<ian@airs.com>
This package is covered by the GNU Public License. See the file `COPYING' for details. If you would like to do something with this package that you feel is reasonable, but you feel is prohibited by the license, contact me to see if we can work it out.
The rest of this section is some descriptive text from the Free Software Foundation.
All the programs, scripts and documents relating to Taylor UUCP are free; this means that everyone is free to use them and free to redistribute them on a free basis. The Taylor UUCP-related programs are not in the public domain; they are copyrighted and there are restrictions on their distribution, but these restrictions are designed to permit everything that a good cooperating citizen would want to do. What is not allowed is to try to prevent others from further sharing any version of these programs that they might get from you.
Specifically, we want to make sure that you have the right to give away copies of the programs that relate to Taylor UUCP, that you receive source code or else can get it if you want it, that you can change these programs or use pieces of them in new free programs, and that you know you can do these things.
To make sure that everyone has such rights, we have to forbid you to deprive anyone else of these rights. For example, if you distribute copies of the Taylor UUCP related programs, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must tell them their rights.
Also, for our own protection, we must make certain that everyone finds out that there is no warranty for the programs that relate to Taylor UUCP. If these programs are modified by someone else and passed on, we want their recipients to know that what they have is not what we distributed, so that any problems introduced by others will not reflect on our reputation.
The precise conditions of the licenses for the programs currently being distributed that relate to Taylor UUCP are found in the General Public Licenses that accompany them.
General introductions to UUCP are available, and perhaps one day I will write one. In the meantime, here is a very brief one that concentrates on the programs provided by Taylor UUCP.
Taylor UUCP is a complete UUCP package. It is covered by the GNU Public License, which means that the source code is always available. It is composed of several programs; most of the names of these programs are based on earlier UUCP packages.
uucp
uucp
program is used to copy file between systems. It is
similar to the standard Unix cp
program, except that you can
refer to a file on a remote system by using `system!' before the
file name. For example, to copy the file `notes.txt' to the system
`airs', you would say `uucp notes.txt airs!~/notes.txt'. In
this example `~' is used to name the UUCP public directory on
`airs'. For more details, see section Invoking uucp.
uux
uux
program is used to request the execution of a program on
a remote system. This is how mail and news are transferred over UUCP.
As with uucp
, programs and files on remote systems may be named
by using `system!'. For example, to run the rnews
program
on `airs', passing it standard input, you would say `uux -
airs!rnews'. The `-' means to read standard input and set things
up such that when rnews
runs on `airs' it will receive the
same standard input. For more details, see section Invoking uux.
Neither uucp
nor uux
actually do any work immediately.
Instead, they queue up requests for later processing. They then start a
daemon process which processes the requests and calls up the appropriate
systems. Normally the system will also start the daemon periodically to
check if there is any work to be done. The advantage of this approach
is that it all happens automatically. You don't have to sit around
waiting for the files to be transferred. The disadvantage is that if
anything goes wrong it might be a while before anybody notices.
uustat
uustat
program does many things. By default it will simply
list all the jobs you have queued with uucp
or uux
that
have not yet been processed. You can use uustat
to remove any of
your jobs from the queue. You can also it use it to show the status of
the UUCP system in various ways, such as showing the connection status
of all the remote systems your system knows about. The system
administrator can use uustat
to automatically discard old jobs
while sending mail to the user who requested them. For more details,
see section Invoking uustat.
uuname
uuname
program by default lists all the remote systems your
system knows about. You can also use it to get the name of your local
system. It is mostly useful for shell scripts. For more details, see
section Invoking uuname.
uulog
uulog
program can be used to display entries in the UUCP log
file. It can select the entries for a particular system or a particular
user. You can use it to see what has happened to your queued jobs in
the past. For more details, see section Invoking uulog.
uuto
uupick
uuto
is a simple shell script interface to uucp
. It will
transfer a file, or the contents of a directory, to a remote system, and
notify a particular user on the remote system when it arrives. The
remote user can then retrieve the file(s) with uupick
. For more
details, see section Invoking uuto, and see section Invoking uupick.
cu
cu
program can be used to call up another system and
communicate with it as though you were directly connected. It can also
do simple file transfers, though it does not provide any error checking.
For more details, section Invoking cu.
These eight programs just described, uucp
, uux
,
uuto
, uupick
, uustat
, uuname
, uulog
,
and cu
are the user programs provided by Taylor UUCP.
uucp
, uux
, and uuto
add requests to the work queue,
uupick
extracts files from the UUCP public directory,
uustat
examines the work queue, uuname
examines the
configuration files, uulog
examines the log files, and cu
just uses the UUCP configuration files.
The real work is actually done by two daemon processes, which are normally run automatically rather than by a user.
uucico
uucico
daemon is the program which actually calls the remote
system and transfers files and requests. uucico
is normally
started automatically by uucp
and uux
. Most systems will
also start it periodically to make sure that all work requests are
handled. uucico
checks the queue to see what work needs to be
done, and then calls the appropriate systems. If the call fails,
perhaps because the phone line is busy, uucico
leaves the
requests in the queue and goes on to the next system to call. It is
also possible to force uucico
to call a remote system even if
there is no work to be done for it, so that it can pick up any work that
may be queued up remotely. For more details, see section Invoking uucico.
uuxqt
uuxqt
daemon processes execution requests made by the
uux
program on remote systems. It also processes requests made
on the local system which require files from a remote system. It is
normally started by uucico
. For more details, see section Invoking uuxqt.
Suppose you, on the system `bantam', want to copy a file to the
system `airs'. You would run the uucp
command locally, with
a command like `uucp notes.txt airs!~/notes.txt'. This would queue
up a request on `bantam' for `airs', and would then start the
uucico
daemon. uucico
would see that there was a request
for `airs' and attempt to call it. When the call succeeded,
another copy of uucico
would be started on `airs'. The two
copies of uucico
would tell each other what they had to do and
transfer the file from `bantam' to `airs'. When the file
transfer was complete the uucico
on `airs' would move it
into the UUCP public directory.
UUCP is often used to transfer mail. This is normally done
automatically by mailer programs. When `bantam' has a mail message
to send to `ian' at `airs', it executes `uux - airs!rmail
ian' and writes the mail message to the uux
process as standard
input. The uux
program, running on `bantam', will read the
standard input and store it, as well as the rmail
request itself,
on the work queue for `airs'. uux
will then start the
uucico
daemon. The uucico
daemon will call up
`airs', just as in the uucp
example, and transfer the work
request and the mail message. The uucico
daemon on `airs'
will put the files on a local work queue. When the communication
session is over, the uucico
daemon on `airs' will start the
uuxqt
daemon. uuxqt
will see the request on the work
queue, and will run `rmail ian' with the mail message as standard
input. The rmail
program, which is not part of the UUCP package,
is then responsible for either putting the message in the right mailbox
on `airs' or forwarding the message on to another system.
Taylor UUCP comes with a few other programs that are useful when installing and configuring UUCP.
uuchk
uuchk
program reads the UUCP configuration files and displays
a rather lengthy description of what it finds. This is useful when
configuring UUCP to make certain that the UUCP package will do what you
expect it to do. For more details, see section Invoking uuchk.
uuconv
uuconv
program can be used to convert UUCP configuration
files from one format to another. This can be useful for administrators
converting from an older UUCP package. Taylor UUCP is able to read and
use old configuration file formats, but some new features can not be
selected using the old formats. For more details, see section Invoking uuconv.
uusched
uusched
script is provided for compatibility with older UUCP
releases. It starts uucico
to call, one at a time, all the
systems for which work has been queued. For more details, see
section Invoking uusched.
tstuu
tstuu
program is a test harness for the UUCP package; it can
help check that the package has been configured and compiled correctly.
However, it uses pseudo-terminals, which means that it is less portable
than the rest of the package. If it works, it can be useful when
initially installing Taylor UUCP. For more details, see section Testing the Compilation.
This chapter describes how to run the UUCP programs.
All of the UUCP programs support a few standard options.
debug
configuration command for details (see section Debugging Levels).
Multiple types may be given, separated by commas, and the `--debug'
option may appear multiple times. A number may also be given, which
will turn on that many types from the foregoing list; for example,
`--debug 2' is equivalent to `--debug abnormal,chat'. To turn
on all types of debugging, use `-x all'.
The uulog
program uses `-X' rather than `-x' to select
the debugging type; for uulog
, `-x' has a different meaning,
for reasons of historical compatibility.
uucp [options] `source-file' `destination-file' uucp [options] `source-file'... `destination-directory'
The uucp
command copies files between systems. Each `file'
argument is either a file name on the local machine or is of the form
`system!file'. The latter is interpreted as being on a remote
system.
When uucp
is used with two non-option arguments, the contents of
the first file are copied to the second. With more than two non-option
arguments, each source file is copied into the destination directory.
A file may be transferred to or from `system2' via `system1' by using `system1!system2!file'.
Any file name that does not begin with `/' or `~' will be prepended with the current directory (unless the `-W' or `--noexpand' options are used). For example, if you are in the directory `/home/ian', then `uucp foo remote!bar' is equivalent to `uucp /home/ian/foo remote!/home/ian/bar'. Note that the resulting file name may not be valid on a remote system.
A file name beginning with a simple `~' starts at the UUCP public
directory; a file name beginning with `~name' starts at the home
directory of the named user. The `~' is interpreted on the
appropriate system. Note that some shells will interpret an initial
`~' before uucp
sees it; to avoid this the `~' must be
quoted.
The shell metacharacters `?' `*' `[' and `]' are interpreted on the appropriate system, assuming they are quoted to prevent the shell from interpreting them first.
The file copy does not take place immediately, but is queued up for the
uucico
daemon; the daemon is started immediately unless the
`-r' or `--nouucico' option is given. The next time the
remote system is called, the file(s) will be copied. See section Invoking uucico.
The file mode is not preserved, except for the execute bit. The resulting file is owned by the uucp user.
The following options may be given to uucp
.
uucico
daemon, the copy
will fail. The files must be readable by the uucico
daemon, and
by the invoking user.
uucico
daemon immediately; merely queue up the
file transfer for later execution.
uustat
.
See section Invoking uustat.
It is possible for some complex operations to produce more than one
jobid, in which case each will be printed on a separate line. For
example
uucp sys1!~user1/file1 sys2!~user2/file2 ~user3will generate two separate jobs, one for the system `sys1' and one for the system `sys2'.
uuto
shell script; see section Invoking uuto. It causes uucp
to interpret the final argument as
`system!user'. The file(s) are sent to
`~/receive/user/local' on the remote system, where
user is from the final argument and local is the local UUCP
system name. Also, uucp
will act as though `--notify user'
were specified.
uux [options] command
The uux
command is used to execute a command on a remote system,
or to execute a command on the local system using files from remote
systems. The command is not executed immediately; the request is queued
until the uucico
daemon calls the system and transfers the
necessary files. The daemon is started automatically unless one of the
`-r' or `--nouucico' options is given.
The actual command execution is done by the uuxqt
daemon on the
appropriate system.
File arguments can be gathered from remote systems to the execution system, as can standard input. Standard output may be directed to a file on a remote system.
The command name may be preceded by a system name followed by an exclamation point if it is to be executed on a remote system. An empty system name is taken as the local system.
Each argument that contains an exclamation point is treated as naming a file. The system which the file is on is before the exclamation point, and the file name on that system follows it. An empty system name is taken as the local system; this form must be used to transfer a file to a command being executed on a remote system. If the file name is not absolute, the current working directory will be prepended to it; the result may not be meaningful on the remote system. A file name may begin with `~/', in which case it is relative to the UUCP public directory on the appropriate system. A file name may begin with `~name/', in which case it is relative to the home directory of the named user on the appropriate system.
Standard input and output may be redirected as usual; the file names
used may contain exclamation points to indicate that they are on remote
systems. Note that the redirection characters must be quoted so that
they are passed to uux
rather than interpreted by the shell.
Append redirection (`>>') does not work.
All specified files are gathered together into a single directory before execution of the command begins. This means that each file must have a distinct name. For example,
uux 'sys1!diff sys2!~user1/foo sys3!~user2/foo >!foo.diff'
will fail because both files will be copied to `sys1' and stored under the name `foo'.
Arguments may be quoted by parentheses to avoid interpretation of
exclamation points. This is useful when executing the uucp
command on a remote system.
Most systems restrict the commands which may be executed using `uux'. Many permit only the execution of `rmail' and `rnews'.
A request to execute an empty command (e.g., `uux sys!') will create a poll file for the specified system; see section Calling Other Systems for an example of why this might be useful.
The following options may be given to uux
.
uucico
daemon,
the copy will fail. The files must be readable by the uucico
daemon, as well as the by the invoker of uux
.
uucico
daemon, the changed versions will be used. The files must
be readable by the uucico
daemon, as well as by the invoker of
uux
.
uuxqt
daemons, including the Taylor UUCP uuxqt
, this is
the default action; for those, `--notification=error' will have no
effect. However, some uuxqt
daemons will send mail if the job
succeeds, unless the `--notification=error' option is used. Some
other uuxqt
daemons will not send mail even if the job fails,
unless the `--notification=error' option is used.
uucico
daemon immediately; merely queue up the
execution request for later processing.
uustat
. See section Invoking uustat. Cancelling any file copies
will make it impossible to complete execution of the job.
Here are some examples of using uux
.
uux -z - sys1!rmail user1
This will execute the command `rmail user1' on the system
`sys1', giving it as standard input whatever is given to uux
as standard input. If a failure occurs, mail will be sent to the user
who ran the command.
uux 'diff -c sys1!~user1/file1 sys2!~user2/file2 >!file.diff'
This will fetch the two named files from system `sys1' and system
`sys2' and execute `diff', putting the result in
`file.diff' in the current directory on the local system. The
current directory must be writable by the uuxqt
daemon for this
to work.
uux 'sys1!uucp ~user1/file1 (sys2!~user2/file2)'
Execute uucp
on the system `sys1' copying `file1' (on
system `sys1') to `sys2'. This illustrates the use of
parentheses for quoting.
uustat -a uustat --all uustat [-eKRiMNQ] [-sS system] [-uU user] [-cC command] [-oy hours] [-B lines] [--executions] [--kill-all] [--rejuvenate-all] [--prompt] [--mail] [--notify] [--no-list] [--system system] [--not-system system] [--user user] [--not-user user] [--command command] [--not-command command] [--older-than hours] [--younger-than hours] [--mail-lines lines] uustat [-kr jobid] [--kill jobid] [--rejuvenate jobid] uustat -q [-sS system] [-oy hours] [--system system] [--not-system system ] [--older-than hours] [--younger-than hours] uustat --list [-sS system] [-oy hours] [--system system ] [--not-system system] [--older-than hours] [--younger-than hours] uustat -m uustat --status uustat -p uustat --ps
The uustat
command can display various types of status
information about the UUCP system. It can also be used to cancel or
rejuvenate requests made by uucp
or uux
.
With no options, uustat
displays all jobs queued up for the
invoking user, as if given the `--user' option with the appropriate
argument.
If any of the `-a', `--all', `-e', `--executions', `-s', `--system', `-S', `--not-system', `-u', `--user', `-U', `--not-user', `-c', `--command', `-C', `--not-command', `-o', `--older-than', `-y', or `--younger-than' options are given, then all jobs which match the combined specifications are displayed.
The `-K' or `--kill-all' option may be used to kill off a selected group of jobs, such as all jobs more than 7 days old.
The following options may be given to uustat
.
uuxqt
rather than uucico
. Queued execution requests may be waiting for
some file to be transferred from a remote system. They are created by
an invocation of uux
.
uucp
or
uux
. A job may only be killed by the user who created the job,
or by the UUCP administrator, or the superuser. The `-k' or
`--kill' options may be used multiple times on the command line to
kill several jobs.
uucp
or uux
. A job may only be
rejuvenated by the user who created the job, or by the UUCP
administrator, or the superuser. The `-r' or `--rejuvenate'
options may be used multiple times on the command line to rejuvenate
several jobs.
uustat --all
Display status of all jobs. A sample output line is as follows:
bugsA027h bugs ian 04-01 13:50 Executing rmail ian@airs.com (sending 12 bytes)
The format is
jobid system user queue-date command (size)
The jobid may be passed to the `--kill' or `--rejuvenate' options. The size indicates how much data is to be transferred to the remote system, and is absent for a file receive request. The `--system', `--not-system', `--user', `--not-user', `--command', `--not-command', `--older-than', and `--younger-than' options may be used to control which jobs are listed.
uustat --executions
Display status of queued up execution requests. A sample output line is as follows:
bugs bugs!ian 05-20 12:51 rmail ian
The format is
system requestor queue-date command
The `--system', `--not-system', `--user', `--not-user', `--command', `--not-command', `--older-than', and `--younger-than' options may be used to control which requests are listed.
uustat --list
Display status for all systems with queued up commands. A sample output line is as follows:
bugs 4C (1 hour) 0X (0 secs) 04-01 14:45 Dial failed
This indicates the system, the number of queued commands, the age of the oldest queued command, the number of queued local executions, the age of the oldest queued execution, the date of the last conversation, and the status of that conversation.
uustat --status
Display conversation status for all remote systems. A sample output line is as follows:
bugs 04-01 15:51 Conversation complete
This indicates the system, the date of the last conversation, and the
status of that conversation. If the last conversation failed,
uustat
will indicate how many attempts have been made to call the
system. If the retry period is currently preventing calls to that
system, uustat
also displays the time when the next call will be
permitted.
uustat --ps
Display the status of all processes holding UUCP locks. The output
format is system dependent, as uustat
simply invokes ps
on
each process holding a lock.
uustat -c rmail -o 168 -K -Q -M -N -W "Queued for over 1 week"
This will kill all `rmail' commands that have been queued up waiting for delivery for over 1 week (168 hours). For each such command, mail will be sent both to the UUCP administrator and to the user who requested the rmail execution. The mail message sent will include the string given by the `-W' option. The `-Q' option prevents any of the jobs from being listed on the terminal, so any output from the program will be error messages.
uuname [-a] [--aliases] uuname -l uuname --local
By default, the uuname
program simply lists the names of all the
remote systems mentioned in the UUCP configuration files.
The uuname
program may also be used to print the UUCP name of the
local system.
The uuname
program is mainly for use by shell scripts.
The following options may be given to uuname
.
uulog [-#] [-n lines] [-sf system] [-u user] [-DSF] [--lines lines] [--system system] [--user user] [--debuglog] [--statslog] [--follow] [--follow=system]
The uulog
program may be used to display the UUCP log file.
Different options may be used to select which parts of the file to
display.
uulog
specifies the debugging
type using `-X' rather than the usual `-x'.
The operation of uulog
depends to some degree upon the type of
log files generated by the UUCP programs. This is a compile time
option. If the UUCP programs have been compiled to use HDB style log
files, uulog
changes in the following ways:
uuxqt
log file.
uuto files... system!user
The uuto
program may be used to conveniently send files to a
particular user on a remote system. It will arrange for mail to be sent
to the remote user when the files arrive on the remote system, and he or
she may easily retrieve the files using the uupick
program
(see section Invoking uupick). Note that uuto
does not provide any
security--any user on the remote system can examine the files.
The last argument specifies the system and user name to which to send the files. The other arguments are the files or directories to be sent.
The uuto
program is actually just a trivial shell script which
invokes the uucp
program with the appropriate arguments. Any
option which may be given to uucp
may also be given to
uuto
. See section Invoking uucp.
uupick [-s system] [--system system]
The uupick
program is used to conveniently retrieve files
transferred by the uuto
program.
For each file transferred by uuto
, uupick
will display the
source system, the file name, and whether the name refers to a regular
file or a directory. It will then wait for the user to specify an
action to take. One of the following commands must be entered:
uupick
.
The `-s' or `--system' option may be used to restrict
uupick
to only present files transferred from a particular
system. The uupick
program also supports the standard UUCP
program options; see section Standard Options.
cu [options] [system | phone | "dir"]
The cu
program is used to call up another system and act as a
dial in terminal. It can also do simple file transfers with no error
checking.
The cu
program takes a single non-option argument.
If the argument is the string `dir' cu will make a direct connection to the port. This may only be used by users with write access to the port, as it permits reprogramming the modem.
Otherwise, if the argument begins with a digit, it is taken to be a phone number to call.
Otherwise, it is taken to be the name of a system to call.
The `-z' or `--system' options may be used to name a system beginning with a digit, and the `-c' or `--phone' options may be used to name a phone number that does not begin with a digit.
The cu
program locates a port to use in the UUCP configuration
files. If a simple system name is given, it will select a port
appropriate for that system. The `-p', `--port', `-l',
`--line', `-s', and `--speed' options may be used to
control the port selection.
When a connection is made to the remote system, cu
forks into two
processes. One reads from the port and writes to the terminal, while
the other reads from the terminal and writes to the port.
The cu
program provides several commands that may be used during
the conversation. The commands all begin with an escape character,
which by default is ~ (tilde). The escape character is only
recognized at the beginning of a line. To send an escape character to
the remote system at the start of a line, it must be entered twice. All
commands are either a single character or a word beginning with %
(percent sign).
The cu
program recognizes the following commands.
cu
variable to the given value. If value is not given, the
variable is set to `true'.
cu
variable to `false'.
The cu
program also supports several variables. They may be
listed with the `~v' command, and set with the `~s' or
`~!' commands.
cu
will delay for a second, after
recognizing the escape character, before printing the name of the local
system. The default is true.
The following options may be given to cu
.
uucico [options]
The uucico
daemon processes file transfer requests queued by
uucp
and uux
. It is started when uucp
or
uux
is run (unless they are given the `-r' or
`--nouucico' options). It is also typically started periodically
using entries in the `crontab' table(s).
When uucico
is invoked with `-r1', `--master',
`-s', `--system', or `-S', the daemon will place a call
to a remote system, running in master mode. Otherwise the daemon will
start in slave mode, accepting a call from a remote system. Typically a
special login name will be set up for UUCP which automatically invokes
uucico
when a remote system calls in and logs in under that name.
When uucico
terminates, it invokes the uuxqt
daemon,
unless the `-q' or `--nouuxqt' options were given;
uuxqt
executes any work orders created by uux
on a remote
system, and any work orders created locally which have received remote
files for which they were waiting.
If a call fails, uucico
will normally refuse to retry the call
until a certain (configurable) amount of time has passed. This may be
overriden by the `-f', `--force', or `-S' options.
The `-l', `--prompt', `-e', or `--loop' options may
be used to force uucico
to produce its own prompts of
`login: ' and `Password:'. When another uucico
daemon
calls in, it will see these prompts and log in as usual. The login name
and password will normally be checked against a separate list kept
specially for uucico
, rather than the `/etc/passwd' file
(see section Configuration File Names). It is possible, on some systems, to
configure uucico
to use `/etc/passwd'. The `-l' or
`--prompt' options will prompt once and then exit; in this mode the
UUCP administrator, or the superuser, may use the `-u' or
`--login' option to force a login name, in which case uucico
will not prompt for one. The `-e' or `--loop' options will
prompt again after the first session is over; in this mode uucico
will permanently control a port.
If uucico
receives a SIGQUIT
, SIGTERM
or
SIGPIPE
signal, it will cleanly abort any current conversation
with a remote system and exit. If it receives a SIGHUP
signal it
will abort any current conversation, but will continue to place calls to
(if invoked with `-r1' or `--master') and accept calls from
(if invoked with `-e' or `--loop') other systems. If it
receives a SIGINT
signal it will finish the current conversation,
but will not place or accept any more calls.
The following options may be given to uucico
.
uucico
to be easily run from
inetd
. The login name and password are checked against the UUCP
password file, which need not be `/etc/passwd'. The `--login'
option may be used to force a login name, in which cause uucico
will only prompt for a password.
kill
to shut it down.
uuxqt
daemon when finished.
uustat
). This can be convenient
for automated polling scripts, which may want to simply attempt to call
every system rather than worry about which particular systems may be
called at the moment. This option also suppresses the log message
indicating that there is no work to be done.
uucico
detaches from the terminal before each call out to another system and
before invoking uuxqt
. This option prevents this.
uucico
to prompt only
for the password, not the login name.
uucico
to use TLI calls to perform I/O.
uuxqt [-c command] [-s system] [--command command] [--system system]
The uuxqt
daemon executes commands requested by uux
from
either the local system or from remote systems. It is started
automatically by the uucico
daemon (unless uucico
is given
the `-q' or `--nouuxqt' options).
There is normally no need to run uuxqt
, since it will be invoked
by uucico
. However, uuxqt
can be invoked directly to
provide greater control over the processing of the work queue.
Multiple invocations of uuxqt
may be run at once, as controlled
by the max-uuxqts
configuration command; see section Miscellaneous config File Commands.
The following options may be given to uuxqt
.
uuchk [-s system] [--system system]
The uuchk
program displays information read from the UUCP
configuration files. It should be used to ensure that UUCP has been
configured correctly.
The `-s' or `--system' options may be used to display the
configuration for just the specified system, rather than for all
systems. The uuchk
program also supports the standard UUCP
program options; see section Standard Options.
uuconv -i type -o type [-p program] [--program program] uuconv --input type --output type [-p program] [--program program]
The uuconv
program converts UUCP configuration files from one
format to another. The type of configuration file to read is specified
using the `-i' or `--input' options. The type of
configuration file to write is specified using the `-o' or
`--output' options.
The supported configuration file types are `taylor', `v2', and `hdb'. For a description of the `taylor' configuration files, see section Taylor UUCP Configuration Files. The other types of configuration files are used by traditional UUCP packages, and are not described in this manual.
An input configuration of type `v2' or `hdb' is read from a compiled in directory (specified by `oldconfigdir' in `Makefile'). An input configuration of type `taylor' is read from a compiled in directory by default, but may be overridden with the standard `-I' or `--config' options (see section Standard Options).
The output configuration is written to files in the directory in which
uuconv
is run.
Some information in the input files may not be representable in the
desired output format, in which case uuconv
will silently discard
it. The output of uuconv
should be carefully checked before it
is used. The uuchk
program may be used for this purpose; see
section Invoking uuchk.
The `-p' or `--program' option may be used to convert specific
cu
configuration information, rather than the default of only
converting the uucp
configuration information; see section The Main Configuration File.
The uuchk
program also supports the standard UUCP program
options; see section Standard Options.
The uusched
program is actually just a shell script which invokes
the uucico
daemon. It is provided for backward compatibility.
It causes uucico
to call all systems for which there is work.
Any option which may be given to uucico
may also be given to
uusched
. See section Invoking uucico.
These are the installation instructions for the Taylor UUCP package.
If you have a source code distribution, you must first compile it for your system. Free versions of Unix, such as Linux, NetBSD, or FreeBSD, often come with pre-compiled binary distributions of UUCP. If you are using a binary distribution, you may skip to the configuration section (see section Configuring Taylor UUCP).
Follow these steps to compile the source code.
uucp
rather than a real person; they should probably
not be owned by root
).
configure
. This script was generated using
the autoconf
program written by David MacKenzie of the Free
Software Foundation. It takes a while to run. It will generate the
file `config.h' based on `config.h.in', and, for each source
code directory, will generate `Makefile' based on
`Makefile.in'.
You can pass certain arguments to configure
in the environment.
Because configure
will compile little test programs to see what
is available on your system, you must tell it how to run your compiler.
It recognizes the following environment variables:
configure
can find
`gcc' it will use it, otherwise it will use `cc'.
configure
will use `-g'.
configure
will use the empty string.
configure
will use the empty string.
configure
finds the BSD install
program,
it will set this to `install -c'; otherwise, it will use `cp'.
sh
, bash
, or ksh
,
invoke configure
as `CC=rcc configure'. If you are using
csh
, do `setenv CC rcc; sh configure'.
On some systems you will want to use `LIBS=-lmalloc'. On Xenix
derived versions of Unix do not use `LIBS=-lx' because this will
bring in the wrong versions of certain routines; if you want to use
`-lx' you must specify `LIBS=-lc -lx'.
If configure
fails for some reason, or if you have a very weird
system, you may have to configure the package by hand. To do this, copy
the file `config.h.in' to `config.h' and edit it for your
system. Then for each source directory (the top directory, and the
subdirectories `lib', `unix', and `uuconf') copy
`Makefile.in' to `Makefile', find the words within @
characters, and set them correctly for your system.
configure
script will default to passing `-posix' to
gcc
. However, using `-posix' changes the environment to
POSIX, and on ISC 3.0, at least, the default for POSIX_NO_TRUNC
is 1. This can lead to a problem when uuxqt
executes
rmail
. IDA sendmail
has dbm configuration files named
`mailertable.{dir,pag}'. Notice these names are 15 characters
long. When uuxqt
compiled with the `-posix' executes
rmail
, which in turn executes sendmail
, the later is run
under the POSIX environment too. This leads to sendmail
bombing
out with `'error opening 'M' database: name too long'
(mailertable.dir)'. It's rather obscure behaviour, and it took me a day
to find out the cause. I don't use the `-posix' switch; instead, I
run gcc
with `-D_POSIX_SOURCE', and add `-lcposix' to
`LIBS'.
configure
worked correctly by checking
`config.h' and the instances of `Makefile'.
configure
script.
If your system supports pseudo-terminals, and you compiled the code to
support the new style of configuration files (HAVE_TAYLOR_CONFIG
was set to 1 in `policy.h'), you should be able to use the
tstuu
program to test the uucico
daemon. If your system
supports STREAMS based pseudo-terminals, you must compile tstuu.c with
`-DHAVE_STREAMS_PTYS'. (The STREAMS based code was contributed by
Marc Boucher).
To run tstuu
, just type `tstuu' with no arguments. You must
run it in the compilation directory, since it runs `./uucp',
`./uux' and `./uucico'. The tstuu
program will run a
lengthy series of tests (it takes over ten minutes on a slow VAX). You
will need a fair amount of space available in `/usr/tmp'. You will
probably want to put it in the background. Do not use ^Z, because
the program traps on SIGCHLD
and winds up dying. The
tstuu
program will create a directory `/usr/tmp/tstuu' and
fill it with configuration files, and create spool directories
`/usr/tmp/tstuu/spool1' and `/usr/tmp/tstuu/spool2'.
If your system does not support the FIONREAD
call, the
`tstuu' program will run very slowly. This may or may not get
fixed in a later version.
The tstuu
program will finish with an execute file named
`X.something' and a data file named `D.something'
in the directory `/usr/tmp/tstuu/spool1' (or, more likely, in
subdirectories, depending on the choice of SPOOLDIR
in
`policy.h'). Two log files will be created in the directory
`/usr/tmp/tstuu'. They will be named `Log1' and `Log2',
or, if you have selected HAVE_HDB_LOGGING
in `policy.h',
`Log1/uucico/test2' and `Log2/uucico/test1'. There should be
no errors in the log files.
You can test uuxqt
with `./uuxqt -I /usr/tmp/tstuu/Config1'.
This should leave a command file `C.something' and a data
file `D.something' in `/usr/tmp/tstuu/spool1' or in
subdirectories. Again, there should be no errors in the log file.
Assuming you compiled the code with debugging enabled, the `-x'
switch can be used to set debugging modes; see the debug
command
for details (see section Debugging Levels). Use `-x all' to turn on
all debugging and generate far more output than you will ever want to
see. The uucico
daemons will put debugging output in the files
`Debug1' and `Debug2' in the directory `/usr/tmp/tstuu'.
After that, you're pretty much on your own.
On some systems you can also use tstuu
to test uucico
against the system uucico
, by using the `-u' switch. For
this to work, change the definitions of ZUUCICO_CMD
and
UUCICO_EXECL
at the top of `tstuu.c' to something
appropriate for your system. The definitions in `tstuu.c' are what
I used for Ultrix 4.0, on which `/usr/lib/uucp/uucico' is
particularly obstinate about being run as a child; I was only able to
run it by creating a login name with no password whose shell was
`/usr/lib/uucp/uucico'. Calling login in this way will leave fake
entries in `wtmp' and `utmp'; if you compile `tstout.c'
(in the `contrib' directory) as a setuid root
program,
tstuu
will run it to clear those entries out. On most systems,
such hackery should not be necessary, although on SCO I had to su to
root
(uucp
might also have worked) before I could run
`/usr/lib/uucp/uucico'.
You can test uucp
and uux
(give them the `-r' switch
to keep them from starting uucico
) to make sure they create the
right sorts of files. Unfortunately, if you don't know what the right
sorts of files are, I'm not going to tell you here.
If you can not run tstuu
, or if it fails inexplicably, don't
worry about it too much. On some systems tstuu
will fail because
of problems using pseudo terminals, which will not matter in normal use.
The real test of the package is talking to another system.
You can install the executable files by becoming root
and typing
`make install'. Or you can look at what `make install' does
and do it by hand. It tries to preserve your old programs, if any, but
it only does this the first time Taylor UUCP is installed (so that if
you install several versions of Taylor UUCP, you can still go back to
your original UUCP programs). You can retrieve the original programs by
typing `make uninstall'.
Note that by default the programs are compiled with debugging
information, and they are not stripped when they are installed. You may
want to strip the installed programs to save disk space. For more
information, see your system documentation for the strip
program.
Of course, simply installing the executable files is not enough. You must also arrange for them to be used correctly.
You will have to decide what types of configuration files you want to use. This package supports a new sort of configuration file; see section Taylor UUCP Configuration Files. It also supports V2 configuration files (`L.sys', `L-devices', etc.) and HDB configuration files (`Systems', `Devices', etc.). No documentation is provided for V2 or HDB configuration files. All types of configuration files can be used at once, if you are so inclined. Currently using just V2 configuration files is not really possible, because there is no way to specify a dialer (there are no built in dialers, and the program does not know how to read `acucap' or `modemcap'); however, V2 configuration files can be used with a new style dial file (see section The Dialer Configuration File), or with a HDB `Dialers' file.
Use of HDB configuration files has two known bugs. A blank line in the
middle of an entry in the `Permissions' file will not be ignored as
it should be. Dialer programs, as found in some versions of HDB, are
not recognized directly. If you must use a dialer program, rather than
an entry in `Devices', you must use the chat-program
command
in a new style dial file; see section The Dialer Configuration File. You will have to invoke
the dialer program via a shell script or another program, since an exit
code of 0 is required to recognize success; the dialHDB
program
in the `contrib' directory may be used for this purpose.
The uuconv
(see section Invoking uuconv) program can be used to
convert from V2 or HDB configuration files to the new style (it can also
do the reverse translation, if you are so inclined). It will not do all
of the work, and the results should be carefully checked, but it can be
quite useful.
If you are installing a new system, you will, of course, have to write the configuration files; see section Taylor UUCP Configuration Files for details on how to do this.
After writing the configuration files, use the uuchk
program to
verify that they are what you expect; see section Invoking uuchk.
After you have written the configuration files, and verified them with
the uuchk
program (see section Invoking uuchk), you must check that
UUCP can correctly contact another system.
Tell uucico
to dial out to the system by using the `-s'
system switch (e.g., `uucico -s uunet'). The log file should tell
you what happens. The exact location of the log file depends upon the
settings in `policy.h' when you compiled the program, and on the
use of the logfile
command in the `config' file. Typical
locations are `/usr/spool/uucp/Log' or a subdirectory under
`/usr/spool/uucp/.Log'.
If you compiled the code with debugging enabled, you can use debugging
mode to get a great deal of information about what sort of data is
flowing back and forth; the various possibilities are described with the
debug
command (see section Debugging Levels). When initially setting
up a connection `-x chat' is probably the most useful (e.g.,
`uucico -s uunet -x chat'); you may also want to use `-x
handshake,incoming,outgoing'. You can use `-x' multiple times on
one command line, or you can give it comma separated arguments as in the
last example. Use `-x all' to turn on all possible debugging
information.
The debugging information is written to a file, normally
`/usr/spool/uucp/Debug', although the default can be changed in
`policy.h', and the `config' file can override the default
with the debugfile
command. The debugging file may contain
passwords and some file contents as they are transmitted over the line,
so the debugging file is only readable by the uucp
user.
You can use the `-f' switch to force uucico
to call out even
if the last call failed recently; using `-S' when naming a system
has the same effect. Otherwise the status file (in the `.Status'
subdirectory of the main spool directory, normally
`/usr/spool/uucp') (see section Status Directory) will prevent too many
attempts from occurring in rapid succession.
On older System V based systems which do not have the setreuid
system call, problems may arise if ordinary users can start an execution
of uuxqt
, perhaps indirectly via uucp
or uux
. UUCP
jobs may wind up executing with a real user ID of the user who invoked
uuxqt
, which can cause problems if the UUCP job checks the real
user ID for security purposes. On such systems, it is safest to put
`run-uuxqt never' (see section Miscellaneous config File Commands) in the
`config' file, so that uucico
never starts uuxqt
, and
invoke uuxqt
directly from a `crontab' file.
Please let me know about any problems you have and how you got around
them. If you do report a problem, please include the version number of
the package you are using, the operating system you are running it on,
and a sample of the debugging file showing the problem (debugging
information is usually what is needed, not just the log file). General
questions such as "why doesn't uucico
dial out" are impossible
to answer without much more information.
By default uucp
and uux
will automatically start up
uucico
to call another system whenever work is queued up.
However, the call may fail, or you may have put in time restrictions
which prevent the call at that time (perhaps because telephone rates are
high) (see section When to Call). Also, a remote system may have work
queued up for your system, but may not be calling you for some reason
(perhaps you have agreed that your system should always place the call).
To make sure that work gets transferred between the systems withing a
reasonable time period, you should arrange to periodically invoke
uucico
.
These periodic invocations are normally triggered by entries in the `crontab' file. The exact format of `crontab' files, and how new entries are added, varies from system to system; check your local documentation (try `man cron').
To attempt to call all systems with outstanding work, use the command `uucico -r1'. To attempt to call a particular system, use the command `uucico -s system'. To attempt to call a particular system, but only if there is work for it, use the command `uucico -C -s system'. (see section Invoking uucico).
A common case is to want to try to call a system at a certain time, with periodic retries if the call fails. A simple way to do this is to create an empty UUCP command file, known as a poll file. If a poll file exists for a system, then `uucico -r1' will place a call to it. If the call succeeds, the poll file will be deleted.
A poll file can be easily created using the `uux' command, by requesting the execution of an empty command. To create a poll file for system, just do something like this:
uux -r system!
The `-r' tells `uux' to not start up `uucico' immediately. Of course, if you do want `uucico' to start up right away, omit the `-r'; if the call fails, the poll file will be left around to cause a later call.
For example, I use the following crontab entries locally:
45 * * * * /bin/echo /usr/lib/uucp/uucico -r1 | /bin/su uucpa 40 4,10,15 * * * /usr/bin/uux -r uunet!
Every hour, at 45 minutes past, this will check if there is any work to
be done, and, if there is, will call the appropriate system. Also, at
4:40am, 10:40am, and 3:40pm, this will create a poll file file for
`uunet', forcing the next run of uucico
to call
`uunet'.
To accept calls from another system, you must arrange matters such that
when that system calls in, it automatically invokes uucico
on
your system.
The most common arrangement is to create a special user name and
password for incoming UUCP calls. This user name typically uses the
same user ID as the regular uucp
user (Unix permits several user
names to share the same user ID). The shell for this user name should
be set to uucico
.
Here is a sample `/etc/passwd' line to accept calls from a remote system named airs:
Uairs:password:4:8:airs UUCP:/usr/spool/uucp:/usr/lib/uucp/uucico
The details may vary on your system. You must use reasonable user and
group ID's. You must use the correct file name for uucico
. The
password must appear in the UUCP configuration files on the remote
system, but will otherwise never be seen or typed by a human.
Note that uucico
appears as the login shell, and that it will be
run with no arguments. This means that it will start in slave mode and
accept an incoming connection. See section Invoking uucico.
On some systems, creating an empty file named `.hushlogin' in the
home directory will skip the printing of various bits of information
when the remote uucico
logs in, speeding up the UUCP connection
process.
For the greatest security, each system which calls in should use a
different user name, each with a different password, and the
called-login
command should be used in the `sys' file to
ensure that the correct login name is used. See section Accepting a Call,
and see section Security.
If you never need to dial out from your system, but only accept incoming
calls, you can arrange for uucico
to handle logins itself,
completely controlling the port, by using the `--endless' option.
See section Invoking uucico.
Taylor UUCP does not include a mail package. All Unix systems come with
some sort of mail delivery agent, typically sendmail
or
MMDF
. Source code is available for some alternative mail
delivery agents, such as IDA sendmail
and smail
.
Taylor UUCP also does not include a news package. The two major Unix
news packages are C-news
and INN
. Both are available in
source code form.
Configuring and using mail delivery agents is a notoriously complex topic, and I will not be discussing it here. Configuring news systems is usually simpler, but I will not be discussing that either. I will merely describe the interactions between the mail and news systems and UUCP.
A mail or news system interacts with UUCP in two ways: sending and receiving.
When mail is to be sent from your machine to another machine via UUCP,
the mail delivery agent will invoke uux
. It will generally run a
command such as `uux - system!rmail address', where
system is the remote system to which the mail is being sent. It
may pass other options to uux
, such as `-r' or `-g'
(see section Invoking uux).
The news system also invokes uux
in order to transfer articles to
another system. The only difference is that news will use uux
to
invoke rnews
on the remote system, rather than rmail
.
You should arrange for your mail and news systems to invoke the Taylor
UUCP version of uux
. If you only have Taylor UUCP, or if you
simply replace any existing version of uux
with the Taylor UUCP
version, this will probably happen automatically. However, if you have
two UUCP packages installed on your system, you will probably have to
modify the mail and news configuration files in some way.
Actually, if both the system UUCP and Taylor UUCP are using the same
spool directory format, the system uux
will probably work fine
with the Taylor uucico
(the reverse is not the case: the Taylor
uux
requires the Taylor uucico
). However, data transfer
will be somewhat more efficient if the Taylor uux
is used.
To receive mail, all that is necessary is for UUCP to invoke
rmail
. Any mail delivery agent will provide an appropriate
version of rmail
; you must simply make sure that it is in the
command path used by UUCP (it almost certainly already is). The default
command path is set in `policy.h', and it may be overridden for a
particular system by the command-path
command
(see section Miscellaneous sys File Commands).
Similarly, for news UUCP must be able to invoke rnews
. Any news
system will provide a version of rnews
, and you must ensure that
is in a directory on the path that UUCP will search.
In general, the layout of the spool directory may be safely ignored.
However, it is documented here for the curious. This description only
covers the SPOOLDIR_TAYLOR
layout. The ways in which the other
spool directory layouts differ are described in the source file
`unix/spool.c'.
Directories and files are only created when they are needed, so a typical system will not have all of the entries described here.
rmail
and rnews
, send an `E' command rather than an execution file
(see section The E Command).
uucico
will
create an execution file on the fly when it receives an `E'
command.
uustat --status
basically just formats and prints the contents of
the status files (see section uustat Examples).
Each status file has a single text line with six fields.
uucico
was unable to open the port.
uucico
is currently talking to the system.
time
system call).
uucico
may attempt another call. This is set based
on the retry time; see section When to Call. The `-f' or `-S'
options to uucico
direct it to ignore this wait time; see
section Invoking uucico.
uuxqt
executes a job requested by uux
, it first
changes the working directory to the `.Xqtdir' subdirectory. This
permits the job to create any sort of temporary file without worrying
about overwriting other files in the spool directory. Any files left
in the `.Xqtdir' subdirectory are removed after each execution is
complete.
uuxqt
are executing simultaneously,
each one executes jobs in a separate directory. The first uses
`.Xqtdir', the second uses `.Xqtdir0001', the third uses
`.Xqtdir0002', and so forth.
uuxqt
encounters an execution file which it is unable to
parse, it saves it in the `.Corrupt' directory, and sends mail
about it to the UUCP administrator.
uuxqt
executes a job, and the job fails, and there is enough
disk space to hold the command file and all the data files, then
uuxqt
saves the files in the `.Failed' directory, and sends
mail about it to the UUCP administrator.
sequence
command is used for a system
(see section Miscellaneous sys File Commands). The sequence number for the system
system is stored in the file `.Sequence/system'. It is
simply stored as a printable number.
uucico
receives the file into
`.Temp/system/temp', where system is the name of
the remote system, and temp is the temporary file name. If a
conversation fails during a file transfer, these files are used to
automatically restart the file transfer from the point of failure.
If the `S' or `E' command does not include a temporary file
name, automatic restart is not possible. In this case, the files are
received into a randomly named file in the `.Temp' directory
itself.
uucico
will store the
data file in the `.Preserve' directory, and send mail to the
requestor describing the failure and naming the saved file.
uucico
acknowledges receipt of a
file, it is possible for the acknowledgement to be lost. If this
happens, the remote system will resend the file. If the file were an
execution request, and uucico
did not keep track of which files
it had already received, this could lead to the execution being
performed twice.
To avoid this problem, when a conversation fails, uucico
records
each file that has been received, but for which the remote system may
not have received the acknowledgement. It records this information by
creating an empty file with the name
`.Received/system/temp', where system is the name
of the remote system, and temp is the temp field of the
`S' or `E' command from the remote system (see section The S Command). Then, if the remote system offers the file again in the next
conversation, uucico
refuses the send request and deletes the
record in the `.Received' directory. This approach only works for
file sends which use a temporary file name, but this is true of all
execution requests.
Lock files for devices and systems are stored in the lock directory,
which may or may not be the same as the spool directory. The lock
directory is set at compilation time by LOCKDIR
in
`policy.h', which may be overridden by the lockdir
command
in the `config' file (see section Miscellaneous config File Commands).
For a description of the names used for device lock files, and the format of the contents of a lock file, see section UUCP Lock Files.
uucico
while talking to a remote system, and are used to prevent multiple
simultaneous conversations with a system.
On systems which limit file names to 14 characters, only the first eight
characters of the system name are used in the lock file name. This
requires that the names of each directly connected remote system be
unique in the first eight characters.
uuxqt
starts up, it uses lock files to determine how many
other uuxqt
daemons are currently running. It first tries to
lock `LCK.XQT.0', then `LCK.XQT.1', and so forth. This is
used to implement the max-uuxqts
command (see section Miscellaneous config File Commands). It is also used to parcel out the `.Xqtdir'
subdirectories (see section Execution Subdirectories).
uuxqt
is invoked with the `-c' or `--command'
option (see section Invoking uuxqt), it creates a lock file named after the
command it is executing. For example, `uuxqt -c rmail' will create
the lock file `LXQ.rmail'. This prevents other uuxqt
daemons from executing jobs of the specified type.
uuxqt
is executing a particular job, it creates a lock file
with the same name as the `X.' file describing the job, but
replacing the initial `X' with `L'. This ensures that if
multiple uuxqt
daemons are running, they do not simultaneously
execute the same job.
fcntl
system
call.
The spool directory may need to be cleaned up periodically. Under some circumstances, files may accumulate in various subdirectories, such as `.Preserve' (see section Other Spool Subdirectories) or `.Corrupt' (see section Execution Subdirectories).
Also, if a remote system stops calling in, you may want to arrange for
any queued up mail to be returned to the sender. This can be done using
the uustat
command (see section Invoking uustat).
The `contrib' directory includes a simple `uuclean' script which may be used as an example of a clean up script. It can be run daily out of `crontab'.
You should periodically trim the UUCP log files, as they will otherwise
grow without limit. The names of the log files are set in
`policy.h', and may be overridden in the configuration file
(see section The Main Configuration File). By default they are are
`/usr/spool/uucp/Log' and `/usr/spool/uucp/Stats'. You may
find the savelog
program in the `contrib' directory to be of
use. There is a manual page for it in `contrib' as well.
This chapter describes the configuration files accepted by the Taylor
UUCP package if compiled with HAVE_TAYLOR_CONFIG
set to 1 in
`policy.h'.
The configuration files are normally found in the directory newconfigdir, which is defined by the `Makefile' variable `newconfigdir'; by default newconfigdir is `/usr/local/conf/uucp'. However, the main configuration file, `config', is the only one which must be in that directory, since it may specify a different location for any or all of the other files. You may run any of the UUCP programs with a different main configuration file by using the `-I' or `--config' option; this can be useful when testing a new configuration. When you use the `-I' option the programs will revoke any setuid privileges.
UUCP uses several different types of configuration files, each describing a different kind of information. The commands permitted in each file are described in detail below. This section is a brief description of some of the different types of files.
The `config' file is the main configuration file. It describes general information not associated with a particular remote system, such as the location of various log files. There are reasonable defaults for everything that may be specified in the `config' file, so you may not actually need one on your system.
There may be only one `config' file, but there may be one or more of each other type of file. The default is one file for each type, but more may be listed in the `config' file.
The `sys' files are used to describe remote systems. Each remote system to which you connect must be listed in a `sys' file. A `sys' file will include information for a system, such as the speed (baud rate) to use, or when to place calls.
For each system you wish to call, you must describe one or more ports; these ports may be defined directly in the `sys' file, or they may be defined in a `port' file.
The `port' files are used to describe ports. A port is a particular hardware connection on your computer. You would normally define as many ports as there are modems attached to your computer. A TCP connection is also described using a port.
The `dial' files are used to describe dialers. Dialer is essentially another word for modem. The `dial' file describes the commands UUCP should use to dial out on a particular type of modem. You would normally define as many dialers as there are types of modems attached to your computer. For example, if you have three Telebit modems used for UUCP, you would probably define three ports and one dialer.
There are other types of configuration files, but these are the important ones. The other types are described below.
All the configuration files follow a simple line-oriented `keyword value' format. Empty lines are ignored, as are leading spaces; unlike HDB, lines with leading spaces are read. The first word on each line is a keyword. The rest of the line is interpreted according to the keyword. Most keywords are followed by numbers, boolean values or simple strings with no embedded spaces.
The # character is used for comments. Everything from a # to the end of the line is ignored unless the # is preceded by a \ (backslash); if the # is preceeded by a \, the \ is removed but the # remains in the line. This can be useful for a phone number containing a #. To enter the sequence `\#', use `\\#'.
The backslash character may be used to continue lines. If the last character in a line is a backslash, the backslash is removed and the line is continued by the next line. The second line is attached to the first with no intervening characters; if you want any whitespace between the end of the first line and the start of the second line, you must insert it yourself.
However, the backslash is not a general quoting character. For example, you cannot use it to get an embedded space in a string argument.
Everything after the keyword must be on the same line. A boolean
may be specified as y, Y, t, or T for true and
n, N, f, or F for false; any trailing characters
are ignored, so true
, false
, etc., are also acceptable.
This section provides few typical examples of configuration files. There are also sample configuration files in the `sample' subdirectory of the distribution.
To start with, here are some examples of uses of the main configuration file, `config'. For a complete description of the commands that are permitted in `config', see section The Main Configuration File.
In many cases you will not need to create a `config' file at all. The most common reason to create one is to give your machine a special UUCP name. Other reasons might be to change the UUCP spool directory, or to permit any remote system to call in.
If you have an internal network of machines, then it is likely that the internal name of your UUCP machine is not the name you want to use when calling other systems. For example, here at `airs.com' our mail/news gateway machine is named `elmer.airs.com' (it is one of several machines all named `localname.airs.com'). If we did not provide a `config' file, then our UUCP name would be `elmer'; however, we actually want it to be `airs'. Therefore, we use the following line in `config':
nodename airs
The UUCP spool directory name is set in `policy.h' when the code is compiled. You might at some point decide that it is appropriate to move the spool directory, perhaps to put it on a different disk partition. You would use the following commands in `config' to change to directories on the partition `/uucp':
spool /uucp/spool pubdir /uucp/uucppublic logfile /uucp/spool/Log debugfile /uucp/spool/Debug
You would then move the contents of the current spool directory to `/uucp/spool'. If you do this, make sure that no UUCP processes are running while you change `config' and move the spool directory.
Suppose you wanted to permit any system to call in to your system and
request files. This is generally known as anonymous UUCP, since
the systems which call in are effectively anonymous. By default,
unknown systems are not permitted to call in. To permit this you must
use the unknown
command in `config'. The unknown
command is followed by any command that may appear in the system file;
for full details, see section The System Configuration File.
I will show two possible anonymous UUCP configurations. The first will let any system call in and download files, but will not permit them to upload files to your system.
# No files may be transferred to this system unknown receive-request no # The public directory is /usr/spool/anonymous unknown pubdir /usr/spool/anonymous # Only files in the public directory may be sent (the default anyhow) unknown remote-send ~
Setting the public directory is convenient for the systems which call in. It permits to request a file by prefixing it with `~/'. For example, assuming your system is known as `server', then to retrieve the file `/usr/spool/anonymous/INDEX' a user on a remote site could just enter `uucp server!~/INDEX ~'; this would transfer `INDEX' from `server''s public directory to the user's local public directory. Note that when using `csh' or `bash' the ! and the second ~ must be quoted.
The next example will permit remote systems to upload files to a special directory named `/usr/spool/anonymous/upload'. Permitting a remote system to upload files permits it to send work requests as well; this example is careful to prohibit commands from unknown systems.
# No commands may be executed (the list of permitted commands is empty) unknown commands # The public directory is /usr/spool/anonymous unknown pubdir /usr/spool/anonymous # Only files in the public directory may be sent; users may not download # files from the upload directory unknown remote-send ~ !~/upload # May only upload files into /usr/spool/anonymous/upload unknown remote-receive ~/upload
A relatively common simple case is a leaf site, a system which only calls or is called by a single remote site. Here is a typical `sys' file that might be used in such a case. For full details on what commands can appear in the `sys' file, see section The System Configuration File.
This is the `sys' file that is used at `airs.com'. We use a single modem to dial out to `uunet'. This example shows how you can specify the port and dialer information directly in the `sys' file for simple cases. It also shows the use of the following:
call-login
call-login
and call-password
allows the default
login chat script to be used. In this case, the login name is specified
in the call-out login file (see section Configuration File Names).
call-timegrade
chat-fail
protocol-parameter
This `sys' file relies on certain defaults. It will allow `uunet' to queue up `rmail' and `rnews' commands. It will allow users to request files from `uunet' into the UUCP public directory. It will also allow `uunet' to request files from the UUCP public directory; in fact `uunet' never requests files, but for additional security we could add the line `request false'.
# The following information is for uunet system uunet # The login name and password are kept in the callout password file call-login * call-password * # We can send anything at any time. time any # During the day we only accept grade `Z' or above; at other times # (not mentioned here) we accept all grades. uunet queues up news # at grade `d', which is lower than `Z'. call-timegrade Z Wk0755-2305,Su1655-2305 # The phone number. phone 7389449 # uunet tends to be slow, so we increase the timeout chat-timeout 120 # We are using a preconfigured Telebit 2500. port type modem port device /dev/ttyd0 port speed 19200 port carrier true port dialer chat "" ATZ\r\d\c OK ATDT\D CONNECT port dialer chat-fail BUSY port dialer chat-fail NO\sCARRIER port dialer complete \d\d+++\d\dATH\r\c port dialer abort \d\d+++\d\dATH\r\c # Increase the timeout and the number of retries. protocol-parameter g timeout 20 protocol-parameter g retries 10
Many organizations have several local machines which are connected by UUCP, and a single machine which connects to the outside world. This single machine is often referred to as a gateway machine.
For this example I will assume a fairly simple case. It should still provide a good general example. There are three machines, `elmer', `comton' and `bugs'. `elmer' is the gateway machine for which I will show the configuration file. `elmer' calls out to `uupsi'. As an additional complication, `uupsi' knows `elmer' as `airs'; this will show how a machine can have one name on an internal network but a different name to the external world. `elmer' has two modems. It also has an TCP connection to `uupsi', but since that is supposed to be reserved for interactive work (it is, perhaps, only a 9600 baud SLIP line) it will only use it if the modems are not available.
A network this small would normally use a single `sys' file.
However, for pedagogical purposes I will show two separate `sys'
files, one for the local systems and one for `uupsi'. This is done
with the sysfile
command in the `config' file. Here is the
`config' file.
# This is config # The local sys file sysfile /usr/local/lib/uucp/sys.local # The remote sys file sysfile /usr/local/lib/uucp/sys.remote
Using the defaults feature of the `sys' file can greatly simplify the listing of local systems. Here is `sys.local'. Note that this assumes that the local systems are trusted; they are permited to request any world readable file and to write files into any world writable directory.
# This is sys.local # Get the login name and password to use from the call-out file call-login * call-password * # The systems must use a particular login called-login Ulocal # Permit sending any world readable file local-send / remote-send / # Permit receiving into any world writable directory local-receive / remote-receive / # Call at any time time any # Use port1, then port2 port port1 alternate port port2 # Now define the systems themselves. Because of all the defaults we # used, there is very little to specify for the systems themselves. system comton phone 5551212 system bugs phone 5552424
The `sys.remote' file describes the `uupsi' connection. The
myname
command is used to change the UUCP name to `airs'
when talking to `uupsi'.
# This is sys.remote # Define uupsi system uupsi # The login name and password are in the call-out file call-login * call-password * # We can call out at any time time any # uupsi uses a special login name called-login Uuupsi # uuspi thinks of us as `airs' myname airs # The phone number phone 5554848 # We use port2 first, then port1, then TCP port port2 alternate port port1 alternate # We don't bother to make a special entry in the port file for TCP, we # just describe the entire port right here. We use a special chat # script over TCP because the usual one confuses some TCP servers. port type TCP address uu.psi.com chat ogin: \L word: \P
The ports are defined in the file `port' (see section The Port Configuration File). For this example they are both connected to the same type of 2400 baud Hayes-compatible modem.
# This is port port port1 type modem device /dev/ttyd0 dialer hayes speed 2400 port port2 type modem device /dev/ttyd1 dialer hayes speed 2400
Dialers are described in the `dial' file (see section The Dialer Configuration File).
# This is dial dialer hayes # The chat script used to dial the phone. \D is the phone number. chat "" ATZ\r\d\c OK ATDT\D CONNECT # If we get BUSY or NO CARRIER we abort the dial immediately chat-fail BUSY chat-fail NO\sCARRIER # When the call is over we make sure we hangup the modem. complete \d\d+++\d\dATH\r\c abort \d\d+++\d\dATH\r\c
Several commands use time strings to specify a range of times. This section describes how to write time strings.
A time string may be a list of simple time strings separated with a vertical bar `|' or a comma `,'.
Each simple time string must begin with `Su', `Mo', `Tu', `We', `Th', `Fr', or `Sa', or `Wk' for any weekday, or `Any' for any day.
Following the day may be a range of hours separated with a hyphen using 24 hour time. The range of hours may cross 0; for example `2300-0700' means any time except 7 AM to 11 PM. If no time is given, calls may be made at any time on the specified day(s).
The time string may also be the single word `Never', which does not
match any time. The time string may also be a single word with a name
defined in a previous timetable
command (see section Miscellaneous config File Commands).
Here are a few sample time strings with an explanation of what they mean.
Chat scripts are used in several different places, such as dialing out on modems or logging in to remote systems. Chat scripts are made up of pairs of strings. The program waits until it sees the first string, known as the expect string, and then sends out the second string, the send string.
Each chat script is defined using a set of commands. These commands
always end in a string beginning with chat
, but may start with
different strings. For example, in the `sys' file there is one set
of commands beginning with chat
and another set beginning with
called-chat
. The prefixes are only used to disambiguate
different types of chat scripts, and this section ignores the prefixes
when describing the commands.
chat strings
chat
command are
pairs of strings separated by whitespace. The first string of each pair
is an expect string, the second is a send string. The program will wait
for the expect string to appear; when it does, the program will send the
send string. If the expect string does not appear within a certain
number of seconds (as set by the chat-timeout
command), the chat
script fails and, typically, the call is aborted. If the final expect
string is seen (and the optional final send string has been sent), the
chat script is successful.
An expect string may contain additional subsend and subexpect strings,
separated by hyphens. If the expect string is not seen, the subsend
string is sent and the chat script continues by waiting for the
subexpect string. This means that a hyphen may not appear in an expect
string; on an ASCII system, use `\055' instead.
An expect string may simply be `""', meaning to skip the expect
phase. Otherwise, the following escape characters may appear in expect
strings:
chat-timeout
(described below) only
for the expect string to which it is attached.
A send string may simply be `""' to skip the send phase.
Otherwise, all of the escape characters legal for expect strings may be
used, and the following escape characters are also permitted:
chat-timeout number
chat-fail string
chat-fail
commands may appear in a single chat script. The
default is to have none.
This permits a chat script to be quickly aborted if an error string is
seen. For example, a script used to dial out on a modem might use the
command `chat-fail BUSY' to stop the chat script immediately if the
string `BUSY' was seen.
The chat-fail
strings are considered in the order they are
listed, so if one string is a suffix of another the longer one should be
listed first. This affects the error message which will be logged. Of
course, if one string is contained within another, but is not a suffix,
the smaller string will always be found before the larger string could
match.
chat-seven-bit boolean
chat-program
, which must ignore parity by itself if necessary.
chat-program strings
chat-program
and chat
are specified, the
program is executed first followed by the chat script.
The first argument to the chat-program
command is the program
name to run. The remaining arguments are passed to the program. The
following escape sequences are recognized in the arguments:
chat-program
define additional escape
sequences.
Arguments other than escape sequences are passed exactly as they appear
in the configuration file, except that sequences of whitespace are
compressed to a single space character (this exception may be removed in
the future).
If the chat-program
command is not used, no program is run.
On Unix, the standard input and standard output of the program will be
attached to the port in use. Anything the program writes to standard
error will be written to the UUCP log file. No other file descriptors
will be open. If the program does not exit with a status of 0, it will
be assumed to have failed. This means that the dialing programs used by
some versions of HDB may not be used directly, but you may be able to
run them via the dialHDB
program in the `contrib' directory.
The program will be run as the uucp
user, and the environment
will be that of the process that started uucico
, so care must be
taken to maintain security.
No search path is used to find the program; a full file name must be
given. If the program is an executable shell script, it will be passed
to `/bin/sh' even on systems which are unable to execute shell
scripts.
Here is a simple example of a chat script that might be used to reset a Hayes compatible modem.
chat "" ATZ OK-ATZ-OK
The first expect string is `""', so it is ignored. The chat script then sends `ATZ'. If the modem responds with `OK', the chat script finishes. If 60 seconds (the default timeout) pass before seeing `OK', the chat script sends another `ATZ'. If it then sees `OK', the chat script succeeds. Otherwise, the chat script fails.
For a more complex chat script example, see section Logging In.
The main configuration file is named `config'.
Since all the values that may be specified in the main configuration file also have defaults, there need not be a main configuration file at all.
Each command in `config' may have a program prefix, which is a
separate word appearing at the beginning of the line. The currently
supported prefixes are `uucp' and `cu'. Any command prefixed
by `uucp' will not be read by the cu
program. Any command
prefixed by `cu' will only be read by the cu
program. For
example, to use a list of systems known only to cu
, list them in
a separate file `file' and put `cu sysfile
`file'' in `config'.
nodename string
hostname string
uuname string
spool string
pubdir string
pubdir
command in the system configuration file; see
section Miscellaneous sys File Commands.
lockdir string
unknown string ...
unknown
command is not used,
unknown systems are not permitted to call in.
strip-login boolean
uucico
is doing its own login
prompting with the `-e', `-l', or `-w' switches, it will
strip the parity bit when it reads the login name and password.
Otherwise all eight bits will be used when checking the strings against
the UUCP password file. The default is true, since some other UUCP
packages send parity bits with the login name and password, and few
systems use eight bit characters in the password file.
strip-proto boolean
uucico
will strip the parity bit
from incoming UUCP protocol commands. Otherwise all eight bits will be
used. This only applies to commands which are not encapsulated in a
link layer protocol. The default is true, which should always be
correct unless your UUCP system names use eight bit characters.
max-uuxqts number
uuxqt
processes which may run at
the same time. Having several uuxqt
processes running at once
can significantly slow down a system, but, since uuxqt
is
automatically started by uucico
, it can happen quite easily. The
default for max-uuxqts
is 0, which means that there is no limit.
If HDB configuration files are being read and the code was compiled
without HAVE_TAYLOR_CONFIG
, then, if the file `Maxuuxqts' in
the configuration directory contains a readable number, it will be used
as the value for max-uuxqts
.
run-uuxqt string or number
uuxqt
should be run by uucico
. This may be a
positive number, in which case uucico
will start a uuxqt
process whenever it receives the given number of execution files from
the remote system, and, if necessary, at the end of the call. The
argument may also be one of the strings `once', `percall', or
`never'. The string `once' means that uucico
will
start uuxqt
once at the end of execution. The string
`percall' means that uucico
will start uuxqt
once per
call that it makes (this is only different from once
when
uucico
is invoked in a way that causes it to make multiple calls,
such as when the `-r1' option is used without the `-s'
option). The string `never' means that uucico
will never
start uuxqt
, in which case uuxqt
should be periodically
run via some other mechanism. The default depends upon which type of
configuration files are being used; if HAVE_TAYLOR_CONFIG
is used
the default is `once', otherwise if HAVE_HDB_CONFIG
is used
the default is `percall', and otherwise, for HAVE_V2_CONFIG
,
the default is `10'.
timetable string string
timetable
defines a timetable that may be used in
subsequently appearing time strings; see section Time Strings. The first
string names the timetable entry; the second is a time string.
The following timetable
commands are predefined. The NonPeak
timetable is included for compatibility. It originally described the
offpeak hours of Tymnet and Telenet, but both have since changed their
schedules.
timetable Evening Wk1705-0755,Sa,Su timetable Night Wk2305-0755,Sa,Su2305-1655 timetable NonPeak Wk1805-0655,Sa,SuIf this command does not appear, then, obviously, no additional timetables will be defined.
v2-files boolean
hdb-files boolean
sysfile strings
sysfile
command may be repeated; each system file has its own set of defaults.
portfile strings
portfile
command may be repeated.
dialfile strings
dialfile
command may be repeated.
dialcodefile strings
dialcodefile
command may be repeated; all the dialcode files will
be read in turn until a dialcode is located.
callfile strings
passwdfile
below is used for incoming calls. The intention of the call out file is
to permit the system file to be publically readable; the call out files
must obviously be kept secure. These files need not be used. Multiple
call out files may be specified on the line, and the callfile
command may be repeated; all the files will be read in turn until the
system is found.
passwdfile strings
uucico
is doing its own login prompting, which it does when given the
`-e', `-l' or `-w' switches. The default is the file
`passwd' in the directory newconfigdir. Each line in the
file(s) has two words: the login name and the password (e.g., Ufoo
foopas
). They may contain escape sequences like those in a chat script
expect string (see section Chat Scripts). The login name is accepted before
the system name is known, so these are independent of which system is
calling in; a particular login may be required for a system by using the
called-login
command in the system file (see section Accepting a Call). These password files are optional, although one must exist if
uucico
is to present its own login prompts.
As a special exception, a colon may be used to separate the login name
from the password, and a colon may be used to terminate the password.
This means that the login name and password may not contain a colon.
This feature, in conjunction with the HAVE_ENCRYPTED_PASSWORDS
macro in `policy.h', permits using a standard Unix
`/etc/passwd' as a UUCP password file, providing the same set of
login names and passwords for both getty
and uucico
.
Multiple password files may be specified on the line, and the
passwdfile
command may be repeated; all the files will be read in
turn until the login name is found.
logfile string
HAVE_HDB_LOGGING
is
defined in `policy.h', then by default a separate log file is used
for each system; using this command to name a log file will cause all
the systems to use it.
statfile string
debugfile string
DEBUG
macro in `policy.h'). If debugging is on, messages
written to the log file are also written to the debugging file to make
it easier to keep the order of actions straight. The debugging file is
different from the log file because information such as passwords can
appear in it, so it must be not be publically readable.
debug string ...
debug
command may be used several times in the
configuration file; every debugging type named will be turned on. When
running any of the programs, the `-x' switch (actually, for
uulog
it's the `-X' switch) may be used to turn on
debugging. The argument to the `-x' switch is one of the strings
listed above, or a number as described above, or a comma separated list
of strings (e.g., `-x chat,handshake'). The `-x' switch may
also appear several times on the command line, in which case all named
debugging types will be turned on. The `-x' debugging is in
addition to any debugging specified by the debug
command; there
is no way to cancel debugging information. The debugging level may also
be set specifically for calls to or from a specific system with the
debug
command in the system file (see section Miscellaneous sys File Commands).
The debugging messages are somewhat idiosyncratic; it may be necessary
to refer to the source code for additional information in some cases.
By default there is a single system configuration, named `sys' in
the directory newconfigdir. This may be overridden by the
sysfile
command in the main configuration file; see
section Configuration File Names.
These files describe all remote systems known to the UUCP package.
The first set of commands in the file, up to the first system
command, specify defaults to be used for all systems in that file. Each
`sys' file uses a different set of defaults.
Subsequently, each set of commands from system
up to the next
system
command describe a particular system. Default values may
be overridden for specific systems.
Each system may then have a series of alternate choices to use when
calling out or calling in. The first set of commands for a particular
system, up to the first alternate
command, provide the first
choice. Subsequently, each set of commands from alternate
up to
the next alternate
command describe an alternate choice for
calling out or calling in.
When a system is called, the commands before the first alternate
are used to select a phone number, port, and so forth; if the call fails
for some reason, the commands between the first alternate
and the
second are used, and so forth. Well, not quite. Actually, each
succeeding alternate will only be used if it is different in some
relevant way (different phone number, different chat script, etc.). If
you want to force the same alternate to be used again (to retry a phone
call more than once, for example), enter the phone number (or any other
relevant field) again to make it appear different.
The alternates can also be used to give different permissions to an
incoming call based on the login name. This will only be done if the
first set of commands, before the first alternate
command, uses
the called-login
command. The list of alternates will be
searched, and the first alternate with a matching called-login
command will be used. If no alternates match, the call will be
rejected.
The alternate
command may also be used in the file-wide defaults
(the set of commands before the first system
command). This
might be used to specify a list of ports which are available for all
systems (for an example of this, see section Gateway Example) or to
specify permissions based on the login name used by the remote system
when it calls in. The first alternate for each system will default to
the first alternate for the file-wide defaults (as modified by the
commands used before the first alternate
command for this
system), the second alternate for each system to the second alternate
for the file-wide defaults (as modified the same way), and so forth. If
a system specifies more alternates than the file-wide defaults, the
trailing ones will default to the last file-wide default alternate. If
a system specifies fewer alternates than the file-wide defaults, the
trailing file-wide default alternates will be used unmodified. The
default-alternates
command may be used to modify this behaviour.
This can all get rather confusing, although it's easier to use than to
describe concisely; the uuchk
program may be used to ensure that
you are getting what you want.
system string
system
command refer to this system.
alternate [string]
alternate
command).
default-alternates boolean
alias string
uucp
and uux
commands, as well as by the remote system
(which can be convenient if a remote system changes its name). The
default is to have no aliases.
myname string
called-login
is used and is not `ANY', then, when a
system logs in with that login name, string is used as the local
system name. Because the local system name must be determined before
the remote system has identified itself, using myname
and
called-login
together for any system will set the local name for
that login; this means that each locally used system name must have a
unique login name associated with it. This allows a system to have
different names for an external and an internal network. The default is
to not use a special local name.
This section describes commands used when placing a call to another system.
time string [number]
time
command is always a fixed amount of time.
The time
command may appear multiple times in a single alternate,
in which case if any time string matches the system may be called. When
the time
command is used for a particular system, any time
or timegrade
commands that appeared in the system defaults are
ignored.
The default time string is `Never'.
timegrade character string [number]
time
command is equivalent to using timegrade
with a grade of z, permitting all jobs. If there are no jobs of a
sufficiently high grade according to the time string, the system will
not be called. Giving the `-s' switch to uucico
to force it
to call a system causes it to assume there is a job of grade 0
waiting to be run.
The optional third argument specifies a retry time in minutes. See the
time
command, above, for more details.
Note that the timegrade
command serves two purposes: 1) if there
is no job of sufficiently high grade the system will not be called, and
2) if the system is called anyway (because the `-s' switch was
given to uucico
) only jobs of sufficiently high grade will be
transferred. However, if the other system calls in, the
timegrade
commands are ignored, and jobs of any grade may be
transferred (but see call-timegrade
and called-timegrade
,
below). Also, the timegrade
command will not prevent the other
system from transferring any job it chooses, regardless of who placed
the call.
The timegrade
command may appear multiple times without using
alternate
. When the timegrade
command is used for a
particular system, any time
or timegrade
commands that
appeared in the system defaults are ignored.
If this command does not appear, there are no restrictions on what grade
of work may be done at what time.
max-retries number
success-wait number
call-timegrade character string
call-timegrade
command may appear multiple times without
using alternate
. If this command does not appear, or if none of
the time strings match, the remote system will be allowed to send
whatever grades of work it chooses.
called-timegrade character string
called-timegrade
command may appear multiple times. If this
command does not appear, or if none of the time strings match, any grade
may be sent to the remote system upon receiving a call.
speed number
baud number
speed
and
port
commands appear, both are used when selecting a port. To
allow calls at more than one speed, the alternate
command must be
used (see section Defaults and Alternates). If this command does not
appear, there is no default; the speed may be specified in the port
file, but if it is not then the natural speed of the port will be used
(whatever that means on the system). Specifying an explicit speed of 0
will request the natural speed of the port (whatever the system sets it
to), overriding any default speed from the defaults at the top of the
file.
port string
speed
command or explicitly using the next version of
port
). There may be many ports with the same name; each will be
tried in turn until an unlocked one is found which matches the desired
speed.
port string ...
port
command, the strings are
treated as a command that might appear in the port file (see section The Port Configuration File). If a port is named (by using a single string following
port
) these commands are ignored; their purpose is to permit
defining the port completely in the system file rather than always
requiring entries in two different files. In order to call out, a port
must be specified using some version of the port
command, or by
using the speed
command to select ports from the port file.
phone string
address string
phone
and address
are equivalent; the duplication is intended to
provide a mnemonic choice depending on the type of port in use.
When used with a modem port, an = character in the phone number
means to wait for a secondary dial tone (although only some modems
support this); a - character means to pause while dialing for 1
second (again, only some modems support this). If the system has more
than one phone number, each one must appear in a different alternate.
The phone
command must appear in order to call out on a modem;
there is no default.
When used with a TCP port, the string names the host to contact. It may
be a domain name or a numeric Internet address. If no address is
specified, the system name is used.
When used with a TLI port, the string is treated as though it were an
expect string in a chat script, allowing the use of escape characters
(see section Chat Scripts). The dialer-sequence
command in the port
file may override this address (see section The Port Configuration File).
When used with a port that not a modem or TCP or TLI, this command is
ignored.
chat strings
chat-timeout number
chat-fail string
chat-seven-bit boolean
chat-program strings
call-login
command.
call-password
command.
chat-program
command. These are `\L' and `\P', which
become the login name and password, respectively, and `\Z', which
becomes the name of the system of being called.
The default chat script is:
chat "" \r\c ogin:-BREAK-ogin:-BREAK-ogin: \L word: \PThis will send a carriage return (the \c suppresses the additional trailing carriage return that would otherwise be sent) and waits for the string `ogin:' (which would be the last part of the `login:' prompt supplied by a Unix system). If it doesn't see `ogin:', it sends a break and waits for `ogin:' again. If it still doesn't see `ogin:', it sends another break and waits for `ogin:' again. If it still doesn't see `ogin:', the chat script aborts and hangs up the phone. If it does see `ogin:' at some point, it sends the login name (as specified by the
call-login
command) followed by a
carriage return (since all send strings are followed by a carriage
return unless \c is used) and waits for the string `word:'
(which would be the last part of the `Password:' prompt supplied by
a Unix system). If it sees `word:', it sends the password and a
carriage return, completing the chat script. The program will then
enter the handshake phase of the UUCP protocol.
This chat script will work for most systems, so you will only be
required to use the call-login
and call-password
commands.
In fact, in the file-wide defaults you could set defaults of
`call-login *' and `call-password *'; you would then just have
to make an entry for each system in the call-out login file.
Some systems seem to flush input after the `login:' prompt, so they
may need a version of this chat script with a \d before the
\L. When using UUCP over TCP, some servers will not be handle the
initial carriage return sent by this chat script; in this case you may
have to specify the simple chat script `ogin: \L word: \P'.
call-login string
call-password string
called-login strings
called-login
commands, in which case the login name will be used
to select which alternate is in effect; this will only work if the first
alternate (before the first alternate
command) uses the
called-login
command.
Additional strings may be specified after the login name; they are a
list of which systems are permitted to use this login name. If this
feature is used, then normally the login name will only be given in a
single called-login
command. Only systems which appear on the
list, or which use an explicit called-login
command, will be
permitted to use that login name. If the same login name is used more
than once with a list of systems, all the lists are concatenated
together. This feature permits you to restrict a login name to a
particular set of systems without requiring you to use the
called-login
command for every single system; you can achieve a
similar effect by using a different system file for each permitted login
name with an appropriate called-login
command in the file-wide
defaults.
callback boolean
uucico
will hang up the connection and prepare to call it back.
The default is false.
called-chat strings
called-chat-timeout number
called-chat-fail string
called-chat-seven-bit boolean
called-chat-program strings
chat
command
(see section Logging In), on the other hand, is used when the remote system
is called. This called chat script might be used to set special modem
parameters that are appropriate to a particular system. It is run after
protocol negotiation is complete, but before the protocol has been
started. For additional escape sequence which may be used besides those
defined for all chat scripts, see section Logging In. There is no default
called chat script. If the called chat script fails, the incoming call
will be aborted.
protocol string
seven-bit
and reliable
commands. If
neither the port nor the dialer use either of these commands, the
default is to assume an eight-bit reliable connection. The commands
`seven-bit true' or `reliable false' might be used in either
the port or the dialer to change this. Each protocol has particular
requirements that must be met before it will be considered during
negotiation with the remote side.
The `t' and `e' protocols are intended for use over TCP or
some other communication path with end to end reliability, as they do no
checking of the data at all. They will only be considered on a TCP port
which is both reliable and eight bit. For technical details, see section UUCP `t' Protocol, and section UUCP `e' Protocol.
The `i' protocol is a bidirectional protocol. It requires an
eight-bit connection. It will run over a half-duplex link, such as
Telebit modems in PEP mode, but for efficient use of such a connection
you must use the half-duplex
command (see section The Port Configuration File).
See section UUCP `i' Protocol.
The `g' protocol is robust, but requires an eight-bit connection.
See section UUCP `g' Protocol.
The `G' protocol is the System V Release 4 version of the `g'
protocol. See section UUCP `G' Protocol.
The `a' protocol is a Zmodem like protocol, contributed by Doug
Evans. It requires an eight-bit connection, but unlike the `g' or
`i' protocol it will work if certain control characters may not be
transmitted.
The `j' protocol is a variant of the `i' protocol which can
avoid certain control characters. The set of characters it avoids can
be set by a parameter. While it technically does not require an eight
bit connection (it could be configured to avoid all characters with the
high bit set) it would be very inefficient to use it over one. It is
useful over a eight-bit connection that will not transmit certain
control characters. See section UUCP `j' Protocol.
The `f' protocol is intended for use with X.25 connections; it
checksums each file as a whole, so any error causes the entire file to
be retransmitted. It requires a reliable connection, but only uses
seven-bit transmissions. It is a streaming protocol, so, while it can
be used on a serial port, the port must be completely reliable and flow
controlled; many aren't. See section UUCP `f' Protocol.
The `v' protocol is the `g' protocol as used by the DOS
program UUPC/Extended. It is provided only so that UUPC/Extended users
can use it; there is no particular reason to select it. See section UUCP `v' Protocol.
The `y' protocol is an efficient streaming protocol. It does error
checking, but when it detects an error it immediately aborts the
connection. This requires a reliable, flow controlled, eight-bit
connection. In practice, it is only useful on a connection that is
nearly always error-free. Unlike the `t' and `e' protocols,
the connection need not be entirely error-free, so the `y' protocol
can be used on a serial port. See section UUCP `y' Protocol.
The protocols will be considered in the order shown above. This means
that if neither the seven-bit
nor the reliable
command are
used, the `t' protocol will be used over a TCP connection and the
`i' protocol will be used over any other type of connection
(subject, of course, to what is supported by the remote system; it may
be assumed that all systems support the `g' protocol).
Note that currently specifying both `seven-bit true' and
`reliable false' will not match any protocol. If this occurs
through a combination of port and dialer specifications, you will have
to use the protocol
command for the system or no protocol will be
selected at all (the only reasonable choice would be `protocol f').
A protocol list may also be specified for a port (see section The Port Configuration File),
but, if there is a list for the system, the list for the port is
ignored.
protocol-parameter character string ...
window
packet-size
remote-packet-size
sync-timeout
sync-retries
timeout
retries
errors
error-decay
errors
.
The default is 10.
ack-frequency
short-packets
which takes a boolean argument:
window
packet-size
startup-retries
init-retries
init-timeout
retries
timeout
garbage
errors
error-decay
errors
.
The default is 10.
remote-window
remote-packet-size
short-packets
escape-control
, which takes a boolean
argument:
timeout
retries
startup-retries
garbage
send-window
escape-control
XON
or XOFF
. The connection must
still transmit eight bit characters other than control characters. The
default is false.
avoid
XON
and XOFF
, which many connections use for
flow control. If the package is configured to use HAVE_BSD_TTY
,
then on some versions of Unix you may have to avoid `\377' as well,
due to the way some implementations of the BSD terminal driver handle
signals.
timeout
retries
timeout
timeout
packet-size
send-request boolean
receive-request boolean
request boolean
call-transfer boolean
called-transfer boolean
transfer boolean
call-local-size number string
call-remote-size number string
called-local-size number string
called-remote-size number string
local-send strings
uucp
or uux
). The directories in the list should be separated by
whitespace. A `~' may be used for the public directory. On a Unix
system, this is typically `/usr/spool/uucppublic'; the public
directory may be set with the pubdir
command. Here is an example
of local-send
:
local-send ~ /usr/spool/ftp/pubListing a directory allows all files within the directory and all subdirectories to be sent. Directories may be excluded by preceding them with an exclamation point. For example:
local-send /usr/ftp !/usr/ftp/private ~means that all files in `/usr/ftp' or the public directory may be sent, except those files in `/usr/ftp/private'. The list of directories is read from left to right, and the last directory to apply takes effect; this means that directories should be listed from top down. The default is the root directory (i.e., any file at all may be sent by local request).
remote-send strings
local-receive strings
remote-receive strings
forward-to strings
uucp
command, it effectively has the ability to forward to
any system.
forward-from strings
uucp
command, it effectively has the ability to request files
from any system.
forward strings
sequence boolean
command-path strings
uux
, not for chat programs. The default is from
`policy.h'.
commands strings
free-space number
uucico
will periodically check the amount of free space. If it
drops below the amount given by the free-space
command, the file
transfer will be aborted. The default amount of space to leave free is
from `policy.h'. This file space checking may not work on all
systems.
pubdir string
debug string ...
debug
command
in the main configuration file (see section Debugging Levels) for more
details. The debugging information specified here is in addition to
that specified in the main configuration file or on the command line.
max-remote-debug string ...
max-remote-debug
command may be used to
put a limit on the debugging level which the system may request, to
avoid filling up the disk with debugging information. Only the
debugging types named in the max-remote-debug
command may be
turned on by the remote system. To prohibit any debugging, use
`max-remote-debug none'.
The following are used as default values for all systems; they can be considered as appearing before the start of the file.
time Never chat "" \r\c ogin:-BREAK-ogin:-BREAK-ogin: \L word: \P chat-timeout 10 callback n sequence n request y transfer y local-send / remote-send ~ local-receive ~ remove-receive ~ command-path [ from `policy.h' ] commands rnews rmail max-remote-debug abnormal,chat,handshake
The port files may be used to name and describe ports. By default there
is a single port file, named `port' in the directory
newconfigdir. This may be overridden by the portfile
command in the main configuration file; see section Configuration File Names.
Any commands in a port file before the first port
command specify
defaults for all ports in the file; however, since the type
command must appear before all other commands for a port, the defaults
are only useful if all ports in the file are of the same type (this
restriction may be lifted in a later version). All commands after a
port
command up to the next port
command then describe
that port. There are different types of ports; each type supports its
own set of commands. Each command indicates which types of ports
support it. There may be many ports with the same name; if a system
requests a port by name then each port with that name will be tried
until an unlocked one is found.
port string
type string
port
command. The type defines
what commands are subsequently allowed. Currently the types are:
uucico
is run as a login shell.
protocol string
protocol-parameter character strings [ any type ]
protocol-parameter
command used for
systems (see section Protocol Selection). This one takes precedence.
seven-bit boolean [ any type ]
reliable boolean [ any type ]
half-duplex boolean [ any type ]
device string [ modem, direct and tli only ]
speed number [modem and direct only ]
baud number [ modem and direct only ]
speed-range number number [ modem only ]
baud-range number number [ modem only ]
speed
(or baud
) command is still used to
determine the speed to run at if the system does not specify a speed.
For example, the command `speed-range 300 19200' means that the
port will match any system which uses a speed from 300 to 19200 baud
(and will use the speed specified by the system); this could be combined
with `speed 2400', which means that when this port is used with a
system that does not specify a speed, the port will be used at 2400
baud.
carrier boolean [ modem and direct only ]
hardflow boolean [ modem and direct only ]
dial-device string [ modem only ]
dialer string [ modem only ]
dialer string ... [ modem only ]
dialer
command, the strings
are treated as a command that might appear in the dial file (see section The Dialer Configuration File). If a dialer is named (by using the first form of this command,
described just above), these commands are ignored. They may be used to
specify dialer information directly in simple situations without needing
to go to a separate file. There is no default. Some sort of dialer
information must be specified to call out on a modem.
dialer-sequence strings [ modem or tcp or tli only ]
phone
command in the system file is used as the final token. The token is
what is used for \D or \T in the dialer chat script. If the
token in this string is \D, the system phone number will be used;
if it is \T, the system phone number will be used after undergoing
dialcodes translation. A missing final token is taken as \D.
This command currently does not work if dial-device
is specified;
to handle this correctly will require a more systematic notion of chat
scripts. Moreover, the complete
and abort
chat scripts,
the protocol parameters, and the carrier
and dtr-toggle
commands are ignored for all but the first dialer.
This command basically lets you specify a sequence of chat scripts to
use. For example, the first dialer might get you to a local network and
the second dialer might describe how to select a machine from the local
network. This lets you break your dialing sequence into simple modules,
and may make it easier to share dialer entries between machines.
This command is to only way to use a chat script with a TCP port. This
can be useful when using a modem which is accessed via TCP.
When this command is used with a TLI port, then if the first dialer is
`TLI' or `TLIS' the first token is used as the address to
connect to. If the first dialer is something else, or if there is no
token, the address given by the address
command is used
(see section Placing the Call). Escape sequences in the address are
expanded as they are for chat script expect strings (see section Chat Scripts). The different between `TLI' and `TLIS' is that the
latter implies the command `stream true'. These contortions are
all for HDB compatibility. Any subsequent dialers are treated as they
are for a TCP port.
lockname string [ modem and direct only ]
lockname LCK..ttycu0
could
be used to force the latter to use the same lock file name as the
former.
service string [ tcp only ]
push strings [ tli only ]
stream boolean [ tli only ]
push
command was not used, the
`tirdwr' module is pushed on to the TLI stream.
server-address string [ tli only ]
t_bind
function. The
value needed may depend upon your particular TLI implementation. Check
the manual pages, and, if necessary, try writing some sample programs.
For AT&T 3B2 System V Release 3 using the Wollongong TCP/IP stack, which
is probably typical, the format of TLI string is `SSPPIIII', where
`SS' is the service number (for TCP, this is 2), `PP' is the
TCP port number, and `IIII' is the Internet address. For example,
to accept a connection from on port 540 from any interface, use
`server-address \x00\x02\x02\x1c\x00\x00\x00\x00'. To only accept
connections from a particular interface, replace the last four digits
with the network address of the interface. (Thanks to Paul Pryor for
the information in this paragraph).
command strings [ pipe only ]
uucico
communicates with it over a pipe. This permits
uucico
or cu
to communicate with another system which can
only be reached through some unusual means. A sample use might be
`command /bin/rlogin -E -8 -l login system'. The
command is run with the full privileges of UUCP; it is responsible for
maintaining security.
The dialer configuration files define dialers. By default there is a
single dialer file, named `dial' in the directory
newconfigdir. This may be overridden by the dialfile
command in the main configuration file; see section Configuration File Names.
Any commands in the file before the first dialer
command specify
defaults for all the dialers in the file. All commands after a
dialer
command up to the next dialer
command are
associated with the named dialer.
dialer string
chat strings
chat-timeout number
chat-fail string
chat-seven-bit boolean
chat-program strings
uucico
daemon will sleep for one second between attempts to
dial out on a modem. If your modem requires a longer wait period, you
must start your chat script with delays (`\d' in a send string).
The chat script will be read from and sent to the port specified by the
dial-device
command for the port, if there is one.
The following escape addition escape sequences may appear in send
strings:
carrier
command in the port file,
\M and \m are ignored. If both the port and the dialer
support carrier, as set by the carrier
command in the port file
and the carrier
command in the dialer file, then every chat
script implicitly begins with \M and ends with \m. There is
no default chat script for dialers.
The following additional escape sequences may be used in
chat-program
:
dialtone string
pause string
carrier boolean
uucico
will require
that carrier be on. One some systems, it will be able to wait for it.
If the argument is false, carrier will not be required. The default is
true.
carrier-wait number
dtr-toggle boolean boolean
complete-chat strings
complete-chat-timeout number
complete-chat-fail string
complete-chat-seven-bit boolean
complete-chat-program strings
complete string
complete-chat
. It is equivalent to
complete-chat "" string
; this has the effect of sending
string to the modem when a call finishes normally.
abort-chat strings
abort-chat-timeout number
abort-chat-fail string
abort-chat-seven-bit boolean
abort-chat-program strings
abort string
abort-chat
. It is equivalent to
abort-chat "" string
; this has the effect of sending
string to the modem when a call is aborted.
protocol-parameter character strings
protocol-parameter
command
in the system configuration file or the port configuration file; see
section Protocol Selection. These parameters take precedence, then those
for the port, then those for the system.
seven-bit boolean
reliable boolean
half-duplex boolean [ any type ]
If your system has a Berkeley style socket library, or a System V style TLI interface library, you can compile the code to permit making connections over TCP. Specifying that a system should be reached via TCP is easy, but nonobvious.
If you are using the new style configuration files (see section Taylor UUCP Configuration Files), add the line `port type tcp' to the entry in the `sys' file. By default UUCP will get the port number by looking up `uucp' in `/etc/services'; if the `uucp' service is not defined, port 540 will be used. You can set the port number to use with the command `port service xxx', where xxx can be either a number or a name to look up in `/etc/services'. You can specify the address of the remote host with `address a.b.c'; if you don't give an address, the remote system name will be used. You should give an explicit chat script for the system when you use TCP; the default chat script begins with a carriage return, which will not work with some UUCP TCP servers.
If you are using V2 configuration files, add a line like this to `L.sys':
sys Any TCP uucp host.domain chat-script
This will make an entry for system sys, to be called at any time, over TCP, using port number `uucp' (as found in `/etc/services'; this may be specified as a number), using remote host `host.domain', with some chat script.
If you are using HDB configuration files, add a line like this to Systems:
sys Any TCP - host.domain chat-script
and a line like this to `Devices':
TCP uucp - -
You only need one line in `Devices' regardless of how many systems you contact over TCP. This will make an entry for system sys, to be called at any time, over TCP, using port number `uucp' (as found in `/etc/services'; this may be specified as a number), using remote host `host.domain', with some chat script.
The uucico
daemon may be run as a TCP server. To use the default
port number, which is a reserved port, uucico
must be invoked by
the superuser (or it must be set user ID to the superuser, but I don't
recommend doing that).
You must define a port, either using the port file (see section The Port Configuration File),
if you are using the new configuration method, or with an entry in
`Devices' if you are using HDB; there is no way to define a port
using V2. If you are using HDB the port must be named `TCP'; a
line as shown above will suffice. You can then start uucico
as
`uucico -p TCP' (after the `-p', name the port; in HDB it must
be `TCP'). This will wait for incoming connections, and fork off a
child for each one. Each connection will be prompted with `login:'
and `Password:'; the results will be checked against the UUCP (not
the system) password file (see section Configuration File Names).
Another way to run a UUCP TCP server is to use the BSD uucpd
program.
Yet another way to run a UUCP TCP server is to use inetd
.
Arrange for inetd
to start up uucico
with the `-l'
switch. This will cause uucico
to prompt with `login:' and
`Password:' and check the results against the UUCP (not the system)
password file (you may want to also use the `-D' switch to avoid a
fork, which in this case is unnecessary).
This discussion of UUCP security applies only to Unix. It is a bit cursory; suggestions for improvement are solicited.
UUCP is traditionally not very secure. Taylor UUCP addresses some security issues, but is still far from being a secure system.
If security is very important to you, then you should not permit any external access to your computer, including UUCP. Any opening to the outside world is a potential security risk.
When local users use UUCP to transfer files, Taylor UUCP can do little
to secure them from each other. You can allow somewhat increased
security by putting the owner of the UUCP programs (normally
uucp
) into a separate group; the use of this is explained in the
following paragraphs, which refer to this separate group as
uucp-group
.
When the uucp
program is invoked to copy a file to a remote
system, it will, by default, copy the file into the UUCP spool
directory. When the uux
program is used, the `-C' switch
must be used to copy the file into the UUCP spool directory. In any
case, once the file has been copied into the spool directory, other
local users will not be able to access it.
When a file is requested from a remote system, UUCP will only permit it
to be placed in a directory which is writable by the requesting user.
The directory must also be writable by UUCP. A local user can create a
directory with a group of uucp-group
and set the mode to permit
group write access. This will allow the file be requested without
permitting it to be viewed by any other user.
There is no provision for security for uucp
requests (as opposed
to uux
requests) made by a user on a remote system. A file sent
over by a remote request may only be placed in a directory which is
world writable, and the file will be world readable and writable. This
will permit any local user to destroy or replace the contents of the
file. A file requested by a remote system must be world readable, and
the directory it is in must be world readable. Any local user will be
able to examine, although not necessarily modify, the file before it is
sent.
There are some security holes and race conditions that apply to the above discussion which I will not elaborate on. They are not hidden from anybody who reads the source code, but they are somewhat technical and difficult (though scarcely impossible) to exploit. Suffice it to say that even under the best of conditions UUCP is not completely secure.
For many sites, security from remote sites is a more important consideration. Fortunately, Taylor UUCP does provide some support in this area.
The greatest security is provided by always dialing out to the other site. This prevents anybody from pretending to be the other site. Of course, only one side of the connection can do this.
If remote dialins must be permitted, then it is best if the dialin line
is used only for UUCP. If this is the case, then you should create a
call-in password file (see section Configuration File Names) and let
uucico
do its own login prompting. For example, to let remote
sites log in on a port named `entry' in the port file (see section The Port Configuration File), you might invoke `uucico -e -p entry'. This would cause
uucico
to enter an endless loop of login prompts and daemon
executions. The advantage of this approach is that even if remote users
break into the system by guessing or learning the password, they will
only be able to do whatever uucico
permits them to do. They will
not be able to start a shell on your system.
If remote users can dial in and log on to your system, then you have a security hazard more serious than that posed by UUCP. But then, you probably knew that already.
Once your system has connected with the remote UUCP, there is a fair
amount of control you can exercise. You can use the remote-send
and remote-receive
commands to control the directories the remote
UUCP can access. You can use the request
command to prevent the
remote UUCP from making any requests of your system at all; however, if
you do this it will not even be able to send you mail or news. If you
do permit remote requests, you should be careful to restrict what
commands may be executed at the remote system's request. The default is
rmail
and rnews
, which will suffice for most systems.
If different remote systems call in and they must be granted different
privileges (perhaps some systems are within the same organization and
some are not) then the called-login
command should be used for
each system to require that they use different login names. Otherwise,
it would be simple for a remote system to use the myname
command
and pretend to be a different system. The sequence
command can
be used to detect when one system pretended to be another, but, since
the sequence numbers must be reset manually after a failed handshake,
this can sometimes be more trouble than it's worth.
This chapter describes how the various UUCP protocols work, and discusses some other internal UUCP issues.
This chapter is quite technical. You do not need to understand it, or even read it, in order to use Taylor UUCP. It is intended for people who are interested in how the UUCP code works.
The information in this chapter is posted monthly to the Usenet newsgroups `comp.mail.uucp', `news.answers', and `comp.answers'. The posting is available from any `news.answers' archive site, such as `rtfm.mit.edu'. If you plan to use this information to write a UUCP program, please make sure you get the most recent version of the posting, in case there have been any corrections.
"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.
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. Taylor UUCP uses any alphanumeric character.
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.
See section When to Call.
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.
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.
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:
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.
execve
system call. For some packages this is
the default anyhow.
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'.
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.
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.
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
(see section Protocol Selection)
or a
port (see section The Port Configuration File).
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'.
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.
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).
After the `C' command response has been received (in the `SY' case) or immediately (in an `SN' case) the master will send another command.
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).
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.
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 (see section UUCP Protocol Sources).
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:
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:
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:
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.
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.
The sending UUCP checks the returned character and acts accordingly.
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.
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.
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.
The `i' protocol was written by Ian Lance Taylor (who also wrote this manual). 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:
(packet << 3) + locchan
(ack << 3) + remchan
(type << 5) + (caller << 4) + len1
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:
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.
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:
(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 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:
The receiver decodes the index bytes as follows (this is the reverse of the operations performed by the sender, presented here for additional clarity):
040 <= index-high < 0176
, the index refers to the
data byte at position (index-high - 040) * 040 +
index-low % 040
.
040 <= index-low < 0100
, then 0200 must be added
to indexed byte.
0100 <= index-low < 0140
, then 020 must be xor'ed
to the indexed byte.
0140 <= index-low < 0177
, then 0200 must be added
to the indexed byte, and 020 must be xor'ed to the indexed byte.
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.
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.
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:
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:
A length field with the high bit set is a control packet. The following control packet types are defined:
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.
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.
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.
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.
This chapter provides the briefest of guides to the Taylor UUCP source code itself.
The code is carefully segregated into a system independent portion and a system dependent portion. The system dependent code is in the `unix' subdirectory, and also in the file `sysh.unx' (also known as `sysdep.h').
With the right configuration parameters, the system independent code
calls only ANSI C functions. Some of the less common ANSI C functions
are also provided in the `lib' directory. The replacement function
strtol
in `lib/strtol.c' assumes that the characters A
to F and a to f appear in strictly sequential order.
The function igradecmp
in `uuconf/grdcmp.c' assumes that the
upper and lower case letters appear in order. Both assumptions are true
for ASCII and EBCDIC, but neither is guaranteed by ANSI C. Disregarding
these caveats, I believe that the system independent portion of the code
is strictly conforming.
That's not too exciting, since all the work is done in the system dependent code. I think that this code can conform to POSIX 1003.1, given the right compilation parameters. I'm a bit less certain about this, though.
The code has been used on a 16 bit segmented system with no function prototypes, so I'm fairly certain that all casts to long and pointers are done when necessary.
I use a modified Hungarian naming convention for my variables and functions. As with all naming conventions, the code is rather opaque if you are not familiar with it, but becomes clear and easy to use with time.
The first character indicates the type of the variable (or function return value). Sometimes additional characters are used. I use the following type prefixes:
A generic pointer (p
) is sometimes a void *
, sometimes a
function pointer in which case the prefix is pf, and sometimes a pointer
to another type, in which case the next character is the type to which
it points (pf is overloaded).
An array of strings (char *[]
) would be named az
(array of
string). If this array were passed to a function, the function
parameter would be named paz
(pointer to array of string).
Note that the variable name prefixes do not necessarily indicate the
type of the variable. For example, a variable prefixed with i may
be int, long or short. Similarly, a variable prefixed with b may
be a char or an int; for example, the return value of getchar
would be caught in an int variable prefixed with b.
For a non-local variable (extern or file static), the first character after the type prefix is capitalized.
Most static variables and functions use another letter after the type prefix to indicate which module they come from. This is to help distinguish different names in the debugger. For example, all static functions in `protg.c', the `g' protocol source code, use a module prefix of `g'. This isn't too useful, as a number of modules use a module prefix of `s'.
I am always grateful for any patches sent in. Much of the flexibility and portability of the code is due to other people. Please do not hesitate to send me any changes you have found necessary or useful.
When sending a patch, please send the output of the Unix diff
program invoked with the `-c' option (if you have the GNU version
of diff
, use the `-p' option). Always invoke diff
with the original file first and the modified file second.
If your diff
does not support `-c' (or you don't have
diff
), send a complete copy of the modified file (if you have
just changed a single function, you can just send the new version of the
function). In particular, please do not send diff
output without
the `-c' option, as it is useless.
If you have made a number of changes, it is very convenient for me if you send each change as a separate mail message. Sometimes I will think that one change is useful but another one is not. If they are in different messages it is much easier for me to apply one but not the other.
I rarely apply the patches directly. Instead I work my way through the hunks and apply each one separately. This ensures that the naming remains consistent, and that I understand all the code.
If you can not follow all these rules, then don't. But if you do, it makes it more likely that I will incorporate your changes. I am not paid for my UUCP work, and my available time is unfortunately very restricted. The package is important to me, and I do what I can, but I can not do all that I would like, much less all that everybody else would like.
Finally, please do not be offended if I do not reply to messages for some time, even a few weeks. I am often behind on my mail, and if I think your message deserves a considered reply I will often put it aside until I have time to deal with it.
This is a list of people who gave help or suggestions while I was working on the Taylor UUCP project. Appearance on this list does not constitute endorsement of the program, particularly since some of the comments were criticisms. I've probably left some people off, and I apologize for any oversight; it does not mean your contribution was unappreciated.
First of all, I would like to thank the people at Infinity Development
Systems (formerly AIRS, which lives on in the domain name) for
permitting me to use their computers and `uunet' access. I would
also like to thank Richard Stallman <rms@gnu.ai.mit.edu>
for
founding the Free Software Foundation, and John Gilmore
<gnu@cygnus.com>
for writing the initial version of gnuucp which
was a direct inspiration for this somewhat larger project. Chip
Salzenberg <chip@tct.com>
has contributed many patches.
Pinard <pinard@iro.umontreal.ca>
tirelessly tested the code and
suggested many improvements. He also put together the initial version
of this manual. Doug Evans contributed the zmodem protocol. Marc
Boucher <marc@CAM.ORG>
contributed the code supporting the pipe
port type. Jorge Cwik jorge@laser.satlink.net
contributed the
`y' protocol code. Finally, Verbus M. Counts
<verbus@westmark.com>
and Centel Federal Systems, Inc., deserve
special thanks, since they actually paid me money to port this code to
System III.
In alphabetical order:
"Earle F. Ake - SAIC"<ake@Dayton.SAIC.COM>
mra@searchtech.com
(Michael Almond)cambler@zeus.calpoly.edu
(Christopher J. Ambler) Brian W. Antoine<briana@tau-ceti.isc-br.com>
jantypas@soft21.s21.com
(John Antypas)james@bigtex.cactus.org
(James Van Artsdalen)jima@netcom.com
(Jim Avera)nba@sysware.DK
(Niels Baggesen)uunet!hotmomma!sdb
(Scott Ballantyne) Zacharias Beckman<zac@dolphin.com>
mike@mbsun.ann-arbor.mi.us
(Mike Bernson)bob@usixth.sublink.org
(Roberto Biancardi)statsci!scott@coco.ms.washington.edu
(Scott Blachowicz)bag%wood2.cs.kiev.ua@relay.ussr.eu.net
(Andrey G Blochintsev)spider@Orb.Nashua.NH.US
(Spider Boardman) Gregory Bond<gnb@bby.com.au>
Marc Boucher<marc@CAM.ORG>
Ard van Breemen<ard@cstmel.hobby.nl>
dean@coplex.com
(Dean Brooks)jbrow@radical.com
(Jim Brownfield)dave@dlb.com
(Dave Buck)gordon@sneaky.lonestar.org
(Gordon Burditt)dburr@sbphy.physics.ucsb.edu
(Donald Burr)mib@gnu.ai.mit.edu
(Michael I Bushnell) Brian Campbell<brianc@quantum.on.ca>
Andrew A. Chernov<ache@astral.msk.su>
jhc@iscp.bellcore.com
(Jonathan Clark)mafc!frank@bach.helios.de
(Frank Conrad) Ed Carp<erc@apple.com>
mpc@mbs.linet.org
(Mark Clements)verbus@westmark.westmark.com
(Verbus M. Counts)cbmvax!snark.thyrsus.com!cowan
(John Cowan) Bob Cunningham<bob@soest.hawaii.edu>
jorge@laser.satlink.net
(Jorge Cwik)kdburg@incoahe.hanse.de
(Klaus Dahlenburg) Damon<d@exnet.co.uk>
celit!billd@UCSD.EDU
(Bill Davidson)hubert@arakis.fdn.org
(Hubert Delahaye)markd@bushwire.apana.org.au
(Mark Delany) Allen Delaney<allen@brc.ubc.ca>
Gerriet M. Denkmanngerriet@hazel.north.de
denny@dakota.alisa.com
(Bob Denny) Drew Derbyshire<ahd@kew.com>
ssd@nevets.oau.org
(Steven S. Dick)gert@greenie.gold.sub.org
(Gert Doering)gemini@geminix.in-berlin.de
(Uwe Doering) Hans-Dieter Doll<hd2@Insel.DE>
deane@deane.teleride.on.ca
(Dean Edmonds) Mark W. Eichin<eichin@cygnus.com>
erik@pdnfido.fidonet.org
Andrew Evans<andrew@airs.com>
dje@cygnus.com
(Doug Evans) Marc Evans<marc@synergytics.com>
Dan Everhart<dan@dyndata.com>
kksys!kegworks!lfahnoe@cs.umn.edu
(Larry Fahnoe) Matthew Farwell<dylan@ibmpcug.co.uk>
fenner@jazz.psu.edu
(Bill Fenner)jaf@inference.com
(Jose A. Fernandez) "David J. Fiander"<golem!david@news.lsuc.on.ca>
Thomas Fischer<batman@olorin.dark.sub.org>
Mister Flash<flash@sam.imash.ras.ru>
louis@marco.de
(Ju"rgen Fluk)erik@eab.retix.com
(Erik Forsberg)andy@scp.caltech.edu
(Andy Fyfe) Lele Gaifax<piggy@idea.sublink.org>
Peter.Galbavy@micromuse.co.uk
hunter@phoenix.pub.uu.oz.au
(James Gardiner [hunter]) Terry Gardner<cphpcom!tjg01>
dgilbert@gamiga.guelphnet.dweomer.org
(David Gilbert)ol@infopro.spb.su
(Oleg Girko)jimmy@tokyo07.info.com
(Jim Gottlieb) Benoit Grange<ben@fizz.fdn.org>
elg@elgamy.jpunix.com
(Eric Lee Green)ryan@cs.umb.edu
(Daniel R. Guilderson)greg@gagme.chi.il.us
(Gregory Gulik) Richard H. Gumpertz<rhg@cps.com>
Scott Guthridge<scooter@cube.rain.com>
Michael Haberler<mah@parrot.prv.univie.ac.at>
Daniel Hagerty<hag@eddie.mit.edu>
jh@moon.nbn.com
(John Harkin)guy@auspex.auspex.com
(Guy Harris)hsw1@papa.attmail.com
(Stephen Harris) Petri Helenius<pete@fidata.fi>
gabe@edi.com
(B. Gabriel Helou) Bob Hemedinger<bob@dalek.mwc.com>
Andrew Herbert<andrew@werple.pub.uu.oz.au>
kherron@ms.uky.edu
(Kenneth Herron) Peter Honeyman<honey@citi.umich.edu>
jhood@smoke.marlboro.vt.us
(John Hood) Mike Ipatow<mip@fido.itc.e-burg.su>
Bill Irwin<bill@twg.bc.ca>
pmcgw!personal-media.co.jp!ishikawa
(Chiaki Ishikawa)ai@easy.in-chemnitz.de
(Andreas Israel)iverson@lionheart.com
(Tim Iverson)bei@dogface.austin.tx.us
(Bob Izenberg)djamiga!djjames@fsd.com
(D.J.James) Rob Janssen<cmgit!rob@relay.nluug.nl>
harvee!esj
(Eric S Johansson) Kevin Johnson<kjj@pondscum.phx.mcd.mot.com>
rj@rainbow.in-berlin.de
(Robert Joop) Alan Judge<aj@dec4ie.IEunet.ie>
chris@cj_net.in-berlin.de
(Christof Junge) Romain Kang<romain@pyramid.com>
tron@Veritas.COM
(Ronald S. Karr) Brendan Kehoe<brendan@cs.widener.edu>
warlock@csuchico.edu
(John Kennedy)kersing@nlmug.nl.mugnet.org
(Jac Kersing)ok@daveg.PFM-Mainz.de
(Olaf Kirch) Gabor Kiss<kissg@sztaki.hu>
gero@gkminix.han.de
(Gero Kuhlmann)rob@pact.nl
(Rob Kurver) "C.A. Lademann"<cal@zls.gtn.com>
kent@sparky.IMD.Sterling.COM
(Kent Landfield) Tin Le<tin@saigon.com>
lebaron@inrs-telecom.uquebec.ca
(Gregory LeBaron)karl@sugar.NeoSoft.Com
(Karl Lehenbauer)alex@hal.rhein-main.de
(Alexander Lehmann)merlyn@digibd.com
(Merlyn LeRoy)clewis@ferret.ocunix.on.ca
(Chris Lewis)gdonl@ssi1.com
(Don Lewis)libove@libove.det.dec.com
(Jay Vassos-Libove)bruce%blilly@Broadcast.Sony.COM
(Bruce Lilly) Godfrey van der Linden<Godfrey_van_der_Linden@NeXT.COM>
Ted Lindgreen<tlindgreen@encore.nl>
andrew@cubetech.com
(Andrew Loewenstern) "Arne Ludwig"<arne@rrzbu.hanse.de>
Matthew Lyle<matt@mips.mitek.com>
djm@eng.umd.edu
(David J. MacKenzie) John R MacMillan<chance!john@sq.sq.com>
jum@helios.de
(Jens-Uwe Mager) Giles D Malet<shrdlu!gdm@provar.kwnet.on.ca>
mem@mv.MV.COM
(Mark E. Mallett)pepe@dit.upm.es
(Jose A. Manas)peter@xpoint.ruessel.sub.org
(Peter Mandrella)martelli@cadlab.sublink.org
(Alex Martelli) W Christopher Martin<wcm@geek.ca.geac.com>
Yanek Martinson<yanek@mthvax.cs.miami.edu>
thomasm@mechti.wupper.de
(Thomas Mechtersheimer)jm@aristote.univ-paris8.fr
(Jean Mehat)me@halfab.freiburg.sub.org
(Udo Meyer)les@chinet.chi.il.us
(Leslie Mikesell)bug@cyberdex.cuug.ab.ca
(Trever Miller)mmitchel@digi.lonestar.org
(Mitch Mitchell) Emmanuel Mogenet<mgix@krainte.jpn.thomson-di.fr>
rmohr@infoac.rmi.de
(Rupert Mohr) Jason Molenda<molenda@sequent.com>
ianm@icsbelf.co.uk
(Ian Moran)jmorriso@bogomips.ee.ubc.ca
(John Paul Morrison)brian@ilinx.wimsey.bc.ca
(Brian J. Murrell)service@infohh.rmi.de
(Dirk Musstopf)lyndon@cs.athabascau.ca
(Lyndon Nerenberg)rolf@saans.north.de
(Rolf Nerstheimer)tom@smart.bo.open.de
(Thomas Neumann)mnichols@pacesetter.com
Richard E. Nickle<trystro!rick@Think.COM>
stephan@sunlab.ka.sub.org
(Stephan Niemz)nolan@helios.unl.edu
(Michael Nolan) david nugent<david@csource.oz.au>
Jim O'Connor<jim@bahamut.fsc.com>
kevin%kosman.uucp@nrc.com
(Kevin O'Gorman) Petri Ojala<ojala@funet.fi>
oneill@cs.ulowell.edu
(Brian 'Doc' O'Neill)Stephen.Page@prg.oxford.ac.uk
abekas!dragoman!mikep@decwrl.dec.com
(Mike Park) Tim Peifferpeiffer@cs.umn.edu
don@blkhole.resun.com
(Don Phillips) "Mark Pizzolato 415-369-9366"<mark@infocomm.com>
John Plate<plate@infotek.dk>
dplatt@ntg.com
(Dave Platt)eldorado@tharr.UUCP
(Mark Powell) Mark Powell<mark@inet-uk.co.uk>
pozar@kumr.lns.com
(Tim Pozar)joey@tessi.UUCP
(Joey Pruett) Paul Pryorptp@fallschurch-acirs2.army.mil
putsch@uicc.com
(Jeff Putsch)ar@nvmr.robin.de
(Andreas Raab) Jarmo Raiha<jarmo@ksvltd.FI>
James Revell<revell@uunet.uu.net>
Scott Reynolds<scott@clmqt.marquette.Mi.US>
mcr@Sandelman.OCUnix.On.Ca
(Michael Richardson) Kenji Rikitake<kenji@rcac.astem.or.jp>
arnold@cc.gatech.edu
(Arnold Robbins)steve@Nyongwa.cam.org
(Steve M. Robbins) Ollivier Robert<Ollivier.Robert@keltia.frmug.fr.net>
Serge Robyns<sr@denkart.be>
Lawrence E. Rosenman<ler@lerami.lerctr.org>
Jeff Ross<jeff@wisdom.bubble.org>
Aleksey P. Rudnev<alex@kiae.su>
"Heiko W.Rupp"<hwr@pilhuhn.ka.sub.org>
wolfgang@wsrcc.com
(Wolfgang S. Rupprecht)tbr@tfic.bc.ca
(Tom Rushworth)jsacco@ssl.com
(Joseph E. Sacco)rsalz@bbn.com
(Rich Salz) Curt Sampson<curt@portal.ca>
sojurn!mike@hobbes.cert.sei.cmu.edu
(Mike Sangrey) Nickolay Saukh<nms@ussr.EU.net>
heiko@lotte.sax.de
(Heiko Schlittermann) Eric Schnoebelen<eric@cirr.com>
russell@alpha3.ersys.edmonton.ab.ca
(Russell Schulz)scott@geom.umn.edu
Igor V. Semenyuk<iga@argrd0.argonaut.su>
Christopher Sawtell<chris@gerty.equinox.gen.nz>
schuler@bds.sub.org
(Bernd Schuler)uunet!gold.sub.org!root
(Christian Seyb)s4mjs!mjs@nirvo.nirvonics.com
(M. J. Shannon Jr.)shields@tembel.org
(Michael Shields)peter@ficc.ferranti.com
(Peter da Silva)vince@victrola.sea.wa.us
(Vince Skahan)frumious!pat
(Patrick Smith)roscom!monty@bu.edu
(Monty Solomon)sommerfeld@orchard.medford.ma.us
(Bill Sommerfeld) Julian Stacey<stacey@guug.de>
evesg@etlrips.etl.go.jp
(Gjoen Stein) Harlan Stenn<harlan@mumps.pfcs.com>
Ralf Stephan<ralf@ark.abg.sub.org>
johannes@titan.westfalen.de
(Johannes Stille)chs@antic.apu.fi
(Hannu Strang)ralf@reswi.ruhr.de
(Ralf E. Stranzenbach)sullivan@Mathcom.com
(S. Sullivan) Shigeya Suzuki<shigeya@dink.foretune.co.jp>
kls@ditka.Chicago.COM
(Karl Swartz)swiers@plains.NoDak.edu
Oleg Tabarovsky<olg@olghome.pccentre.msk.su>
ikeda@honey.misystems.co.jp
(Takatoshi Ikeda) John Theus<john@theus.rain.com>
rd@aii.com
(Bob Thrush) ppKarsten Thygesen<karthy@dannug.dk>
Graham Toal<gtoal@pizzabox.demon.co.uk>
rmtodd@servalan.servalan.com
(Richard Todd) Martin Tomes<mt00@controls.eurotherm.co.uk>
Len Tower<tower-prep@ai.mit.edu>
Mark Towfiq<justice!towfiq@Eingedi.Newton.MA.US>
mju@mudos.ann-arbor.mi.us
(Marc Unangst) Matthias Urlichs<urlichs@smurf.noris.de>
Tomi Vainio<tomppa@fidata.fi>
a3@a3.xs4all.nl
(Adri Verhoef) Andrew Vignaux<ajv@ferrari.datamark.co.nz>
vogel@omega.ssw.de
(Andreas Vogel) Dima Volodin<dvv@hq.demos.su>
jos@bull.nl
(Jos Vos)jv@nl.net
(Johan Vromans) David Vrona<dave@sashimi.wwa.com>
Marcel.Waldvogel@nice.usergroup.ethz.ch
(Marcel Waldvogel)steve@nshore.org
(Stephen J. Walick)syd@dsinc.dsi.com
(Syd Weinstein)gerben@rna.indiv.nluug.nl
(Gerben Wierda)jbw@cs.bu.edu
(Joe Wells)frnkmth!twwells.com!bill
(T. William Wells) Peter Wemm<Peter_Wemm@zeus.dialix.oz.au>
mauxci!eci386!woods@apple.com
(Greg A. Woods)John.Woods@proteon.com
(John Woods) Michael Yu.Yaroslavtsev<mike@yaranga.ipmce.su>
Alexei K. Yushin<root@july.elis.crimea.ua>
jon@console.ais.org
(Jon Zeeff) Matthias Zepf<agnus@amylnd.stgt.sub.org>
Eric Ziegast<uunet!ziegast>
Jump to: . - / - a - c - d - e - f - g - h - i - j - l - m - n - p - r - s - t - u - v - x - y
Jump to: a - b - c - d - f - h - l - m - n - p - r - s - t - u - v
This document was generated on 7 November 1999 using texi2html 1.56k.