comparison man/tramp.texi @ 64772:060177e931f8

Sync with Tramp 2.0.50.

64771:ef0eb1c0d920

@node Overview
@chapter An overview of @value{tramp}
@cindex overview

After the installation of @value{tramp} into your @value{emacsname}, you
will be able to access files on remote machines as though they were
local.  Access to the remote file system for editing files, version
control, and @command{dired} are transparently enabled.

Your access to the remote machine can be with the @command{rsh},
@command{rlogin}, @command{telnet} programs or with any similar
connection method.  This connection must pass @acronym{ASCII}
successfully to be usable but need not be 8-bit clean.

Communication with this process happens through an
@value{emacsname} buffer, that is, the output from the remote end
goes into a buffer.

@item
The remote host may prompt for a login name (for @command{telnet}).  The
login name is given in the file name, so @value{tramp} sends the login name and
a newline.

@item
The remote host may prompt for a password or pass phrase (for
@command{rsh} or for @command{telnet} after sending the login name).
@value{tramp} displays the prompt in the minibuffer, asking you for the

@item
Again, @value{tramp} transfers the file contents to the remote host either
inline or out-of-band.  This is the reverse of what happens when reading
the file.

@end itemize

I hope this has provided you with a basic overview of what happens
behind the scenes when you open a file with @value{tramp}.

@item @option{sshx}
@cindex method sshx
@cindex sshx method
@cindex Cygwin (with sshx method)

As you would expect, this is similar to @option{ssh}, only a little
different.  Whereas @option{ssh} opens a normal interactive shell on
the remote host, this option uses @samp{ssh -t -t @var{host} -l
@var{user} /bin/sh} to open a connection.  This is useful for users

in without such questions.

This is also useful for Windows users where @command{ssh}, when
invoked from an @value{emacsname} buffer, tells them that it is not
allocating a pseudo tty.  When this happens, the login shell is wont
to not print any shell prompt, which confuses @value{tramp} mightily.  For
reasons unknown, some Windows ports for @command{ssh} (maybe the
Cygwin one) require the doubled @samp{-t} option.

This supports the @samp{-p} kludge.


@item @option{krlogin}

@item @option{scpx} --- @command{ssh} and @command{scp}
@cindex method scpx
@cindex scpx method
@cindex scp (with scpx method)
@cindex ssh (with scpx method)
@cindex Cygwin (with scpx method)

As you would expect, this is similar to @option{scp}, only a little
different.  Whereas @option{scp} opens a normal interactive shell on
the remote host, this option uses @samp{ssh -t -t @var{host} -l
@var{user} /bin/sh} to open a connection.  This is useful for users

This is also useful for Windows users where @command{ssh}, when
invoked from an @value{emacsname} buffer, tells them that it is not
allocating a pseudo tty.  When this happens, the login shell is wont
to not print any shell prompt, which confuses @value{tramp} mightily.
Maybe this applies to the Cygwin port of SSH.

This method supports the @samp{-p} hack.


@item @option{pscp} --- @command{plink} and @command{pscp}

(add-to-list
 'tramp-multi-connection-function-alist
 '("sshf" tramp-multi-connect-rlogin "ssh %h -l %u -p 4400%n"))
@end lisp

Now you can use an @code{sshf} hop which connects to port 4400 instead of
the standard port.


@node Default Method
@section Selecting a default method

Another consideration with the selection of transfer methods is the
environment you will use them in and, especially when used over the
Internet, the security implications of your preferred method.

The @command{rsh} and @command{telnet} methods send your password as
plain text as you log in to the remote machine, as well as transferring
the files in such a way that the content can easily be read from other
machines.

If you need to connect to remote systems that are accessible from the
Internet, you should give serious thought to using @command{ssh} based
methods to connect. These provide a much higher level of security,
making it a non-trivial exercise for someone to obtain your password or
read the content of the files you are editing.


@subsection Which method is the right one for me?
@cindex choosing the right method

My suggestion is to use an inline method.  For large files, out-of-band
methods might be more efficient, but I guess that most people will want
to edit mostly small files.

I guess that these days, most people can access a remote machine by
using @code{ssh}.  So I suggest that you use the @code{ssh} method.
So, type @kbd{C-x C-f
@value{prefix}ssh@value{postfixsinglehop}root@@otherhost@value{postfix}/etc/motd
@key{RET}} to edit the @file{/etc/motd} file on the other host.

If you can't use @code{ssh} to log in to the remote host, then select a
method that uses a program that works.  For instance, Windows users
might like the @code{plink} method which uses the PuTTY implementation
of @code{ssh}.  Or you use Kerberos and thus like @code{krlogin}.

For the special case of editing files on the local host as another
user, see the @code{su} or @code{sudo} method.  It offers shortened
syntax for the @samp{root} account, like
@file{@value{prefix}su@value{postfixsinglehop}@value{postfix}/etc/motd}.

People who edit large files may want to consider @code{scp} instead of
@code{ssh}, or @code{pscp} instead of @code{plink}.  These out-of-band
methods are faster than inline methods for large files.  Note, however,
that out-of-band methods suffer from some limitations.  Please try
first whether you really get a noticeable speed advantage from using an
out-of-band method!  Maybe even for large files, inline methods are
fast enough.


@node Customizing Methods
@section Using Non-Standard Methods
@cindex customizing methods

@emph{all} situations.  The other strategy is to require you to set up
the remote host such that it behaves like @value{tramp} expects.  This might
be inconvenient because you have to invest a lot of effort into shell
setup before you can begin to use @value{tramp}.

The package, therefore, pursues a combined approach.  It tries to figure
out some of the more common setups, and only requires you to avoid
really exotic stuff.  For example, it looks through a list of
directories to find some programs on the remote host.  And also, it
knows that it is not obvious how to check whether a file exists, and
therefore it tries different possibilities.  (On some hosts and shells,
the command @code{test -e} does the trick, on some hosts the shell
builtin doesn't work but the program @code{/usr/bin/test -e} or
@code{/bin/test -e} works.  And on still other hosts, @code{ls -d} is
the right way to do this.)

Below you find a discussion of a few things that @value{tramp} does not deal
with, and that you therefore have to set up correctly.

@table @asis

different user.  The default value of
@code{tramp-shell-prompt-pattern} is the same as the default value of
@code{shell-prompt-pattern}, which is reported to work well in many
circumstances.

@item @code{tset} and other questions
@cindex Unix command tset
@cindex tset Unix command

Some people invoke the @code{tset} program from their shell startup
scripts which asks the user about the terminal type of the shell.
Maybe some shells ask other questions when they are started.  @value{tramp}
does not know how to answer these questions.  There are two approaches
for dealing with this problem.  One approach is to take care that the
shell does not ask any questions when invoked from @value{tramp}.  You can

@item Non-Bourne commands in @file{.profile}

After logging in to the remote host, @value{tramp} issues the command
@code{exec /bin/sh}.  (Actually, the command is slightly different.)
When @code{/bin/sh} is executed, it reads some init files, such as
@file{~/.shrc} or @file{~/.profile}.

Now, some people have a login shell which is not @code{/bin/sh} but a
Bourne-ish shell such as bash or ksh.  Some of these people might put
their shell setup into the files @code{~/.shrc} or @code{~/.profile}.
This way, it is possible for non-Bourne constructs to end up in those
files.  Then, @code{exec /bin/sh} might cause the Bourne shell to barf
on those constructs.

As an example, imagine somebody putting @code{export FOO=bar} into the
file @file{~/.profile}.  The standard Bourne shell does not understand
this syntax and will emit a syntax error when it reaches this line.

Another example is the tilde (@code{~}) character, say when adding
@file{~/bin} to @code{$PATH}.  Many Bourne shells will not expand this
character, and since there is usually no directory whose name consists
of the single character tilde, strange things will happen.

What can you do about this?

Well, one possibility is to make sure that everything in @file{~/.shrc}
and @file{~/.profile} on all remote hosts is Bourne-compatible.  In the
above example, instead of @code{export FOO=bar}, you might use
@code{FOO=bar; export FOO} instead.

The other possibility is to put your non-Bourne shell setup into some
other files.  For example, bash reads the file @file{~/.bash_profile}
instead of @file{~/.profile}, if the former exists.  So bash
aficionados just rename their @file{~/.profile} to
@file{~/.bash_profile} on all remote hosts, and Bob's your uncle.

The @value{tramp} developers would like to circumvent this problem, so if you
have an idea about it, please tell us.  However, we are afraid it is not
that simple: before saying @code{exec /bin/sh}, @value{tramp} does not know
which kind of shell it might be talking to.  It could be a Bourne-ish
shell like ksh or bash, or it could be a csh derivative like tcsh, or
it could be zsh, or even rc.  If the shell is Bourne-ish already, then
it might be prudent to omit the @code{exec /bin/sh} step.  But how to
find out if the shell is Bourne-ish?

@end table


@node Auto-save and Backup

This section needs a lot of work!  Please help.

@cindex method sshx with Cygwin
@cindex sshx method with Cygwin
If you use the Cygwin installation of ssh (you have to explicitly select
it in the installer), then it should work out of the box to just select
@code{sshx} as the connection method.  You can find information about
setting up Cygwin in their FAQ at @uref{http://cygwin.com/faq/}.

@cindex method scpx with Cygwin
@cindex scpx method with Cygwin
If you wish to use the @code{scpx} connection method, then you might
have the problem that @value{emacsname} calls @code{scp} with a
Windows filename such as @code{c:/foo}.  The Cygwin version of
@code{scp} does not know about Windows filenames and interprets this
as a remote filename on the host @code{c}.

One possible workaround is to write a wrapper script for @code{scp}
which converts the Windows filename to a Cygwinized filename.

I guess that another workaround is to run @value{emacsname} under
Cygwin, or to run a Cygwinized @value{emacsname}.

@cindex Cygwin and ssh-agent
@cindex SSH_AUTH_SOCK and @value{emacsname} on Windows
If you want to use either @code{ssh} based method on Windows, then you
might encounter problems with @code{ssh-agent}.  Using this program,
you can avoid typing the pass-phrase every time you log in.  However,
if you start @value{emacsname} from a desktop shortcut, then the
environment variable @code{SSH_AUTH_SOCK} is not set and so
@value{emacsname} and thus @value{tramp} and thus @code{ssh} and
@code{scp} started from @value{tramp} cannot communicate with
@code{ssh-agent}.  It works better to start @value{emacsname} from
the shell.

If anyone knows how to start @code{ssh-agent} under Windows in such a
way that desktop shortcuts can profit, please holler.  I don't really
know anything at all about Windows@dots{}


@node Usage

@item
There's this @file{~/.sh_history} file on the remote host which keeps
growing and growing.  What's that?

Sometimes, @value{tramp} starts @code{ksh} on the remote host for tilde
expansion.  Maybe @code{ksh} saves the history by default.  @value{tramp}
tries to turn off saving the history, but maybe you have to help.  For
example, you could put this in your @file{.kshrc}:

@example
if [ -f $HOME/.sh_history ] ; then
   /bin/rm $HOME/.sh_history
fi

@chapter Debatable Issues and What Was Decided

@itemize @bullet
@item The uuencode method does not always work.

Due to the design of @value{tramp}, the encoding and decoding programs need to
read from stdin and write to stdout.  On some systems, @code{uudecode -o
-} will read stdin and write the decoded file to stdout, on other
systems @code{uudecode -p} does the same thing.  But some systems have
uudecode implementations which cannot do this at all---it is not
possible to call these uudecode implementations with suitable parameters
so that they write to stdout.

Of course, this could be circumvented: the @code{begin foo 644} line
could be rewritten to put in some temporary file name, then
@code{uudecode} could be called, then the temp file could be printed and
deleted.

But I have decided that this is too fragile to reliably work, so on some
systems you'll have to do without the uuencode methods.

@item @value{tramp} does not work on XEmacs 20.

disabled.  This can be achieved by setting file permissions @code{000}
to the files @file{.../xemacs-packages/lisp/tramp/auto-autoloads.el*}.

In case of unified filenames, all @value{emacsname} download sites
are added to @code{tramp-default-method-alist} with default method
@code{ftp} @xref{Default Method}.  These settings shouldn't be touched
for proper working of the @value{emacsname} package system.

The syntax for unified filenames is described in the @value{tramp} manual
for @value{emacsothername}.
@end ifset

@end itemize

@node Concept Index
@comment node-name,    next,  previous,      up
@unnumbered Concept Index

64772:060177e931f8

@node Overview
@chapter An overview of @value{tramp}
@cindex overview

After the installation of @value{tramp} into your @value{emacsname},
you will be able to access files on remote machines as though they
were local.  Access to the remote file system for editing files,
version control, and @code{dired} are transparently enabled.

Your access to the remote machine can be with the @command{rsh},
@command{rlogin}, @command{telnet} programs or with any similar
connection method.  This connection must pass @acronym{ASCII}
successfully to be usable but need not be 8-bit clean.

Communication with this process happens through an
@value{emacsname} buffer, that is, the output from the remote end
goes into a buffer.

@item
The remote host may prompt for a login name (for @command{telnet}).
The login name is given in the file name, so @value{tramp} sends the
login name and a newline.

@item
The remote host may prompt for a password or pass phrase (for
@command{rsh} or for @command{telnet} after sending the login name).
@value{tramp} displays the prompt in the minibuffer, asking you for the

@item
Again, @value{tramp} transfers the file contents to the remote host either
inline or out-of-band.  This is the reverse of what happens when reading
the file.
@end itemize

I hope this has provided you with a basic overview of what happens
behind the scenes when you open a file with @value{tramp}.

@item @option{sshx}
@cindex method sshx
@cindex sshx method

As you would expect, this is similar to @option{ssh}, only a little
different.  Whereas @option{ssh} opens a normal interactive shell on
the remote host, this option uses @samp{ssh -t -t @var{host} -l
@var{user} /bin/sh} to open a connection.  This is useful for users

in without such questions.

This is also useful for Windows users where @command{ssh}, when
invoked from an @value{emacsname} buffer, tells them that it is not
allocating a pseudo tty.  When this happens, the login shell is wont
to not print any shell prompt, which confuses @value{tramp} mightily.
For reasons unknown, some Windows ports for @command{ssh} require the
doubled @samp{-t} option.

This supports the @samp{-p} kludge.


@item @option{krlogin}

@item @option{scpx} --- @command{ssh} and @command{scp}
@cindex method scpx
@cindex scpx method
@cindex scp (with scpx method)
@cindex ssh (with scpx method)

As you would expect, this is similar to @option{scp}, only a little
different.  Whereas @option{scp} opens a normal interactive shell on
the remote host, this option uses @samp{ssh -t -t @var{host} -l
@var{user} /bin/sh} to open a connection.  This is useful for users

This is also useful for Windows users where @command{ssh}, when
invoked from an @value{emacsname} buffer, tells them that it is not
allocating a pseudo tty.  When this happens, the login shell is wont
to not print any shell prompt, which confuses @value{tramp} mightily.

This method supports the @samp{-p} hack.


@item @option{pscp} --- @command{plink} and @command{pscp}

(add-to-list
 'tramp-multi-connection-function-alist
 '("sshf" tramp-multi-connect-rlogin "ssh %h -l %u -p 4400%n"))
@end lisp

Now you can use an @option{sshf} hop which connects to port 4400 instead of
the standard port.


@node Default Method
@section Selecting a default method

Another consideration with the selection of transfer methods is the
environment you will use them in and, especially when used over the
Internet, the security implications of your preferred method.

The @option{rsh} and @option{telnet} methods send your password as
plain text as you log in to the remote machine, as well as
transferring the files in such a way that the content can easily be
read from other machines.

If you need to connect to remote systems that are accessible from the
Internet, you should give serious thought to using @option{ssh} based
methods to connect. These provide a much higher level of security,
making it a non-trivial exercise for someone to obtain your password
or read the content of the files you are editing.


@subsection Which method is the right one for me?
@cindex choosing the right method

My suggestion is to use an inline method.  For large files, out-of-band
methods might be more efficient, but I guess that most people will want
to edit mostly small files.

I guess that these days, most people can access a remote machine by
using @command{ssh}.  So I suggest that you use the @option{ssh}
method.  So, type @kbd{C-x C-f
@value{prefix}ssh@value{postfixsinglehop}root@@otherhost@value{postfix}/etc/motd
@key{RET}} to edit the @file{/etc/motd} file on the other host.

If you can't use @option{ssh} to log in to the remote host, then
select a method that uses a program that works.  For instance, Windows
users might like the @option{plink} method which uses the PuTTY
implementation of @command{ssh}.  Or you use Kerberos and thus like
@option{krlogin}.

For the special case of editing files on the local host as another
user, see the @option{su} or @option{sudo} methods.  They offer
shortened syntax for the @samp{root} account, like
@file{@value{prefix}su@value{postfixsinglehop}@value{postfix}/etc/motd}.

People who edit large files may want to consider @option{scp} instead
of @option{ssh}, or @option{pscp} instead of @option{plink}.  These
out-of-band methods are faster than inline methods for large files.
Note, however, that out-of-band methods suffer from some limitations.
Please try first whether you really get a noticeable speed advantage
from using an out-of-band method!  Maybe even for large files, inline
methods are fast enough.


@node Customizing Methods
@section Using Non-Standard Methods
@cindex customizing methods

@emph{all} situations.  The other strategy is to require you to set up
the remote host such that it behaves like @value{tramp} expects.  This might
be inconvenient because you have to invest a lot of effort into shell
setup before you can begin to use @value{tramp}.

The package, therefore, pursues a combined approach.  It tries to
figure out some of the more common setups, and only requires you to
avoid really exotic stuff.  For example, it looks through a list of
directories to find some programs on the remote host.  And also, it
knows that it is not obvious how to check whether a file exists, and
therefore it tries different possibilities.  (On some hosts and
shells, the command @command{test -e} does the trick, on some hosts
the shell builtin doesn't work but the program @command{/usr/bin/test
-e} or @command{/bin/test -e} works.  And on still other hosts,
@command{ls -d} is the right way to do this.)

Below you find a discussion of a few things that @value{tramp} does not deal
with, and that you therefore have to set up correctly.

@table @asis

different user.  The default value of
@code{tramp-shell-prompt-pattern} is the same as the default value of
@code{shell-prompt-pattern}, which is reported to work well in many
circumstances.

@item @command{tset} and other questions
@cindex Unix command tset
@cindex tset Unix command

Some people invoke the @command{tset} program from their shell startup
scripts which asks the user about the terminal type of the shell.
Maybe some shells ask other questions when they are started.  @value{tramp}
does not know how to answer these questions.  There are two approaches
for dealing with this problem.  One approach is to take care that the
shell does not ask any questions when invoked from @value{tramp}.  You can

@item Non-Bourne commands in @file{.profile}

After logging in to the remote host, @value{tramp} issues the command
@command{exec /bin/sh}.  (Actually, the command is slightly
different.)  When @command{/bin/sh} is executed, it reads some init
files, such as @file{~/.shrc} or @file{~/.profile}.

Now, some people have a login shell which is not @code{/bin/sh} but a
Bourne-ish shell such as bash or ksh.  Some of these people might put
their shell setup into the files @file{~/.shrc} or @file{~/.profile}.
This way, it is possible for non-Bourne constructs to end up in those
files.  Then, @command{exec /bin/sh} might cause the Bourne shell to
barf on those constructs.

As an example, imagine somebody putting @command{export FOO=bar} into
the file @file{~/.profile}.  The standard Bourne shell does not
understand this syntax and will emit a syntax error when it reaches
this line.

Another example is the tilde (@code{~}) character, say when adding
@file{~/bin} to @code{$PATH}.  Many Bourne shells will not expand this
character, and since there is usually no directory whose name consists
of the single character tilde, strange things will happen.

What can you do about this?

Well, one possibility is to make sure that everything in @file{~/.shrc}
and @file{~/.profile} on all remote hosts is Bourne-compatible.  In the
above example, instead of @command{export FOO=bar}, you might use
@command{FOO=bar; export FOO} instead.

The other possibility is to put your non-Bourne shell setup into some
other files.  For example, bash reads the file @file{~/.bash_profile}
instead of @file{~/.profile}, if the former exists.  So bash
aficionados just rename their @file{~/.profile} to
@file{~/.bash_profile} on all remote hosts, and Bob's your uncle.

The @value{tramp} developers would like to circumvent this problem, so
if you have an idea about it, please tell us.  However, we are afraid
it is not that simple: before saying @command{exec /bin/sh},
@value{tramp} does not know which kind of shell it might be talking
to.  It could be a Bourne-ish shell like ksh or bash, or it could be a
csh derivative like tcsh, or it could be zsh, or even rc.  If the
shell is Bourne-ish already, then it might be prudent to omit the
@command{exec /bin/sh} step.  But how to find out if the shell is
Bourne-ish?

@end table


@node Auto-save and Backup

This section needs a lot of work!  Please help.

@cindex method sshx with Cygwin
@cindex sshx method with Cygwin
The recent Cygwin installation of @command{ssh} works only with a
Cygwinized @value{emacsname}.  You can check it by typing @kbd{M-x
eshell}, and starting @kbd{ssh test.machine}.  The problem is evident
if you see a message like this:

@example
Pseudo-terminal will not be allocated because stdin is not a terminal.
@end example

Older @command{ssh} versions of Cygwin are told to cooperate with
@value{tramp} selecting @option{sshx} as the connection method.  You
can find information about setting up Cygwin in their FAQ at
@uref{http://cygwin.com/faq/}.

@cindex method scpx with Cygwin
@cindex scpx method with Cygwin
If you wish to use the @option{scpx} connection method, then you might
have the problem that @value{emacsname} calls @command{scp} with a
Windows filename such as @code{c:/foo}.  The Cygwin version of
@command{scp} does not know about Windows filenames and interprets this
as a remote filename on the host @code{c}.

One possible workaround is to write a wrapper script for @option{scp}
which converts the Windows filename to a Cygwinized filename.

@cindex Cygwin and ssh-agent
@cindex SSH_AUTH_SOCK and @value{emacsname} on Windows
If you want to use either @option{ssh} based method on Windows, then
you might encounter problems with @command{ssh-agent}.  Using this
program, you can avoid typing the pass-phrase every time you log in.
However, if you start @value{emacsname} from a desktop shortcut, then
the environment variable @code{SSH_AUTH_SOCK} is not set and so
@value{emacsname} and thus @value{tramp} and thus @command{ssh} and
@command{scp} started from @value{tramp} cannot communicate with
@command{ssh-agent}.  It works better to start @value{emacsname} from
the shell.

If anyone knows how to start @command{ssh-agent} under Windows in such a
way that desktop shortcuts can profit, please holler.  I don't really
know anything at all about Windows@dots{}


@node Usage

@item
There's this @file{~/.sh_history} file on the remote host which keeps
growing and growing.  What's that?

Sometimes, @value{tramp} starts @command{ksh} on the remote host for
tilde expansion.  Maybe @command{ksh} saves the history by default.
@value{tramp} tries to turn off saving the history, but maybe you have
to help.  For example, you could put this in your @file{.kshrc}:

@example
if [ -f $HOME/.sh_history ] ; then
   /bin/rm $HOME/.sh_history
fi

@chapter Debatable Issues and What Was Decided

@itemize @bullet
@item The uuencode method does not always work.

Due to the design of @value{tramp}, the encoding and decoding programs
need to read from stdin and write to stdout.  On some systems,
@command{uudecode -o -} will read stdin and write the decoded file to
stdout, on other systems @command{uudecode -p} does the same thing.
But some systems have uudecode implementations which cannot do this at
all---it is not possible to call these uudecode implementations with
suitable parameters so that they write to stdout.

Of course, this could be circumvented: the @code{begin foo 644} line
could be rewritten to put in some temporary file name, then
@command{uudecode} could be called, then the temp file could be
printed and deleted.

But I have decided that this is too fragile to reliably work, so on some
systems you'll have to do without the uuencode methods.

@item @value{tramp} does not work on XEmacs 20.

disabled.  This can be achieved by setting file permissions @code{000}
to the files @file{.../xemacs-packages/lisp/tramp/auto-autoloads.el*}.

In case of unified filenames, all @value{emacsname} download sites
are added to @code{tramp-default-method-alist} with default method
@option{ftp} @xref{Default Method}.  These settings shouldn't be touched
for proper working of the @value{emacsname} package system.

The syntax for unified filenames is described in the @value{tramp} manual
for @value{emacsothername}.
@end ifset
@end itemize

@node Concept Index
@comment node-name,    next,  previous,      up
@unnumbered Concept Index