Indicates that the calling UUCP understands the Taylor UUCP
size negotiation extension. Not supported by traditional
UUCP packages. Supported by UUPlus. The optional number is
a bitmask of features supported by the calling UUCP, and is
described below.
called: `\020ROK\000'
There are actually several possible responses.
`ROK'
The calling UUCP is acceptable, and the handshake proceeds to
the protocol negotiation. Some options may also appear; see
below.
`ROKN[NUMBER]'
The calling UUCP is acceptable, it specified `-N', and the
called UUCP also understands the Taylor UUCP size limiting
extensions. The optional number is a bitmask of features
supported by the called UUCP, and is described below.
`RLCK'
The called UUCP already has a lock for the calling UUCP,
which normally indicates the two machines are already
communicating.
`RCB'
The called UUCP will call back. This may be used to avoid
impostors (but only one machine out of each pair should call
back, or no conversation will ever begin).
`RBADSEQ'
The call sequence number is wrong (see the `-Q' discussion
above).
`RLOGIN'
The calling UUCP is using the wrong login name.
`RYou are unknown to me'
The calling UUCP is not known to the called UUCP, and the
called UUCP does not permit connections from unknown systems.
Some versions of UUCP just drop the line rather than sending
this message.
If the response is `ROK', the following options are supported by
System V Release 4 UUCP and QFT.
`-R'
The called UUCP knows how to restart failed file
transmissions.
`-ULIMIT'
Reports the ulimit value of the called UUCP. The limit is
specified as a base 16 number in C notation. This number is
the number of 512 byte blocks in the largest file which the
called UUCP can create. The calling UUCP may not send a file
larger than this. Also supported by UUPlus. Taylor UUCP
understands this option, but does not generate it.
`-xLEVEL'
I'm not sure just what this means. It may request the
calling UUCP to set its debugging level to the specified
value.
If the response is not `ROK' (or `ROKN') both sides hang up the
phone, abandoning the call.
called: `\020Pprotocols\000'
Note that the called UUCP outputs two strings in a row. The
protocols string is a list of UUCP protocols supported by the
caller. Each UUCP protocol has a single character name. These
protocols are discussed in more detail later in this document.
For example, the called UUCP might send `\020Pgf\000'.
caller: `\020Uprotocol\000'
The calling UUCP selects which protocol to use out of the protocols
offered by the called UUCP. If there are no mutually supported
protocols, the calling UUCP sends `\020UN\000' and both sides hang
up the phone. Otherwise the calling UUCP sends something like
`\020Ug\000'.
Most UUCP packages will consider each locally supported protocol in turn
and select the first one supported by the called UUCP. With some
versions of HDB UUCP, this can be modified by giving a list of protocols
after the device name in the `Devices' file or the `Systems' file. For
example, to select the `e' protocol in `Systems',
airs Any ACU,e ...
or in Devices,
ACU,e ttyXX ...
Taylor UUCP provides the `protocol' command which may be used either
for a system or a port. UUPlus allows specification of the protocol
string on a per-system basis in the `SYSTEMS' file.
The optional number following a `-N' sent by the calling system, or an
`ROKN' sent by the called system, is a bitmask of features supported by
the UUCP package. The optional number was introduced in Taylor UUCP
version 1.04. The number is sent as an octal number with a leading
zero. The following bits are currently defined. A missing number
should be taken as `011'.
`01'
UUCP supports size negotiation.
`02'
UUCP supports file restart.
`04'
UUCP supports the `E' command.
`010'
UUCP requires the file size in the `S' and `R' commands to be in
base 10. This bit is used by default if no number appears, but
should not be explicitly sent.
`020'
UUCP expects a dummy string between the notify field and the size
field in an `S' command. This is true of SVR4 UUCP. This bit
should not be used.
After the protocol has been selected and the initial handshake has been
completed, both sides turn on the selected protocol. For some protocols
(notably `g') a further handshake is done at this point.
UUCP Protocol Commands
----------------------
Each protocol supports a method for sending a command to the remote
system. This method is used to transmit a series of commands between
the two UUCP packages. At all times, one package is the master and the
other is the slave. Initially, the calling UUCP is the master.
If a protocol error occurs during the exchange of commands, both sides
move immediately to the final handshake.
The master will send one of five commands: `S', `R', `X', `E', or `H'.
Any file name referred to below is either an absolute file name
beginning with `/', a public directory file name beginning with `~/', a
file name relative to a user's home directory beginning with `~USER/',
or a spool directory file name. File names in the spool directory are
not absolute, but instead are converted to file names within the spool
directory by UUCP. They always begin with `C.' (for a command file
created by `uucp' or `uux'), `D.' (for a data file created by `uucp',
`uux' or by an execution, or received from another system for an
execution), or `X.' (for an execution file created by `uux' or received
from another system).
The S Command
.............
master: `S FROM TO USER -OPTIONS TEMP MODE NOTIFY SIZE'
The `S' and the `-' are literal characters. This is a request by
the master to send a file to the slave.
FROM
The name of the file to send. If the `C' option does not
appear in OPTIONS, the master will actually open and send
this file. Otherwise the file has been copied to the spool
directory, where it is named TEMP. The slave ignores this
field unless TO is a directory, in which case the basename of
FROM will be used as the file name. If FROM is a spool
directory filename, it must be a data file created for or by
an execution, and must begin with `D.'.
TO
The name to give the file on the slave. If this field names
a directory the file is placed within that directory with the
basename of FROM. A name ending in `/' is taken to be a
directory even if one does not already exist with that name.
If TO begins with `X.', an execution file will be created on
the slave. Otherwise, if TO begins with `D.' it names a data
file to be used by some execution file. Otherwise, TO should
not be in the spool directory.
USER
The name of the user who requested the transfer.
OPTIONS
A list of options to control the transfer. The following
options are defined (all options are single characters):
`C'
The file has been copied to the spool directory (the
master should use TEMP rather than FROM).
`c'
The file has not been copied to the spool directory
(this is the default).
`d'
The slave should create directories as necessary (this
is the default).
`f'
The slave should not create directories if necessary,
but should fail the transfer instead.
`m'
The master should send mail to USER when the transfer is
complete.
`n'
The slave should send mail to NOTIFY when the transfer is
complete.
TEMP
If the `C' option appears in OPTIONS, this names the file to
be sent. Otherwise if FROM is in the spool directory, TEMP
is the same as FROM. Otherwise TEMP may be a dummy string,
such as `D.0'. After the transfer has been succesfully
completed, the master will delete the file TEMP.
MODE
This is an octal number giving the mode of the file on the
master. If the file is not in the spool directory, the slave
will always create it with mode 0666, except that if (MODE &
0111) is not zero (the file is executable), the slave will
create the file with mode 0777. If the file is in the spool
directory, some UUCP packages will use the algorithm above
and some will always create the file with mode 0600. This
field is ignored by UUPlus, since it is meaningless on DOS;
UUPlus uses 0666 for outgoing files.
NOTIFY
This field may not be present, and in any case is only
meaningful if the `n' option appears in OPTIONS. If the `n'
option appears, then, when the transfer is successfully
completed, the slave will send mail to NOTIFY, which must be
a legal mailing address on the slave. If a SIZE field will
appear but the `n' option does not appear, NOTIFY will always
be present, typically as the string `dummy' or simply a pair
of double quotes.
SIZE
This field is only present when doing Taylor UUCP or SVR4
UUCP size negotiation. It is the size of the file in bytes.
Taylor UUCP version 1.03 sends the size as a decimal integer,
while versions 1.04 and up, and all other UUCP packages that
support size negotiation, send the size in base 16 with a
leading 0x.
The slave then responds with an `S' command response.
`SY START'
The slave is willing to accept the file, and file transfer
begins. The START field will only be present when using file
restart. It specifies the byte offset into the file at which
to start sending. If this is a new file, START will be 0x0.
`SN2'
The slave denies permission to transfer the file. This can
mean that the destination directory may not be accessed, or
that no requests are permitted. It implies that the file
transfer will never succeed.
`SN4'
The slave is unable to create the necessary temporary file.
This implies that the file transfer might succeed later.
`SN6'
This is only used by Taylor UUCP size negotiation. It means
that the slave considers the file too large to transfer at
the moment, but it may be possible to transfer it at some
other time.
`SN7'
This is only used by Taylor UUCP size negotiation. It means
that the slave considers the file too large to ever transfer.
`SN8'
This is only used by Taylor UUCP. It means that the file was
already received in a previous conversation. This can happen
if the receive acknowledgement was lost after it was sent by
the receiver but before it was received by the sender.
`SN9'
This is only used by Taylor UUCP (versions 1.05 and up) and
UUPlus (versions 2.0 and up). It means that the remote
system was unable to open another channel (see the discussion
of the `i' protocol for more information about channels).
This implies that the file transfer might succeed later.
`SN10'
This is reportedly used by SVR4 UUCP to mean that the file
size is too large.
If the slave responds with `SY', a file transfer begins. When the
file transfer is complete, the slave sends a `C' command response.
`CY'
The file transfer was successful.
`CYM'
The file transfer was successful, and the slave wishes to
become the master; the master should send an `H' command,
described below.
`CN5'
The temporary file could not be moved into the final
location. This implies that the file transfer will never
succeed.
After the `C' command response has been received (in the `SY' case) or
immediately (in an `SN' case) the master will send another command.
The R Command
.............
master: `R FROM TO USER -OPTIONS SIZE'
The `R' and the `-' are literal characters. This is a request by
the master to receive a file from the slave. I do not know how
SVR4 UUCP or QFT implement file transfer restart in this case.
FROM
This is the name of the file on the slave which the master
wishes to receive. It must not be in the spool directory,
and it may not contain any wildcards.
TO
This is the name of the file to create on the master. I do
not believe that it can be a directory. It may only be in
the spool directory if this file is being requested to
support an execution either on the master or on some system
other than the slave.
USER
The name of the user who requested the transfer.
OPTIONS
A list of options to control the transfer. The following
options are defined (all options are single characters):
`d'
The master should create directories as necessary (this
is the default).
`f'
The master should not create directories if necessary,
but should fail the transfer instead.
`m'
The master should send mail to USER when the transfer is
complete.
SIZE
This only appears if Taylor UUCP size negotiation is being
used. It specifies the largest file which the master is
prepared to accept (when using SVR4 UUCP or QFT, this was
specified in the `-U' option during the initial handshake).
The slave then responds with an `R' command response. UUPlus does
not support `R' requests, and always responds with `RN2'.
`RY MODE [SIZE]'
The slave is willing to send the file, and file transfer
begins. The MODE argument is the octal mode of the file on
the slave. The master treats this just as the slave does the
MODE argument in the send command, q.v. I am told that SVR4
UUCP sends a trailing SIZE argument. For some versions of
BSD UUCP, the MODE argument may have a trailing `M' character
(e.g., `RY 0666M'). This means that the slave wishes to
become the master.
`RN2'
The slave is not willing to send the file, either because it
is not permitted or because the file does not exist. This
implies that the file request will never succeed.
`RN6'
This is only used by Taylor UUCP size negotiation. It means
that the file is too large to send, either because of the
size limit specifies by the master or because the slave
considers it too large. The file transfer might succeed
later, or it might not (this may be cleared up in a later
release of Taylor UUCP).
`RN9'
This is only used by Taylor UUCP (versions 1.05 and up) and
FSUUCP (versions 1.5 and up). It means that the remote
system was unable to open another channel (see the discussion
of the `i' protocol for more information about channels).
This implies that the file transfer might succeed later.
If the slave responds with `RY', a file transfer begins. When the
file transfer is complete, the master sends a `C' command. The
slave pretty much ignores this, although it may log it.
`CY'
The file transfer was successful.
`CN5'
The temporary file could not be moved into the final location.
After the `C' command response has been sent (in the `RY' case) or
immediately (in an `RN' case) the master will send another command.
The X Command
.............
master: `X FROM TO USER -OPTIONS'
The `X' and the `-' are literal characters. This is a request by
the master to, in essence, execute uucp on the slave. The slave
should execute `uucp FROM TO'.
FROM
This is the name of the file or files on the slave which the
master wishes to transfer. Any wildcards are expanded on the
slave. If the master is requesting that the files be
transferred to itself, the request would normally contain
wildcard characters, since otherwise an `R' command would
suffice. The master can also use this command to request
that the slave transfer files to a third system.
TO
This is the name of the file or directory to which the files
should be transferred. This will normally use a UUCP name.
For example, if the master wishes to receive the files
itself, it would use `master!path'.
USER
The name of the user who requested the transfer.
OPTIONS
A list of options to control the transfer. It is not clear
which, if any, options are supported by most UUCP packages.
The slave then responds with an `X' command response. FSUUCP does
not support `X' requests, and always responds with `XN'.
`XY'
The request was accepted, and the appropriate file transfer
commands have been queued up for later processing.
`XN'
The request was denied. No particular reason is given.
In either case, the master will then send another command.
The E Command
.............
master: `E FROM TO USER -OPTIONS TEMP MODE NOTIFY SIZE COMMAND'
The `E' command is only supported by Taylor UUCP 1.04 and up. It
is used to make an execution request without requiring a separate
`X.*' file. It is only used when the command to be executed
requires a single input file which is passed to it as standard
input.
All the fields have the same meaning as they do for an `S' command,
except for OPTIONS and COMMAND.
OPTIONS
A list of options to control the transfer. The following
options are defined (all options are single characters):
`C'
The file has been copied to the spool directory (the
master should use TEMP rather than FROM).
`c'
The file has not been copied to the spool directory
(this is the default).
`N'
No mail message should be sent, even if the command
fails. This is the equivalent of the `N' command in an
`X.*' file.
`Z'
A mail message should be sent if the command fails (this
is generally the default in any case). This is the
equivalent of the `Z' command in an `X.*' file.
`R'
Mail messages about the execution should be sent to the
address in the NOTIFY field. This is the equivalent of
the `R' command in an `X.*' file.
`e'
The execution should be done with `/bin/sh'. This is the
equivalent of the `e' command in an `X.*' file.
COMMAND
The command which should be executed. This is the equivalent
of the `C' command in an `X.*' file.
The slave then responds with an `E' command response. These are
the same as the `S' command responses, but the initial character is
`E' rather than `S'.
If the slave responds with `EY', the file transfer begins. When
the file transfer is complete, the slave sends a `C' command
response, just as for the `S' command. After a successful file
transfer, the slave is responsible for arranging for the command
to be executed. The transferred file is passed as standard input,
as though it were named in the `I' and `F' commands of an `X.*'
file.
After the `C' command response has been received (in the `EY'
case) or immediately (in an `EN' case) the master will send another
command.
The H Command
.............
master: `H'
This is used by the master to hang up the connection. The slave
will respond with an `H' command response.
`HY'
The slave agrees to hang up the connection. In this case the
master sends another `HY' command. In some UUCP packages the
slave will then send a third `HY' command. At this point the
protocol is shut down, and the final handshake is begun.
`HN'
The slave does not agree to hang up. In this case the master
and the slave exchange roles. The next command will be sent
by the former slave, which is the new master. The roles may
be reversed several times during a single connection.
The Final Handshake
-------------------
After the protocol has been shut down, the final handshake is performed.
This handshake has no real purpose, and some UUCP packages simply drop
the connection rather than do it (in fact, some will drop the connection
immediately after both sides agree to hangup, without even closing down
the protocol).
caller: `\020OOOOOO\000'
called: `\020OOOOOOO\000'
That is, the calling UUCP sends six `O' characters and the called UUCP
replies with seven `O' characters. Some UUCP packages always send six
`O' characters.
------------------------------
From: UUCP `g' Protocol
Subject: UUCP `g' Protocol
UUCP `g' Protocol
=================
The `g' protocol is a packet based flow controlled error correcting
protocol that requires an eight bit clear connection. It is the
original UUCP protocol, and is supported by all UUCP implementations.
Many implementations of it are only able to support small window and
packet sizes, specifically a window size of 3 and a packet size of 64
bytes, but the protocol itself can support up to a window size of 7 and
a packet size of 4096 bytes. Complaints about the inefficiency of the
`g' protocol generally refer to specific implementations, rather than
to the correctly implemented protocol.
The `g' protocol was originally designed for general packet drivers,
and thus contains some features that are not used by UUCP, including an
alternate data channel and the ability to renegotiate packet and window
sizes during the communication session.
The `g' protocol is spoofed by many Telebit modems. When spoofing is
in effect, each Telebit modem uses the `g' protocol to communicate with
the attached computer, but the data between the modems is sent using a
Telebit proprietary error correcting protocol. This allows for very
high throughput over the Telebit connection, which, because it is
half-duplex, would not normally be able to handle the `g' protocol very
well at all. When a Telebit is spoofing the `g' protocol, it forces
the packet size to be 64 bytes and the window size to be 3.
This discussion of the `g' protocol explains how it works, but does not
discuss useful error handling techniques. Some discussion of this can
be found in Jamie E. Hanrahan's paper, cited above.
All `g' protocol communication is done with packets. Each packet
begins with a six byte header. Control packets consist only of the
header. Data packets contain additional data.
The header is as follows:
`\020'
Every packet begins with a `^P'.
K (1 <= K <= 9)
The K value is always 9 for a control packet. For a data packet,
the K value indicates how much data follows the six byte header.
The amount of data is 2 ** (K + 4), where ** indicates
exponentiation. Thus a K value of 1 means 32 data bytes and a K
value of 8 means 4096 data bytes. The K value for a data packet
must be between 1 and 8 inclusive.
checksum low byte
checksum high byte
The checksum value is described below.
control byte
The control byte indicates the type of packet, and is described
below.
xor byte
This byte is the xor of K, the checksum low byte, the checksum
high byte and the control byte (i.e., the second, third, fourth and
fifth header bytes). It is used to ensure that the header data is
valid.
The control byte in the header is composed of three bit fields, referred
to here as TT (two bits), XXX (three bits) and YYY (three bits). The
control is TTXXXYYY, or `(TT << 6) + (XXX << 3) + YYY'.
The TT field takes on the following values:
`0'
This is a control packet. In this case the K byte in the header
must be 9. The XXX field indicates the type of control packet;
these types are described below.
`1'
This is an alternate data channel packet. This is not used by
UUCP.
`2'
This is a data packet, and the entire contents of the attached data
field (whose length is given by the K byte in the header) are
valid. The XXX and YYY fields are described below.
`3'
This is a short data packet. Let the length of the data field (as
given by the K byte in the header) be L. Let the first byte in
the data field be B1. If B1 is less than 128 (if the most
significant bit of B1 is 0), then there are `L - B1' valid bytes
of data in the data field, beginning with the second byte. If `B1
>= 128', let B2 be the second byte in the data field. Then there
are `L - ((B1 & 0x7f) + (B2 << 7))' valid bytes of data in the
data field, beginning with the third byte. In all cases L bytes
of data are sent (and all data bytes participate in the checksum
calculation) but some of the trailing bytes may be dropped by the
receiver. The XXX and YYY fields are described below.
In a data packet (short or not) the XXX field gives the sequence number
of the packet. Thus sequence numbers can range from 0 to 7, inclusive.
The YYY field gives the sequence number of the last correctly received
packet.
Each communication direction uses a window which indicates how many
unacknowledged packets may be transmitted before waiting for an
acknowledgement. The window may range from 1 to 7, and may be different
in each direction. For example, if the window is 3 and the last packet
acknowledged was packet number 6, packet numbers 7, 0 and 1 may be sent
but the sender must wait for an acknowledgement before sending packet
number 2. This acknowledgement could come as the YYY field of a data
packet, or as the YYY field of a `RJ' or `RR' control packet (described
below).
Each packet must be transmitted in order (the sender may not skip
sequence numbers). Each packet must be acknowledged, and each packet
must be acknowledged in order.
In a control packet, the XXX field takes on the following values:
1 `CLOSE'
The connection should be closed immediately. This is typically
sent when one side has seen too many errors and wants to give up.
It is also sent when shutting down the protocol. If an unexpected
`CLOSE' packet is received, a `CLOSE' packet should be sent in
reply and the `g' protocol should halt, causing UUCP to enter the
final handshake.
2 `RJ' or `NAK'
The last packet was not received correctly. The YYY field
contains the sequence number of the last correctly received packet.
3 `SRJ'
Selective reject. The YYY field contains the sequence number of a
packet that was not received correctly, and should be
retransmitted. This is not used by UUCP, and most implementations
will not recognize it.
4 `RR' or `ACK'
Packet acknowledgement. The YYY field contains the sequence
number of the last correctly received packet.
5 `INITC'
Third initialization packet. The YYY field contains the maximum
window size to use.
6 `INITB'
Second initialization packet. The YYY field contains the packet
size to use. It requests a size of 2 ** (YYY + 5). Note that
this is not the same coding used for the K byte in the packet
header (it is 1 less). Most UUCP implementations that request a
packet size larger than 64 bytes can handle any packet size up to
that specified.
7 `INITA'
First initialization packet. The YYY field contains the maximum
window size to use.
To compute the checksum, call the control byte (the fifth byte in the
header) C.
The checksum of a control packet is simply `0xaaaa - C'.
The checksum of a data packet is `0xaaaa - (CHECK ^ C)', where `^'
denotes exclusive or, and CHECK is the result of the following routine
as run on the contents of the data field (every byte in the data field
participates in the checksum, even for a short data packet). Below is
the routine used by an early version of Taylor UUCP; it is a slightly
modified version of a routine which John Gilmore patched from G.L.
Chesson's original paper. The `z' argument points to the data and the
`c' argument indicates how much data there is.
int
igchecksum (z, c)
register const char *z;
register int c;
{
register unsigned int ichk1, ichk2;
ichk1 = 0xffff;
ichk2 = 0;
do
{
register unsigned int b;
/* Rotate ichk1 left. */
if ((ichk1 & 0x8000) == 0)
ichk1 <<= 1;
else
{
ichk1 <<= 1;
++ichk1;
}
/* Add the next character to ichk1. */
b = *z++ & 0xff;
ichk1 += b;
/* Add ichk1 xor the character position in the buffer counting from
the back to ichk2. */
ichk2 += ichk1 ^ c;
/* If the character was zero, or adding it to ichk1 caused an
overflow, xor ichk2 to ichk1. */
if (b == 0 || (ichk1 & 0xffff) < b)
ichk1 ^= ichk2;
}
while (--c > 0);
return ichk1 & 0xffff;
}
When the `g' protocol is started, the calling UUCP sends an `INITA'
control packet with the window size it wishes the called UUCP to use.
The called UUCP responds with an `INITA' packet with the window size it
wishes the calling UUCP to use. Pairs of `INITB' and `INITC' packets
are then similarly exchanged. When these exchanges are completed, the
protocol is considered to have been started.
Note that the window and packet sizes are not a negotiation. Each
system announces the window and packet size which the other system
should use. It is possible that different window and packet sizes will
be used in each direction. The protocol works this way on the theory
that each system knows how much data it can accept without getting
overrun. Therefore, each system tells the other how much data to send
before waiting for an acknowledgement.
When a UUCP package transmits a command, it sends one or more data
packets. All the data packets will normally be complete, although some
UUCP packages may send the last one as a short packet. The command
string is sent with a trailing null byte, to let the receiving package
know when the command is finished. Some UUCP packages require the last
byte of the last packet sent to be null, even if the command ends
earlier in the packet. Some packages may require all the trailing bytes
in the last packet to be null, but I have not confirmed this.
When a UUCP package sends a file, it will send a sequence of data
packets. The end of the file is signalled by a short data packet
containing zero valid bytes (it will normally be preceeded by a short
data packet containing the last few bytes in the file).
Note that the sequence numbers cover the entire communication session,
including both command and file data.
When the protocol is shut down, each UUCP package sends a `CLOSE'
control packet.
------------------------------
From: UUCP `f' Protocol
Subject: UUCP `f' Protocol
UUCP `f' Protocol
=================
The `f' protocol is a seven bit protocol which checksums an entire file
at a time. It only uses the characters between `\040' and `\176'
(ASCII `space' and `~') inclusive, as well as the carriage return
character. It can be very efficient for transferring text only data,
but it is very inefficient at transferring eight bit data (such as
compressed news). It is not flow controlled, and the checksum is
fairly insecure over large files, so using it over a serial connection
requires handshaking (XON/XOFF can be used) and error correcting
modems. Some people think it should not be used even under those
circumstances.
I believe that the `f' protocol originated in BSD versions of UUCP. It
was originally intended for transmission over X.25 PAD links.
The `f' protocol has no startup or finish protocol. However, both
sides typically sleep for a couple of seconds before starting up,
because they switch the terminal into XON/XOFF mode and want to allow
the changes to settle before beginning transmission.
When a UUCP package transmits a command, it simply sends a string
terminated by a carriage return.
When a UUCP package transmits a file, each byte B of the file is
translated according to the following table:
0 <= B <= 037: 0172, B + 0100 (0100 to 0137)
040 <= B <= 0171: B ( 040 to 0171)
0172 <= B <= 0177: 0173, B - 0100 ( 072 to 077)
0200 <= B <= 0237: 0174, B - 0100 (0100 to 0137)
0240 <= B <= 0371: 0175, B - 0200 ( 040 to 0171)
0372 <= B <= 0377: 0176, B - 0300 ( 072 to 077)
That is, a byte between `\040' and `\171' inclusive is transmitted as
is, and all other bytes are prefixed and modified as shown.
When all the file data is sent, a seven byte sequence is sent: two bytes
of `\176' followed by four ASCII bytes of the checksum as printed in
base 16 followed by a carriage return. For example, if the checksum
was 0x1234, this would be sent: `\176\1761234\r'.
The checksum is initialized to 0xffff. For each byte that is sent it is
modified as follows (where B is the byte before it has been transformed
as described above):
/* Rotate the checksum left. */
if ((ichk & 0x8000) == 0)
ichk <<= 1;
else
{
ichk <<= 1;
++ichk;
}
/* Add the next byte into the checksum. */
ichk += B;
When the receiving UUCP sees the checksum, it compares it against its
own calculated checksum and replies with a single character followed by
a carriage return.
`G'
The file was received correctly.
`R'
The checksum did not match, and the file should be resent from the
beginning.
`Q'
The checksum did not match, but too many retries have occurred and
the communication session should be abandoned.
The sending UUCP checks the returned character and acts accordingly.
------------------------------
From: UUCP `t' Protocol
Subject: UUCP `t' Protocol
UUCP `t' Protocol
=================
The `t' protocol is intended for use on links which provide reliable
end-to-end connections, such as TCP. It does no error checking or flow
control, and requires an eight bit clear channel.
I believe the `t' protocol originated in BSD versions of UUCP.
When a UUCP package transmits a command, it first gets the length of the
command string, C. It then sends `((C / 512) + 1) * 512' bytes (the
smallest multiple of 512 which can hold C bytes plus a null byte)
consisting of the command string itself followed by trailing null bytes.
When a UUCP package sends a file, it sends it in blocks. Each block
contains at most 1024 bytes of data. Each block consists of four bytes
containing the amount of data in binary (most significant byte first,
the same format as used by the Unix function `htonl') followed by that
amount of data. The end of the file is signalled by a block containing
zero bytes of data.
------------------------------
From: UUCP `e' Protocol
Subject: UUCP `e' Protocol
UUCP `e' Protocol
=================
The `e' protocol is similar to the `t' protocol. It does no flow
control or error checking and is intended for use over networks
providing reliable end-to-end connections, such as TCP.
The `e' protocol originated in versions of HDB UUCP.
When a UUCP package transmits a command, it simply sends the command as
an ASCII string terminated by a null byte.
When a UUCP package transmits a file, it sends the complete size of the
file as an ASCII decimal number. The ASCII string is padded out to 20
bytes with null bytes (i.e. if the file is 1000 bytes long, it sends
`1000\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0\0'). It then sends the entire file.
------------------------------
From: UUCP `G' Protocol
Subject: UUCP `G' Protocol
UUCP `G' Protocol
=================
The `G' protocol is used by SVR4 UUCP. It is identical to the `g'
protocol, except that it is possible to modify the window and packet
sizes. The SVR4 implementation of the `g' protocol reportedly is fixed
at a packet size of 64 and a window size of 7. Supposedly SVR4 chose
to implement a new protocol using a new letter to avoid any potential
incompatibilities when using different packet or window sizes.
Most implementations of the `g' protocol that accept packets larger
than 64 bytes will also accept packets smaller than whatever they
requested in the `INITB' packet. The SVR4 `G' implementation is an
exception; it will only accept packets of precisely the size it
requests in the INITB packet.
------------------------------
From: UUCP `i' Protocol
Subject: UUCP `i' Protocol
UUCP `i' Protocol
=================
The `i' protocol was written by Ian Lance Taylor (who also wrote this
FAQ). It was first used by Taylor UUCP version 1.04.
It is a sliding window packet protocol, like the `g' protocol, but it
supports bidirectional transfers (i.e., file transfers in both
directions simultaneously). It requires an eight bit clear connection.
Several ideas for the protocol were taken from the paper `A
High-Throughput Message Transport System' by P. Lauder. I don't know
where the paper was published, but the author's e-mail address is
`piers@cs.su.oz.au'. The `i' protocol does not adopt his main idea,
which is to dispense with windows entirely. This is because some links
still do require flow control and, more importantly, because using
windows sets a limit to the amount of data which the protocol must be
able to resend upon request. To reduce the costs of window
acknowledgements, the protocol uses a large window and only requires an
ack at the halfway point.
Each packet starts with a six byte header, optionally followed by data
bytes with a four byte checksum. There are currently five defined
packet types (`DATA', `SYNC', `ACK', `NAK', `SPOS', `CLOSE') which are
described below. Although any packet type may include data, any data
provided with an `ACK', `NAK' or `CLOSE' packet is ignored.
Every `DATA', `SPOS' and `CLOSE' packet has a sequence number. The
sequence numbers are independent for each side. The first packet sent
by each side is always number 1. Each packet is numbered one greater
than the previous packet, modulo 32.
Every packet has a local channel number and a remote channel number.
For all packets at least one channel number is zero. When a UUCP
command is sent to the remote system, it is assigned a non-zero local
channel number. All packets associated with that UUCP command sent by
the local system are given the selected local channel number. All
associated packets sent by the remote system are given the selected
number as the remote channel number. This permits each UUCP command to
be uniquely identified by the channel number on the originating system,
and therefore each UUCP package can associate all file data and UUCP
command responses with the appropriate command. This is a requirement
for bidirectional UUCP transfers.
The protocol maintains a single global file position, which starts at 0.
For each incoming packet, any associated data is considered to occur at
the current file position, and the file position is incremented by the
amount of data contained. The exception is a packet of type `SPOS',
which is used to change the file position. The reason for keeping
track of the file position is described below.
The header is as follows:
`\007'
Every packet begins with `^G'.
`(PACKET << 3) + LOCCHAN'
The five bit packet number combined with the three bit local
channel number. `DATA', `SPOS' and `CLOSE' packets use the packet
sequence number for the PACKET field. `NAK' packet types use the
PACKET field for the sequence number to be resent. `ACK' and
`SYNC' do not use the PACKET field, and generally leave it set to
0. Packets which are not associated with a UUCP command from the
local system use a local channel number of 0.
`(ACK << 3) + REMCHAN'
The five bit packet acknowledgement combined with the three bit
remote channel number. The packet acknowledgement is the number
of the last packet successfully received; it is used by all packet
types. Packets which are not sent in response to a UUCP command
from the remote system use a remote channel number of 0.
`(TYPE << 5) + (CALLER << 4) + LEN1'
The three bit packet type combined with the one bit packet
direction combined with the upper four bits of the data length.
The packet direction bit is always 1 for packets sent by the
calling UUCP, and 0 for packets sent by the called UUCP. This
prevents confusion caused by echoed packets.
LEN2
The lower eight bits of the data length. The twelve bits of data
length permit packets ranging in size from 0 to 4095 bytes.
CHECK
The exclusive or of the second through fifth bytes of the header.
This provides an additional check that the header is valid.
If the data length is non-zero, the packet is immediately followed by
the specified number of data bytes. The data bytes are followed by a
four byte CRC 32 checksum, with the most significant byte first. The
CRC is calculated over the contents of the data field.
The defined packet types are as follows:
0 `DATA'
This is a plain data packet.
1 `SYNC'
`SYNC' packets are exchanged when the protocol is initialized, and
are described further below. `SYNC' packets do not carry sequence
numbers (that is, the PACKET field is ignored).
2 `ACK'
This is an acknowledgement packet. Since `DATA' packets also carry
packet acknowledgements, `ACK' packets are only used when one side
has no data to send. `ACK' packets do not carry sequence numbers.
3 `NAK'
This is a negative acknowledgement. This is sent when a packet is
received incorrectly, and means that the packet number appearing
in the PACKET field must be resent. `NAK' packets do not carry
sequence numbers (the PACKET field is already used).
4 `SPOS'
This packet changes the file position. The packet contains four
bytes of data holding the file position, most significant byte
first. The next packet received will be considered to be at the
named file position.
5 `CLOSE'
When the protocol is shut down, each side sends a `CLOSE' packet.
This packet does have a sequence number, which could be used to
ensure that all packets were correctly received (this is not
needed by UUCP, however, which uses the higher level `H' command
with an `HY' response).
When the protocol starts up, both systems send a `SYNC' packet. The
`SYNC' packet includes at least three bytes of data. The first two
bytes are the maximum packet size the remote system should send, most
significant byte first. The third byte is the window size the remote
system should use. The remote system may send packets of any size up
to the maximum. If there is a fourth byte, it is the number of
channels the remote system may use (this must be between 1 and 7,
inclusive). Additional data bytes may be defined in the future.
The window size is the number of packets that may be sent before a
packet is acknowledged. There is no requirement that every packet be
acknowledged; any acknowledgement is considered to acknowledge all
packets through the number given. In the current implementation, if one
side has no data to send, it sends an `ACK' when half the window is
received.
Note that the `NAK' packet corresponds to the unused `g' protocol `SRJ'
packet type, rather than to the `RJ' packet type. When a `NAK' is
received, only the named packet should be resent, not any subsequent
packets.
Note that if both sides have data to send, but a packet is lost, it is
perfectly reasonable for one side to continue sending packets, all of
which will acknowledge the last packet correctly received, while the
system whose packet was lost will be unable to send a new packet because
the send window will be full. In this circumstance, neither side will
time out and one side of the communication will be effectively shut down
for a while. Therefore, any system with outstanding unacknowledged
packets should arrange to time out and resend a packet even if data is
being received.
Commands are sent as a sequence of data packets with a non-zero local
channel number. The last data packet for a command includes a trailing
null byte (normally a command will fit in a single data packet). Files
are sent as a sequence of data packets ending with one of length zero.
The channel numbers permit a more efficient implementation of the UUCP
file send command. Rather than send the command and then wait for the
`SY' response before sending the file, the file data is sent beginning
immediately after the `S' command is sent. If an `SN' response is
received, the file send is aborted, and a final data packet of length
zero is sent to indicate that the channel number may be reused. If an
`SY' reponse with a file position indicator is received, the file send
adjusts to the file position; this is why the protocol maintains a
global file position.
Note that the use of channel numbers means that each UUCP system may
send commands and file data simultaneously. Moreover, each UUCP system
may send multiple files at the same time, using the channel number to
disambiguate the data. Sending a file before receiving an
acknowledgement for the previous file helps to eliminate the round trip
delays inherent in other UUCP protocols.
------------------------------
From: UUCP `j' Protocol
Subject: UUCP `j' Protocol
UUCP `j' Protocol
=================
The `j' protocol is a variant of the `i' protocol. It was also written
by Ian Lance Taylor, and first appeared in Taylor UUCP version 1.04.
The `j' protocol is a version of the `i' protocol designed for
communication links which intercept a few characters, such as XON or
XOFF. It is not efficient to use it on a link which intercepts many
characters, such as a seven bit link. The `j' protocol performs no
error correction or detection; that is presumed to be the responsibility
of the `i' protocol.
When the `j' protocol starts up, each system sends a printable ASCII
string indicating which characters it wants to avoid using. The string
begins with the ASCII character `^' (octal 136) and ends with the ASCII
character `~' (octal 176). After sending this string, each system
looks for the corresponding string from the remote system. The strings
are composed of escape sequences: `\ooo', where `o' is an octal digit.
For example, sending the string `^\021\023~' means that the ASCII XON
and XOFF characters should be avoided. The union of the characters
described in both strings (the string which is sent and the string
which is received) is the set of characters which must be avoided in
this conversation. Avoiding a printable ASCII character (octal 040 to
octal 176, inclusive) is not permitted.
After the exchange of characters to avoid, the normal `i' protocol
start up is done, and the rest of the conversation uses the normal `i'
protocol. However, each `i' protocol packet is wrapped to become a `j'
protocol packet.
Each `j' protocol packet consists of a seven byte header, followed by
data bytes, followed by index bytes, followed by a one byte trailer.
The packet header looks like this:
`^'
Every packet begins with the ASCII character `^', octal 136.
HIGH
LOW
These two characters give the total number of bytes in the packet.
Both HIGH and LOW are printable ASCII characters. The length of
the packet is `(HIGH - 040) * 0100 + (LOW - 040)', where `040 <=
HIGH < 0177' and `040 <= LOW < 0140'. This permits a length of
6079 bytes, but there is a further restriction on packet size
described below.
`='
The ASCII character `=', octal 075.
DATA-HIGH
DATA-LOW
These two characters give the total number of data bytes in the
packet. The encoding is as described for HIGH and LOW. The number
of data bytes is the size of the `i' protocol packet wrapped inside
this `j' protocol packet.
`@'
The ASCII character `@', octal 100.
The header is followed by the number of data bytes given in DATA-HIGH
and DATA-LOW. These data bytes are the `i' protocol packet which is
being wrapped in the `j' protocol packet. However, each character in
the `i' protocol packet which the `j' protocol must avoid is
transformed into a printable ASCII character (recall that avoiding a
printable ASCII character is not permitted). Two index bytes are used
for each character which must be transformed.
The index bytes immediately follow the data bytes. The index bytes are
created in pairs. Each pair of index bytes encodes the location of a
character in the `i' protocol packet which was transformed to become a
printable ASCII character. Each pair of index bytes also encodes the
precise transformation which was performed.
When the sender finds a character which must be avoided, it will
transform it using one or two operations. If the character is 0200 or
greater, it will subtract 0200. If the resulting character is less than
020, or is equal to 0177, it will xor by 020. The result is a printable
ASCII character.
The zero based byte index of the character within the `i' protocol
packet is determined. This index is turned into a two byte printable
ASCII index, INDEX-HIGH and INDEX-LOW, such that the index is
`(INDEX-HIGH - 040) * 040 + (INDEX-LOW - 040)'. INDEX-LOW is
restricted such that `040 <= INDEX-LOW < 0100'. INDEX-HIGH is not
permitted to be 0176, so `040 <= INDEX-HIGH < 0176'. INDEX-LOW is then
modified to encode the transformation:
* If the character transformation only had to subtract 0200, then
INDEX-LOW is used as is.
* If the character transformation only had to xor by 020, then 040
is added to INDEX-LOW.
* If both operations had to be performed, then 0100 is added to
INDEX-LOW. However, if the value of INDEX-LOW was initially 077,
then adding 0100 would result in 0177, which is not a printable
ASCII character. For that special case, INDEX-HIGH is set to
0176, and INDEX-LOW is set to the original value of INDEX-HIGH.
The receiver decodes the index bytes as follows (this is the reverse of
the operations performed by the sender, presented here for additional
clarity):
* The first byte in the index is INDEX-HIGH, and the second is
INDEX-LOW.
* If `040 <= INDEX-HIGH < 0176', the index refers to the data byte
at position `(INDEX-HIGH - 040) * 040 + INDEX-LOW % 040'.
* If `040 <= INDEX-LOW < 0100', then 0200 must be added to indexed
byte.
* If `0100 <= INDEX-LOW < 0140', then 020 must be xor'ed to the
indexed byte.
* If `0140 <= INDEX-LOW < 0177', then 0200 must be added to the
indexed byte, and 020 must be xor'ed to the indexed byte.
* If `INDEX-HIGH == 0176', the index refers to the data byte at
position `(INDEX-LOW - 040) * 040 + 037'. 0200 must be added to
the indexed byte, and 020 must be xor'ed to the indexed byte.
This means the largest `i' protocol packet which may be wrapped inside
a `j' protocol packet is `(0175 - 040) * 040 + (077 - 040) == 3007'
bytes.
The final character in a `j' protocol packet, following the index
bytes, is the ASCII character `~' (octal 176).
The motivation behind using an indexing scheme, rather than escape
characters, is to avoid data movement. The sender may simply add a
header and a trailer to the `i' protocol packet. Once the receiver has
loaded the `j' protocol packet, it may scan the index bytes,
transforming the data bytes, and then pass the data bytes directly on to
the `i' protocol routine.
------------------------------
From: UUCP `x' Protocol
Subject: UUCP `x' Protocol
UUCP `x' Protocol
=================
The `x' protocol is used in Europe (and probably elsewhere) with
machines that contain an builtin X.25 card and can send eight bit data
transparently across X.25 circuits, without interference from the X.28
or X.29 layers. The protocol sends packets of 512 bytes, and relies on
a write of zero bytes being read as zero bytes without stopping
communication. It first appeared in the original System V UUCP
implementation.
------------------------------
From: UUCP `y' Protocol
Subject: UUCP `y' Protocol
UUCP `y' Protocol
=================
The `y' protocol was developed by Jorge Cwik for use in FX UUCICO, a PC
uucico program. It is designed for communication lines which handle
error correction and flow control. It requires an eight bit clean
connection. It performs error detection, but not error correction:
when an error is detected, the line is dropped. It is a streaming
protocol, like the `f' protocol; there are no packet acknowledgements,
so the protocol is efficient over a half-duplex communication line such
as PEP.
Every packet contains a six byte header:
sequence low byte
sequence high byte
A two byte sequence number, in little endian order. The first
sequence number is 0. Since the first packet is always a sync
packet (described below) the sequence number of the first data
packet is always 1. Each system counts sequence numbers
independently.
length low byte
length high byte
A two byte data length, in little endian order. If the high bit
of the sixteen bit field is clear, this is the number of data
bytes which follow the six byte header. If the high bit is set,
there is no data, and the length field is a type of control packet.
checksum low byte
checksum high byte
A two byte checksum, in little endian order. The checksum is
computed over the data bytes. The checksum algorithm is described
below. If there are no data bytes, the checksum is sent as 0.
When the protocol starts up, each side must send a sync packet. This is
a packet with a normal six byte header followed by data. The sequence
number of the sync packet should be 0. Currently at least four bytes of
data must be sent with the sync packet. Additional bytes should be
ignored. They are defined as follows:
version
The version number of the protocol. Currently this must be 1.
Larger numbers should be ignored; it is the responsibility of the
newer version to accommodate the older one.
packet size
The maximum data length to use divided by 256. This is sent as a
single byte. The maximum data length permitted is 32768, which
would be sent as 128. Customarily both systems will use the same
maximum data length, the lower of the two requested.
flags low byte
flags high byte
Two bytes of flags. None are currently defined. These bytes
should be sent as 0, and ignored by the receiver.
A length field with the high bit set is a control packet. The
following control packet types are defined:
0xfffe `YPKT_ACK'
Acknowledges correct receipt of a file.
0xfffd `YPKT_ERR'
Indicates an incorrect checksum.
0xfffc `YPKT_BAD'
Indicates a bad sequence number, an invalid length, or some other
error.
If a control packet other than `YPKT_ACK' is received, the connection
is dropped. If a checksum error is detected for a received packet, a
`YPKT_ERR' control packet is sent, and the connection is dropped. If a
packet is received out of sequence, a `YPKT_BAD' control packet is
sent, and the connection is dropped.
The checksum is initialized to 0xffff. For each data byte in a packet
it is modified as follows (where B is the byte before it has been
transformed as described above):
/* Rotate the checksum left. */
if ((ichk & 0x8000) == 0)
ichk <<= 1;
else
{
ichk <<= 1;
++ichk;
}
/* Add the next byte into the checksum. */
ichk += B;
This is the same algorithm as that used by the `f' protocol.
A command is sent as a sequence of data packets followed by a null byte.
In the normal case, a command will fit into a single packet. The packet
should be exactly the length of the command plus a null byte. If the
command is too long, more packets are sent as required.
A file is sent as a sequence of data packets, ending with a zero length
packet. The data packets may be of any length greater than zero and
less than or equal to the maximum permitted packet size specified in the
initial sync packet.
After the zero length packet ending a file transfer has been received,
the receiving system sends a `YPKT_ACK' control packet. The sending
system waits for the `YPKT_ACK' control packet before continuing; this
wait should be done with a large timeout, since there may be a
considerable amount of data buffered on the communication path.
------------------------------
From: UUCP `d' Protocol
Subject: UUCP `d' Protocol
UUCP `d' Protocol
=================
The `d' protocol is apparently used for DataKit muxhost (not RS-232)
connections. No file size is sent. When a file has been completely
transferred, a write of zero bytes is done; this must be read as zero
bytes on the other end.
------------------------------
From: UUCP `h' Protocol
Subject: UUCP `h' Protocol
UUCP `h' Protocol
=================
The `h' protocol is apparently used in some places with HST modems. It
does no error checking, and is not that different from the `t'
protocol. I don't know the details.
------------------------------
From: UUCP `v' Protocol
Subject: UUCP `v' Protocol
UUCP `v' Protocol
=================
The `v' protocol is used by UUPC/extended, a PC UUCP program. It is
simply a version of the `g' protocol which supports packets of any
size, and also supports sending packets of different sizes during the
same conversation. There are many `g' protocol implementations which
support both, but there are also many which do not. Using `v' ensures
that everything is supported.
------------------------------
From: Thanks
Subject: Thanks
Besides the papers and information acknowledged at the top of this
article, the following people have contributed help, advice,
suggestions and information:
Earle Ake 513-429-6500 <ake@Dayton.SAIC.COM>
chris@uuplus.com (Christopher J. Ambler)
jhc@iscp.bellcore.com (Jonathan Clark)
jorge@laser.satlink.net (Jorge Cwik)
celit!billd@UCSD.EDU (Bill Davidson)
"Drew Derbyshire" <ahd@kew.com>
erik@pdnfido.fidonet.org
Matthew Farwell <dylan@ibmpcug.co.uk>
dgilbert@gamiga.guelphnet.dweomer.org (David Gilbert)
kherron@ms.uky.edu (Kenneth Herron)
Mike Ipatow <mip@fido.itc.e-burg.su>
Romain Kang <romain@pyramid.com>
"Jonathan I. Kamens" <jik@GZA.COM>
"David J. MacKenzie" <djm@eng.umd.edu>
jum@helios.de (Jens-Uwe Mager)
peter@xpoint.ruessel.sub.org (Peter Mandrella)
david nugent <david@csource.oz.au>
Stephen.Page@prg.oxford.ac.uk
joey@tessi.UUCP (Joey Pruett)
James Revell <revell@uunet.uu.net>
Larry Rosenman <ler@lerami.lerctr.org>
Rich Salz <rsalz@bbn.com>
evesg@etlrips.etl.go.jp (Gjoen Stein)
kls@ditka.Chicago.COM (Karl Swartz)
Dima Volodin <dvv@hq.demos.su>
John.Woods@proteon.com (John Woods)
jon@console.ais.org (Jon Zeeff)
Eric Ziegast <ziegast@uunet.uu.net>
------------------------------
End of UUCP Internals Frequently Asked Questions
******************************
--
Ian Taylor | ian@airs.com | First to identify quote wins free e-mail message:
``According to [this theory], when [light] reaches the retina, it beats on it,
and, for example, 483B beats a second give red, 727B beats violet, and so on.
So those who are color-blind are those who cannot count the beats, I suppose!''
|
Закладки на сайте Проследить за страницей |
Created 1996-2024 by Maxim Chirkov Добавить, Поддержать, Вебмастеру |