1 \input texinfo @c -*-texinfo-*-
2 @setfilename ../../info/tramp
4 @settitle TRAMP User Manual
7 @c This is *so* much nicer :)
10 @c In the Tramp CVS, the version number is auto-frobbed from
11 @c configure.ac, so you should edit that file and run
12 @c "autoconf && ./configure" to change the version number.
14 @c Additionally, flags are set with respect to the Emacs flavor; and
15 @c depending whether Tramp is packaged into (X)Emacs, or standalone.
17 @include trampver.texi
19 @c Macro for formatting a filename according to the respective syntax.
20 @c xxx and yyy are auxiliary macros in order to omit leading and
21 @c trailing whitespace. Not very elegant, but I don't know it better.
27 @macro yyy {one, two}@c
35 @macro trampfn {method, user, host, localname}@c
36 @value{prefix}@yyy{\method\,@value{postfixhop}}@yyy{\user\,@@}\host\@value{postfix}\localname\@c
40 Copyright @copyright{} 1999-2012 Free Software Foundation, Inc.
43 Permission is granted to copy, distribute and/or modify this document
44 under the terms of the GNU Free Documentation License, Version 1.3 or
45 any later version published by the Free Software Foundation; with no
46 Invariant Sections, with the Front-Cover texts being ``A GNU Manual'',
47 and with the Back-Cover Texts as in (a) below. A copy of the license
48 is included in the section entitled ``GNU Free Documentation License''.
50 (a) The FSF's Back-Cover Text is: ``You have the freedom to
51 copy and modify this GNU manual. Buying copies from the FSF
52 supports it in developing GNU and promoting software freedom.''
56 @c Entries for @command{install-info} to use
57 @dircategory @value{emacsname} network features
59 * TRAMP: (tramp). Transparent Remote Access, Multiple Protocol
60 @value{emacsname} remote file access via rsh and rcp.
64 @title @value{tramp} version @value{trampver} User Manual
65 @author by Daniel Pittman
66 @author based on documentation by Kai Gro@ss{}johann
74 @node Top, Overview, (dir), (dir)
75 @top @value{tramp} version @value{trampver} User Manual
77 This file documents @value{tramp} version @value{trampver}, a remote file
78 editing package for @value{emacsname}.
80 @value{tramp} stands for `Transparent Remote (file) Access, Multiple
81 Protocol'. This package provides remote file editing, similar to
82 @value{ftppackagename}.
84 The difference is that @value{ftppackagename} uses FTP to transfer
85 files between the local and the remote host, whereas @value{tramp} uses a
86 combination of @command{rsh} and @command{rcp} or other work-alike
87 programs, such as @command{ssh}/@command{scp}.
89 You can find the latest version of this document on the web at
90 @uref{http://www.gnu.org/software/tramp/}.
92 @c Pointer to the other Emacs flavor is necessary only in case of
93 @c standalone installation.
95 The manual has been generated for @value{emacsname}.
97 If you want to read the info pages for @value{emacsothername}, you
98 should read in @ref{Installation} how to create them.
101 If you're using the other Emacs flavor, you should read the
102 @uref{@value{emacsotherfilename}, @value{emacsothername}} pages.
107 The latest release of @value{tramp} is available for
108 @uref{ftp://ftp.gnu.org/gnu/tramp/, download}, or you may see
109 @ref{Obtaining Tramp} for more details, including the CVS server
112 @value{tramp} also has a @uref{http://savannah.gnu.org/projects/tramp/,
113 Savannah Project Page}.
116 There is a mailing list for @value{tramp}, available at
117 @email{tramp-devel@@gnu.org}, and archived at
118 @uref{http://lists.gnu.org/archive/html/tramp-devel/, the
119 @value{tramp} Mail Archive}.
121 Older archives are located at
122 @uref{http://sourceforge.net/mailarchive/forum.php?forum=tramp-devel,
123 SourceForge Mail Archive} and
124 @uref{http://www.mail-archive.com/emacs-rcp@@ls6.cs.uni-dortmund.de/,
126 @c in HTML output, there's no new paragraph.
135 * Overview:: What @value{tramp} can and cannot do.
139 * Obtaining Tramp:: How to obtain @value{tramp}.
140 * History:: History of @value{tramp}.
141 @ifset installchapter
142 * Installation:: Installing @value{tramp} with your @value{emacsname}.
144 * Configuration:: Configuring @value{tramp} for use.
145 * Usage:: An overview of the operation of @value{tramp}.
146 * Bug Reports:: Reporting Bugs and Problems.
147 * Frequently Asked Questions:: Questions and answers from the mailing list.
148 * Function Index:: @value{tramp} functions.
149 * Variable Index:: User options and variables.
150 * Concept Index:: An item for each concept.
154 * Files directories and localnames:: How file names, directories and localnames are mangled and managed.
155 * Traces and Profiles:: How to Customize Traces.
156 * Issues:: Debatable Issues and What Was Decided.
158 * GNU Free Documentation License:: The license for this documentation.
161 --- The Detailed Node Listing ---
163 @ifset installchapter
164 Installing @value{tramp} with your @value{emacsname}
166 * Installation parameters:: Parameters in order to control installation.
167 * Load paths:: How to plug-in @value{tramp} into your environment.
171 Configuring @value{tramp} for use
173 * Connection types:: Types of connections made to remote machines.
174 * Inline methods:: Inline methods.
175 * External methods:: External methods.
177 * GVFS based methods:: GVFS based external methods.
180 * Gateway methods:: Gateway methods.
182 * Default Method:: Selecting a default method.
183 * Default User:: Selecting a default user.
184 * Default Host:: Selecting a default host.
185 * Multi-hops:: Connecting to a remote host using multiple hops.
186 * Customizing Methods:: Using Non-Standard Methods.
187 * Customizing Completion:: Selecting config files for user/host name completion.
188 * Password handling:: Reusing passwords for several connections.
189 * Connection caching:: Reusing connection related information.
190 * Remote Programs:: How @value{tramp} finds and uses programs on the remote machine.
191 * Remote shell setup:: Remote shell setup hints.
192 * Windows setup hints:: Issues with Cygwin ssh.
193 * Auto-save and Backup:: Auto-save and Backup.
197 * Filename Syntax:: @value{tramp} filename conventions.
198 * Alternative Syntax:: URL-like filename syntax.
199 * Filename completion:: Filename completion.
200 * Ad-hoc multi-hops:: Declaring multiple hops in the file name.
201 * Remote processes:: Integration with other @value{emacsname} packages.
202 * Cleanup remote connections:: Cleanup remote connections.
204 How file names, directories and localnames are mangled and managed
206 * Localname deconstruction:: Breaking a localname into its components.
208 * External packages:: Integration with external Lisp packages.
215 @chapter An overview of @value{tramp}
218 After the installation of @value{tramp} into your @value{emacsname}, you
219 will be able to access files on remote machines as though they were
220 local. Access to the remote file system for editing files, version
221 control, and @code{dired} are transparently enabled.
223 Your access to the remote machine can be with the @command{rsh},
224 @command{rlogin}, @command{telnet} programs or with any similar
225 connection method. This connection must pass @acronym{ASCII}
226 successfully to be usable but need not be 8-bit clean.
228 The package provides support for @command{ssh} connections out of the
229 box, one of the more common uses of the package. This allows
230 relatively secure access to machines, especially if @command{ftp}
233 Under Windows, @value{tramp} is integrated with the PuTTY package,
234 using the @command{plink} program.
236 The majority of activity carried out by @value{tramp} requires only that
237 the remote login is possible and is carried out at the terminal. In
238 order to access remote files @value{tramp} needs to transfer their content
239 to the local machine temporarily.
241 @value{tramp} can transfer files between the machines in a variety of ways.
242 The details are easy to select, depending on your needs and the
243 machines in question.
245 The fastest transfer methods for large files rely on a remote file
246 transfer package such as @command{rcp}, @command{scp}, @command{rsync}
247 or (under Windows) @command{pscp}.
249 If the remote copy methods are not suitable for you, @value{tramp} also
250 supports the use of encoded transfers directly through the shell.
251 This requires that the @command{mimencode} or @command{uuencode} tools
252 are available on the remote machine. These methods are generally
253 faster for small files.
255 @value{tramp} is still under active development and any problems you encounter,
256 trivial or major, should be reported to the @value{tramp} developers.
260 @subsubheading Behind the scenes
261 @cindex behind the scenes
262 @cindex details of operation
265 This section tries to explain what goes on behind the scenes when you
266 access a remote file through @value{tramp}.
268 Suppose you type @kbd{C-x C-f} and enter part of an @value{tramp} file name,
269 then hit @kbd{@key{TAB}} for completion. Suppose further that this is
270 the first time that @value{tramp} is invoked for the host in question. Here's
275 @value{tramp} discovers that it needs a connection to the host. So it
276 invokes @samp{telnet @var{host}} or @samp{rsh @var{host} -l
277 @var{user}} or a similar tool to connect to the remote host.
278 Communication with this process happens through an
279 @value{emacsname} buffer, that is, the output from the remote end
283 The remote host may prompt for a login name (for @command{telnet}).
284 The login name is given in the file name, so @value{tramp} sends the
285 login name and a newline.
288 The remote host may prompt for a password or pass phrase (for
289 @command{rsh} or for @command{telnet} after sending the login name).
290 @value{tramp} displays the prompt in the minibuffer, asking you for the
291 password or pass phrase.
293 You enter the password or pass phrase. @value{tramp} sends it to the remote
294 host, followed by a newline.
297 @value{tramp} now waits for the shell prompt or for a message that the login
300 If @value{tramp} sees neither of them after a certain period of time
301 (a minute, say), then it issues an error message saying that it
302 couldn't find the remote shell prompt and shows you what the remote
305 If @value{tramp} sees a @samp{login failed} message, it tells you so,
306 aborts the login attempt and allows you to try again.
309 Suppose that the login was successful and @value{tramp} sees the shell prompt
310 from the remote host. Now @value{tramp} invokes @command{/bin/sh} because
311 Bourne shells and C shells have different command
312 syntaxes.@footnote{Invoking @command{/bin/sh} will fail if your login
313 shell doesn't recognize @samp{exec /bin/sh} as a valid command.
314 Maybe you use the Scheme shell @command{scsh}@dots{}}
316 After the Bourne shell has come up, @value{tramp} sends a few commands to
317 ensure a good working environment. It turns off echoing, it sets the
318 shell prompt, and a few other things.
321 Now the remote shell is up and it good working order. Remember, what
322 was supposed to happen is that @value{tramp} tries to find out what files exist
323 on the remote host so that it can do filename completion.
325 So, @value{tramp} basically issues @command{cd} and @command{ls} commands and
326 also sometimes @command{echo} with globbing. Another command that is
327 often used is @command{test} to find out whether a file is writable or a
328 directory or the like. The output of each command is parsed for the
332 Suppose you are finished with filename completion, have entered @kbd{C-x
333 C-f}, a full file name and hit @kbd{@key{RET}}. Now comes the time to
334 transfer the file contents from the remote host to the local host so
335 that you can edit them.
337 See above for an explanation of how @value{tramp} transfers the file contents.
339 For inline transfers, @value{tramp} issues a command like @samp{mimencode -b
340 /path/to/remote/file}, waits until the output has accumulated in the
341 buffer that's used for communication, then decodes that output to
342 produce the file contents.
344 For external transfers, @value{tramp} issues a command like the
347 rcp user@@host:/path/to/remote/file /tmp/tramp.4711
349 It then reads the local temporary file @file{/tmp/tramp.4711} into a
350 buffer and deletes the temporary file.
353 You now edit the buffer contents, blithely unaware of what has happened
354 behind the scenes. (Unless you have read this section, that is.) When
355 you are finished, you type @kbd{C-x C-s} to save the buffer.
358 Again, @value{tramp} transfers the file contents to the remote host
359 either inline or external. This is the reverse of what happens when
363 I hope this has provided you with a basic overview of what happens
364 behind the scenes when you open a file with @value{tramp}.
368 @node Obtaining Tramp
369 @chapter Obtaining Tramp.
370 @cindex obtaining Tramp
372 @value{tramp} is freely available on the Internet and the latest
373 release may be downloaded from @uref{ftp://ftp.gnu.org/gnu/tramp/}.
374 This release includes the full documentation and code for
375 @value{tramp}, suitable for installation. But Emacs (22 or later)
376 includes @value{tramp} already, and there is a @value{tramp} package
377 for XEmacs, as well. So maybe it is easier to just use those. But if
378 you want the bleeding edge, read on@dots{...}
380 For the especially brave, @value{tramp} is available from CVS. The CVS
381 version is the latest version of the code and may contain incomplete
382 features or new issues. Use these versions at your own risk.
384 Instructions for obtaining the latest development version of @value{tramp}
385 from CVS can be found by going to the Savannah project page at the
386 following URL and then clicking on the CVS link in the navigation bar
390 @uref{http://savannah.gnu.org/projects/tramp/}
393 Or follow the example session below:
396 ] @strong{cd ~/@value{emacsdir}}
397 ] @strong{export CVS_RSH="ssh"}
398 ] @strong{cvs -z3 -d:pserver:anonymous@@cvs.savannah.gnu.org:/sources/tramp co tramp}
402 You should now have a directory @file{~/@value{emacsdir}/tramp}
403 containing the latest version of @value{tramp}. You can fetch the latest
404 updates from the repository by issuing the command:
407 ] @strong{cd ~/@value{emacsdir}/tramp}
408 ] @strong{export CVS_RSH="ssh"}
409 ] @strong{cvs update -d}
413 Once you've got updated files from the CVS repository, you need to run
414 @command{autoconf} in order to get an up-to-date @file{configure}
418 ] @strong{cd ~/@value{emacsdir}/tramp}
424 @chapter History of @value{tramp}
426 @cindex development history
428 Development was started end of November 1998. The package was called
429 @file{rssh.el}, back then. It only provided one method to access a
430 file, using @command{ssh} to log in to a remote host and using
431 @command{scp} to transfer the file contents. After a while, the name
432 was changed to @file{rcp.el}, and now it's @value{tramp}. Along the way,
433 many more methods for getting a remote shell and for transferring the
434 file contents were added. Support for VC was added.
436 After that, there were added the multi-hop methods in April 2000 and
437 the unification of @value{tramp} and Ange-FTP filenames in July 2002.
438 In July 2004, multi-hop methods have been replaced by proxy hosts.
439 Running commands on remote hosts was introduced in December 2005.
441 Support of gateways exists since April 2007.
444 GVFS integration started in February 2009.
447 Remote commands on Windows hosts are available since September 2011.
449 Ad-hoc multi-hop methods (with a changed syntax) have been reenabled
452 In December 2001, @value{tramp} has been added to the XEmacs package
453 repository. Being part of the Emacs repository happened in June 2002,
454 the first release including @value{tramp} was Emacs 22.1.
456 @value{tramp} is also a Debian GNU/Linux package since February 2001.
459 @c Installation chapter is necessary only in case of standalone
460 @c installation. Text taken from trampinst.texi.
461 @ifset installchapter
462 @include trampinst.texi
466 @chapter Configuring @value{tramp} for use
467 @cindex configuration
469 @cindex default configuration
470 @value{tramp} is (normally) fully functional when it is initially
471 installed. It is initially configured to use the @command{scp}
472 program to connect to the remote host. So in the easiest case, you
473 just type @kbd{C-x C-f} and then enter the filename
474 @file{@trampfn{, user, machine, /path/to.file}}.
476 On some hosts, there are problems with opening a connection. These are
477 related to the behavior of the remote shell. See @xref{Remote shell
478 setup}, for details on this.
480 If you do not wish to use these commands to connect to the remote
481 host, you should change the default connection and transfer method
482 that @value{tramp} uses. There are several different methods that @value{tramp}
483 can use to connect to remote machines and transfer files
484 (@pxref{Connection types}).
486 If you don't know which method is right for you, see @xref{Default
491 * Connection types:: Types of connections made to remote machines.
492 * Inline methods:: Inline methods.
493 * External methods:: External methods.
495 * GVFS based methods:: GVFS based external methods.
498 * Gateway methods:: Gateway methods.
500 * Default Method:: Selecting a default method.
501 Here we also try to help those who
502 don't have the foggiest which method
504 * Default User:: Selecting a default user.
505 * Default Host:: Selecting a default host.
506 * Multi-hops:: Connecting to a remote host using multiple hops.
507 * Customizing Methods:: Using Non-Standard Methods.
508 * Customizing Completion:: Selecting config files for user/host name completion.
509 * Password handling:: Reusing passwords for several connections.
510 * Connection caching:: Reusing connection related information.
511 * Remote Programs:: How @value{tramp} finds and uses programs on the remote machine.
512 * Remote shell setup:: Remote shell setup hints.
513 * Windows setup hints:: Issues with Cygwin ssh.
514 * Auto-save and Backup:: Auto-save and Backup.
518 @node Connection types
519 @section Types of connections made to remote machines
520 @cindex connection types, overview
522 There are two basic types of transfer methods, each with its own
523 advantages and limitations. Both types of connection make use of a
524 remote shell access program such as @command{rsh}, @command{ssh} or
525 @command{telnet} to connect to the remote machine.
527 This connection is used to perform many of the operations that @value{tramp}
528 requires to make the remote file system transparently accessible from
529 the local machine. It is only when visiting files that the methods
532 @cindex inline methods
533 @cindex external methods
534 @cindex methods, inline
535 @cindex methods, external
536 Loading or saving a remote file requires that the content of the file
537 be transferred between the two machines. The content of the file can
538 be transferred using one of two methods: the @dfn{inline method} over
539 the same connection used to log in to the remote machine, or the
540 @dfn{external method} through another connection using a remote copy
541 program such as @command{rcp}, @command{scp} or @command{rsync}.
543 The performance of the external methods is generally better than that
544 of the inline methods, at least for large files. This is caused by
545 the need to encode and decode the data when transferring inline.
547 The one exception to this rule are the @command{scp} based transfer
548 methods. While these methods do see better performance when actually
549 transferring files, the overhead of the cryptographic negotiation at
550 startup may drown out the improvement in file transfer times.
552 External methods should be configured such a way that they don't
553 require a password (with @command{ssh-agent}, or such alike). Modern
554 @command{scp} implementations offer options to reuse existing
555 @command{ssh} connections, see method @command{scpc}. If it isn't
556 possible, you should consider @ref{Password handling}, otherwise you
557 will be prompted for a password every copy action.
561 @section Inline methods
562 @cindex inline methods
563 @cindex methods, inline
565 The inline methods in @value{tramp} are quite powerful and can work in
566 situations where you cannot use an external transfer program to connect.
567 Inline methods are the only methods that work when connecting to the
568 remote machine via telnet. (There are also strange inline methods which
569 allow you to transfer files between @emph{user identities} rather than
572 These methods depend on the existence of a suitable encoding and
573 decoding command on remote machine. Locally, @value{tramp} may be able to
574 use features of @value{emacsname} to decode and encode the files or
575 it may require access to external commands to perform that task.
579 @cindex base-64 encoding
580 @value{tramp} checks the availability and usability of commands like
581 @command{mimencode} (part of the @command{metamail} package) or
582 @command{uuencode} on the remote host. The first reliable command
583 will be used. The search path can be customized, see @ref{Remote
586 If both commands aren't available on the remote host, @value{tramp}
587 transfers a small piece of Perl code to the remote host, and tries to
588 apply it for encoding and decoding.
590 The variable @var{tramp-inline-compress-start-size} controls, whether
591 a file shall be compressed before encoding. This could increase
592 transfer speed for large text files.
600 Connect to the remote host with @command{rsh}. Due to the unsecure
601 connection it is recommended for very local host topology only.
603 On operating systems which provide the command @command{remsh} instead
604 of @command{rsh}, you can use the method @option{remsh}. This is true
605 for HP-UX or Cray UNICOS, for example.
612 Connect to the remote host with @command{ssh}. This is identical to
613 the previous option except that the @command{ssh} package is used,
614 making the connection more secure.
616 There are also two variants, @option{ssh1} and @option{ssh2}, that
617 call @samp{ssh -1} and @samp{ssh -2}, respectively. This way, you can
618 explicitly select whether you want to use the SSH protocol version 1
619 or 2 to connect to the remote host. (You can also specify in
620 @file{~/.ssh/config}, the SSH configuration file, which protocol
621 should be used, and use the regular @option{ssh} method.)
623 All the methods based on @command{ssh} have an additional feature: you
624 can specify a host name which looks like @file{host#42} (the real host
625 name, then a hash sign, then a port number). This means to connect to
626 the given host but to also pass @code{-p 42} as arguments to the
627 @command{ssh} command.
630 @item @option{telnet}
631 @cindex method telnet
632 @cindex telnet method
634 Connect to the remote host with @command{telnet}. This is as unsecure
635 as the @option{rsh} method.
642 This method does not connect to a remote host at all, rather it uses
643 the @command{su} program to allow you to edit files as another user.
644 That means, the specified host name in the file name must be either
645 @samp{localhost} or the host name as returned by the function
646 @command{(system-name)}. For an exception of this rule see
654 This is similar to the @option{su} method, but it uses @command{sudo}
655 rather than @command{su} to become a different user.
657 Note that @command{sudo} must be configured to allow you to start a
658 shell as the user. It would be nice if it was sufficient if
659 @command{ls} and @command{mimencode} were allowed, but that is not
660 easy to implement, so I haven't got around to it, yet.
667 As you would expect, this is similar to @option{ssh}, only a little
668 different. Whereas @option{ssh} opens a normal interactive shell on
669 the remote host, this option uses @samp{ssh -t -t @var{host} -l
670 @var{user} /bin/sh} to open a connection. This is useful for users
671 where the normal login shell is set up to ask them a number of
672 questions when logging in. This procedure avoids these questions, and
673 just gives @value{tramp} a more-or-less `standard' login shell to work
676 Note that this procedure does not eliminate questions asked by
677 @command{ssh} itself. For example, @command{ssh} might ask ``Are you
678 sure you want to continue connecting?'' if the host key of the remote
679 host is not known. @value{tramp} does not know how to deal with such a
680 question (yet), therefore you will need to make sure that you can log
681 in without such questions.
683 This is also useful for Windows users where @command{ssh}, when
684 invoked from an @value{emacsname} buffer, tells them that it is not
685 allocating a pseudo tty. When this happens, the login shell is wont
686 to not print any shell prompt, which confuses @value{tramp} mightily.
688 This supports the @samp{-p} argument.
691 @item @option{krlogin}
692 @cindex method krlogin
693 @cindex krlogin method
694 @cindex Kerberos (with krlogin method)
696 This method is also similar to @option{ssh}. It only uses the
697 @command{krlogin -x} command to log in to the remote host.
703 @cindex Kerberos (with ksu method)
705 This is another method from the Kerberos suite. It behaves like @option{su}.
712 This method is mostly interesting for Windows users using the PuTTY
713 implementation of SSH. It uses @samp{plink -ssh} to log in to the
716 This supports the @samp{-P} argument.
718 Additionally, the methods @option{plink1} and @option{plink2} are
719 provided, which call @samp{plink -1 -ssh} or @samp{plink -2 -ssh} in
720 order to use SSH protocol version 1 or 2 explicitly.
722 CCC: Do we have to connect to the remote host once from the command
723 line to accept the SSH key? Maybe this can be made automatic?
725 CCC: Say something about the first shell command failing. This might
726 be due to a wrong setting of @code{tramp-rsh-end-of-line}.
729 @item @option{plinkx}
730 @cindex method plinkx
731 @cindex plinkx method
733 Another method using PuTTY on Windows. Instead of host names, it
734 expects PuTTY session names, calling @samp{plink -load @var{session}
735 -t"}. User names are relevant only in case the corresponding session
736 hasn't defined a user name. Different port numbers must be defined in
742 @node External methods
743 @section External methods
744 @cindex methods, external
745 @cindex external methods
747 The external methods operate through multiple channels, using the
748 remote shell connection for many actions while delegating file
749 transfers to an external transfer utility.
751 This saves the overhead of encoding and decoding that multiplexing the
752 transfer through the one connection has with the inline methods.
754 Since external methods need their own overhead opening a new channel,
755 all files which are smaller than @var{tramp-copy-size-limit} are still
756 transferred with the corresponding inline method. It should provide a
757 fair trade-off between both approaches.
760 @item @option{rcp} --- @command{rsh} and @command{rcp}
763 @cindex rcp (with rcp method)
764 @cindex rsh (with rcp method)
766 This method uses the @command{rsh} and @command{rcp} commands to connect
767 to the remote machine and transfer files. This is probably the fastest
768 connection method available.
770 The alternative method @option{remcp} uses the @command{remsh} and
771 @command{rcp} commands. It should be applied on machines where
772 @command{remsh} is used instead of @command{rsh}.
775 @item @option{scp} --- @command{ssh} and @command{scp}
778 @cindex scp (with scp method)
779 @cindex ssh (with scp method)
781 Using @command{ssh} to connect to the remote host and @command{scp} to
782 transfer files between the machines is the best method for securely
783 connecting to a remote machine and accessing files.
785 The performance of this option is also quite good. It may be slower than
786 the inline methods when you often open and close small files however.
787 The cost of the cryptographic handshake at the start of an @command{scp}
788 session can begin to absorb the advantage that the lack of encoding and
791 There are also two variants, @option{scp1} and @option{scp2}, that
792 call @samp{ssh -1} and @samp{ssh -2}, respectively. This way, you can
793 explicitly select whether you want to use the SSH protocol version 1
794 or 2 to connect to the remote host. (You can also specify in
795 @file{~/.ssh/config}, the SSH configuration file, which protocol
796 should be used, and use the regular @option{scp} method.)
798 All the @command{ssh} based methods support the @samp{-p} feature
799 where you can specify a port number to connect to in the host name.
800 For example, the host name @file{host#42} tells @value{tramp} to
801 specify @samp{-p 42} in the argument list for @command{ssh}, and to
802 specify @samp{-P 42} in the argument list for @command{scp}.
805 @item @option{sftp} --- @command{ssh} and @command{sftp}
808 @cindex sftp (with sftp method)
809 @cindex ssh (with sftp method)
811 That is mostly the same method as @option{scp}, but using
812 @command{sftp} as transfer command. So the same remarks are valid.
814 This command does not work like @value{ftppackagename}, where
815 @command{ftp} is called interactively, and all commands are send from
816 within this session. Instead of, @command{ssh} is used for login.
818 This method supports the @samp{-p} argument.
821 @item @option{rsync} --- @command{ssh} and @command{rsync}
824 @cindex rsync (with rsync method)
825 @cindex ssh (with rsync method)
827 Using the @command{ssh} command to connect securely to the remote
828 machine and the @command{rsync} command to transfer files is almost
829 identical to the @option{scp} method.
831 While @command{rsync} performs much better than @command{scp} when
832 transferring files that exist on both hosts, this advantage is lost if
833 the file exists only on one side of the connection. A file can exists
834 on both the remote and local host, when you copy a file from/to a
835 remote host. When you just open a file from the remote host (or write
836 a file there), a temporary file on the local side is kept as long as
837 the corresponding buffer, visiting this file, is alive.
839 This method supports the @samp{-p} argument.
842 @item @option{scpx} --- @command{ssh} and @command{scp}
845 @cindex scp (with scpx method)
846 @cindex ssh (with scpx method)
848 As you would expect, this is similar to @option{scp}, only a little
849 different. Whereas @option{scp} opens a normal interactive shell on
850 the remote host, this option uses @samp{ssh -t -t @var{host} -l
851 @var{user} /bin/sh} to open a connection. This is useful for users
852 where the normal login shell is set up to ask them a number of
853 questions when logging in. This procedure avoids these questions, and
854 just gives @value{tramp} a more-or-less `standard' login shell to work
857 This is also useful for Windows users where @command{ssh}, when
858 invoked from an @value{emacsname} buffer, tells them that it is not
859 allocating a pseudo tty. When this happens, the login shell is wont
860 to not print any shell prompt, which confuses @value{tramp} mightily.
862 This method supports the @samp{-p} argument.
865 @item @option{scpc} --- @command{ssh} and @command{scp}
868 @cindex scp (with scpc method)
869 @cindex ssh (with scpc method)
871 Newer versions of @option{ssh} (for example OpenSSH 4) offer an option
872 @option{ControlMaster}. This allows @option{scp} to reuse an existing
873 @option{ssh} channel, which increases performance.
875 Before you use this method, you should check whether your @option{ssh}
876 implementation supports this option. Try from the command line
879 ssh localhost -o ControlMaster=yes /bin/true
882 If that command succeeds silently, then you can use @option{scpc}; but
886 command-line: line 0: Bad configuration option: ControlMaster
889 then you cannot use it. Note, that the option
890 @option{ControlPersist}, if it is supported by your @option{ssh}
891 version, must be set to @option{no}.
893 This method supports the @samp{-p} argument.
896 @item @option{rsyncc} --- @command{ssh} and @command{rsync}
897 @cindex method rsyncc
898 @cindex rsyncc method
899 @cindex rsync (with rsyncc method)
900 @cindex ssh (with rsyncc method)
902 Like the @option{scpc} method, @option{rsyncc} improves the underlying
903 @command{ssh} connection by the option @option{ControlMaster}. This
904 allows @command{rsync} to reuse an existing @command{ssh} channel,
905 which increases performance.
907 This method supports the @samp{-p} argument.
910 @item @option{pscp} --- @command{plink} and @command{pscp}
913 @cindex pscp (with pscp method)
914 @cindex plink (with pscp method)
915 @cindex PuTTY (with pscp method)
917 This method is similar to @option{scp}, but it uses the
918 @command{plink} command to connect to the remote host, and it uses
919 @command{pscp} for transferring the files. These programs are part
920 of PuTTY, an SSH implementation for Windows.
922 This method supports the @samp{-P} argument.
925 @item @option{psftp} --- @command{plink} and @command{psftp}
928 @cindex psftp (with psftp method)
929 @cindex plink (with psftp method)
930 @cindex PuTTY (with psftp method)
932 As you would expect, this method is similar to @option{sftp}, but it
933 uses the @command{plink} command to connect to the remote host, and it
934 uses @command{psftp} for transferring the files. These programs are
935 part of PuTTY, an SSH implementation for Windows.
937 This method supports the @samp{-P} argument.
940 @item @option{fcp} --- @command{fsh} and @command{fcp}
943 @cindex fsh (with fcp method)
944 @cindex fcp (with fcp method)
946 This method is similar to @option{scp}, but it uses the @command{fsh}
947 command to connect to the remote host, and it uses @command{fcp} for
948 transferring the files. @command{fsh/fcp} are a front-end for
949 @command{ssh} which allow for reusing the same @command{ssh} session
950 for submitting several commands. This avoids the startup overhead of
951 @command{scp} (which has to establish a secure connection whenever it
952 is called). Note, however, that you can also use one of the inline
953 methods to achieve a similar effect.
955 This method uses the command @samp{fsh @var{host} -l @var{user}
956 /bin/sh -i} to establish the connection, it does not work to just say
957 @command{fsh @var{host} -l @var{user}}.
962 There is no inline method using @command{fsh} as the multiplexing
963 provided by the program is not very useful in our context. @value{tramp}
964 opens just one connection to the remote host and then keeps it open,
972 This is not a native @value{tramp} method. Instead, it forwards all
973 requests to @value{ftppackagename}.
975 This works only for unified filenames, see @ref{Issues}.
979 @item @option{smb} --- @command{smbclient}
983 This is another not natural @value{tramp} method. It uses the
984 @command{smbclient} command on different Unices in order to connect to
985 an SMB server. An SMB server might be a Samba (or CIFS) server on
986 another UNIX host or, more interesting, a host running MS Windows. So
987 far, it is tested against MS Windows NT, MS Windows 2000, MS Windows
988 XP, MS Windows Vista, and MS Windows 7.
990 The first directory in the localname must be a share name on the remote
991 host. Remember that the @code{$} character, in which default shares
992 usually end, must be written @code{$$} due to environment variable
993 substitution in file names. If no share name is given (i.e. remote
994 directory @code{/}), all available shares are listed.
996 Since authorization is done on share level, you will always be
997 prompted for a password if you access another share on the same host.
998 This can be suppressed by @ref{Password handling}.
1000 For authorization, MS Windows uses both a user name and a domain name.
1001 Because of this, the @value{tramp} syntax has been extended: you can
1002 specify a user name which looks like @code{user%domain} (the real user
1003 name, then a percent sign, then the domain name). So, to connect to
1004 the machine @code{melancholia} as user @code{daniel} of the domain
1005 @code{BIZARRE}, and edit @file{.emacs} in the home directory (share
1006 @code{daniel$}) I would specify the filename @file{@trampfn{smb,
1007 daniel%BIZARRE, melancholia, /daniel$$/.emacs}}.
1009 Depending on the Windows domain configuration, a Windows user might be
1010 considered as domain user per default. In order to connect as local
1011 user, the WINS name of that machine must be given as domain name.
1012 Usually, it is the machine name in capital letters. In the example
1013 above, the local user @code{daniel} would be specified as
1014 @file{@trampfn{smb, daniel%MELANCHOLIA, melancholia, /daniel$$/.emacs}}.
1016 The domain name as well as the user name are optional. If no user
1017 name is specified at all, the anonymous user (without password
1018 prompting) is assumed. This is different from all other @value{tramp}
1019 methods, where in such a case the local user name is taken.
1021 The @option{smb} method supports the @samp{-p} argument.
1023 @strong{Please note:} If @value{emacsname} runs locally under MS
1024 Windows, this method isn't available. Instead, you can use UNC
1025 file names like @file{//melancholia/daniel$$/.emacs}. The only
1026 disadvantage is that there's no possibility to specify another user
1032 @node GVFS based methods
1033 @section GVFS based external methods
1034 @cindex methods, gvfs
1035 @cindex gvfs based methods
1038 The connection methods described in this section are based on GVFS
1039 @uref{http://en.wikipedia.org/wiki/GVFS}. Via GVFS, the remote
1040 filesystem is mounted locally through FUSE. @value{tramp} uses
1041 this local mounted directory internally.
1043 The communication with GVFS is implemented via D-Bus messages.
1044 Therefore, your @value{emacsname} must have D-Bus integration,
1045 @pxref{Top, , D-Bus, dbus}.
1054 This method provides access to WebDAV files and directories. There
1055 exists also the external method @option{davs}, which uses SSL
1056 encryption for the access.
1058 Both methods support the port number specification as discussed above.
1065 OBEX is an FTP-like access protocol for simple devices, like cell
1066 phones. For the time being, @value{tramp} only supports OBEX over Bluetooth.
1069 @item @option{synce}
1070 @cindex method synce
1071 @cindex synce method
1073 The @option{synce} method allows communication with Windows Mobile
1074 devices. Beside GVFS for mounting remote files and directories via
1075 FUSE, it also needs the SYNCE-GVFS plugin.
1078 @defopt tramp-gvfs-methods
1079 This customer option, a list, defines the external methods which
1080 shall be used with GVFS. Per default, these are @option{dav},
1081 @option{davs}, @option{obex} and @option{synce}. Other possible
1082 values are @option{ftp}, @option{sftp} and @option{smb}.
1088 @node Gateway methods
1089 @section Gateway methods
1090 @cindex methods, gateway
1091 @cindex gateway methods
1093 Gateway methods are not methods to access a remote host directly.
1094 These methods are intended to pass firewalls or proxy servers.
1095 Therefore, they can be used for proxy host declarations
1096 (@pxref{Multi-hops}) only.
1098 A gateway method must always come along with a method which supports
1099 port setting. This is because @value{tramp} targets the accompanied
1100 method to @file{localhost#random_port}, from where the firewall or
1101 proxy server is accessed.
1103 Gateway methods support user name and password declarations. These
1104 are used to authenticate towards the corresponding firewall or proxy
1105 server. They can be passed only if your friendly administrator has
1106 granted your access.
1109 @item @option{tunnel}
1110 @cindex method tunnel
1111 @cindex tunnel method
1113 This method implements an HTTP tunnel via the @command{CONNECT}
1114 command (see RFC 2616, 2817). Any HTTP 1.1 compliant (proxy) server
1115 shall support this command.
1117 As authentication method, only @option{Basic Authentication} (see RFC
1118 2617) is implemented so far. If no port number is given in the
1119 declaration, port @option{8080} is used for the proxy server.
1122 @item @option{socks}
1123 @cindex method socks
1124 @cindex socks method
1126 The @command{socks} method provides access to SOCKSv5 servers (see
1127 RFC 1928). @option{Username/Password Authentication} according to RFC
1130 The default port number of the socks server is @option{1080}, if not
1131 specified otherwise.
1137 @node Default Method
1138 @section Selecting a default method
1139 @cindex default method
1141 @vindex tramp-default-method
1142 When you select an appropriate transfer method for your typical usage
1143 you should set the variable @code{tramp-default-method} to reflect that
1144 choice. This variable controls which method will be used when a method
1145 is not specified in the @value{tramp} file name. For example:
1148 (setq tramp-default-method "ssh")
1151 @vindex tramp-default-method-alist
1152 You can also specify different methods for certain user/host
1153 combinations, via the variable @code{tramp-default-method-alist}. For
1154 example, the following two lines specify to use the @option{ssh}
1155 method for all user names matching @samp{john} and the @option{rsync}
1156 method for all host names matching @samp{lily}. The third line
1157 specifies to use the @option{su} method for the user @samp{root} on
1158 the machine @samp{localhost}.
1161 (add-to-list 'tramp-default-method-alist '("" "john" "ssh"))
1162 (add-to-list 'tramp-default-method-alist '("lily" "" "rsync"))
1163 (add-to-list 'tramp-default-method-alist
1164 '("\\`localhost\\'" "\\`root\\'" "su"))
1168 See the documentation for the variable
1169 @code{tramp-default-method-alist} for more details.
1171 External methods are normally preferable to inline methods, giving
1174 @xref{Inline methods}.
1175 @xref{External methods}.
1177 Another consideration with the selection of transfer methods is the
1178 environment you will use them in and, especially when used over the
1179 Internet, the security implications of your preferred method.
1181 The @option{rsh} and @option{telnet} methods send your password as
1182 plain text as you log in to the remote machine, as well as
1183 transferring the files in such a way that the content can easily be
1184 read from other machines.
1186 If you need to connect to remote systems that are accessible from the
1187 Internet, you should give serious thought to using @option{ssh} based
1188 methods to connect. These provide a much higher level of security,
1189 making it a non-trivial exercise for someone to obtain your password
1190 or read the content of the files you are editing.
1193 @subsection Which method is the right one for me?
1194 @cindex choosing the right method
1196 Given all of the above, you are probably thinking that this is all fine
1197 and good, but it's not helping you to choose a method! Right you are.
1198 As a developer, we don't want to boss our users around but give them
1199 maximum freedom instead. However, the reality is that some users would
1200 like to have some guidance, so here I'll try to give you this guidance
1201 without bossing you around. You tell me whether it works @dots{}
1203 My suggestion is to use an inline method. For large files, external
1204 methods might be more efficient, but I guess that most people will
1205 want to edit mostly small files. And if you access large text files,
1206 compression (driven by @var{tramp-inline-compress-start-size}) shall
1207 still result in good performance.
1209 I guess that these days, most people can access a remote machine by
1210 using @command{ssh}. So I suggest that you use the @option{ssh}
1211 method. So, type @kbd{C-x C-f @trampfn{ssh, root, otherhost,
1212 /etc/motd} @key{RET}} to edit the @file{/etc/motd} file on the other
1215 If you can't use @option{ssh} to log in to the remote host, then
1216 select a method that uses a program that works. For instance, Windows
1217 users might like the @option{plink} method which uses the PuTTY
1218 implementation of @command{ssh}. Or you use Kerberos and thus like
1221 For the special case of editing files on the local host as another
1222 user, see the @option{su} or @option{sudo} methods. They offer
1223 shortened syntax for the @samp{root} account, like
1224 @file{@trampfn{su, , , /etc/motd}}.
1226 People who edit large files may want to consider @option{scpc} instead
1227 of @option{ssh}, or @option{pscp} instead of @option{plink}. These
1228 external methods are faster than inline methods for large files.
1229 Note, however, that external methods suffer from some limitations.
1230 Please try first whether you really get a noticeable speed advantage
1231 from using an external method! Maybe even for large files, inline
1232 methods are fast enough.
1236 @section Selecting a default user
1237 @cindex default user
1239 The user part of a @value{tramp} file name can be omitted. Usually,
1240 it is replaced by the user name you are logged in. Often, this is not
1241 what you want. A typical use of @value{tramp} might be to edit some
1242 files with root permissions on the local host. This case, you should
1243 set the variable @code{tramp-default-user} to reflect that choice.
1247 (setq tramp-default-user "root")
1250 @code{tramp-default-user} is regarded as obsolete, and will be removed
1253 @vindex tramp-default-user-alist
1254 You can also specify different users for certain method/host
1255 combinations, via the variable @code{tramp-default-user-alist}. For
1256 example, if you always have to use the user @samp{john} in the domain
1257 @samp{somewhere.else}, you can specify the following:
1260 (add-to-list 'tramp-default-user-alist
1261 '("ssh" ".*\\.somewhere\\.else\\'" "john"))
1265 See the documentation for the variable
1266 @code{tramp-default-user-alist} for more details.
1268 One trap to fall in must be known. If @value{tramp} finds a default
1269 user, this user will be passed always to the connection command as
1270 parameter (for example @command{ssh here.somewhere.else -l john}. If
1271 you have specified another user for your command in its configuration
1272 files, @value{tramp} cannot know it, and the remote access will fail.
1273 If you have specified in the given example in @file{~/.ssh/config} the
1277 Host here.somewhere.else
1282 than you must discard selecting a default user by @value{tramp}. This
1283 will be done by setting it to @code{nil} (or @samp{lily}, likewise):
1286 (add-to-list 'tramp-default-user-alist
1287 '("ssh" "\\`here\\.somewhere\\.else\\'" nil))
1290 The last entry in @code{tramp-default-user-alist} could be your
1291 default user you'll apply predominantly. You shall @emph{append} it
1292 to that list at the end:
1295 (add-to-list 'tramp-default-user-alist '(nil nil "jonas") t)
1300 @section Selecting a default host
1301 @cindex default host
1303 @vindex tramp-default-host
1304 Finally, it is even possible to omit the host name part of a
1305 @value{tramp} file name. This case, the value of the variable
1306 @code{tramp-default-host} is used. Per default, it is initialized
1307 with the host name your local @value{emacsname} is running.
1309 If you, for example, use @value{tramp} mainly to contact the host
1310 @samp{target} as user @samp{john}, you can specify:
1313 (setq tramp-default-user "john"
1314 tramp-default-host "target")
1317 Then the simple file name @samp{@trampfn{ssh, , ,}} will connect you
1318 to John's home directory on target.
1320 Note, however, that the most simplification @samp{/::} won't work,
1321 because @samp{/:} is the prefix for quoted file names.
1326 @section Connecting to a remote host using multiple hops
1330 Sometimes, the methods described before are not sufficient.
1331 Sometimes, it is not possible to connect to a remote host using a
1332 simple command. For example, if you are in a secured network, you
1333 might have to log in to a bastion host first before you can connect to
1334 the outside world. Of course, the target host may also require a
1337 @vindex tramp-default-proxies-alist
1338 @defopt tramp-default-proxies-alist
1339 In order to specify multiple hops, it is possible to define a proxy
1340 host to pass through, via the variable
1341 @code{tramp-default-proxies-alist}. This variable keeps a list of
1342 triples (@var{host} @var{user} @var{proxy}).
1344 The first matching item specifies the proxy host to be passed for a
1345 file name located on a remote target matching @var{user}@@@var{host}.
1346 @var{host} and @var{user} are regular expressions or @code{nil}, which
1347 is interpreted as a regular expression which always matches.
1349 @var{proxy} must be a Tramp filename which localname part is ignored.
1350 Method and user name on @var{proxy} are optional, which is interpreted
1351 with the default values.
1353 The method must be an inline or gateway method (@pxref{Inline
1354 methods}, @pxref{Gateway methods}).
1357 The method must be an inline method (@pxref{Inline methods}).
1359 If @var{proxy} is @code{nil}, no additional hop is required reaching
1360 @var{user}@@@var{host}.
1362 If you, for example, must pass the host @samp{bastion.your.domain} as
1363 user @samp{bird} for any remote host which is not located in your local
1367 (add-to-list 'tramp-default-proxies-alist
1368 '("\\." nil "@trampfn{ssh, bird, bastion.your.domain,}"))
1369 (add-to-list 'tramp-default-proxies-alist
1370 '("\\.your\\.domain\\'" nil nil))
1373 Please note the order of the code. @code{add-to-list} adds elements at the
1374 beginning of a list. Therefore, most relevant rules must be added last.
1376 Proxy hosts can be cascaded. If there is another host called
1377 @samp{jump.your.domain}, which is the only one in your local domain who
1378 is allowed connecting @samp{bastion.your.domain}, you can add another
1382 (add-to-list 'tramp-default-proxies-alist
1383 '("\\`bastion\\.your\\.domain\\'"
1385 "@trampfn{ssh, , jump.your.domain,}"))
1388 @var{proxy} can contain the patterns @code{%h} or @code{%u}. These
1389 patterns are replaced by the strings matching @var{host} or
1390 @var{user}, respectively.
1392 If you, for example, wants to work as @samp{root} on hosts in the
1393 domain @samp{your.domain}, but login as @samp{root} is disabled for
1394 non-local access, you might add the following rule:
1397 (add-to-list 'tramp-default-proxies-alist
1398 '("\\.your\\.domain\\'" "\\`root\\'" "@trampfn{ssh, , %h,}"))
1401 Opening @file{@trampfn{sudo, , randomhost.your.domain,}} would connect
1402 first @samp{randomhost.your.domain} via @code{ssh} under your account
1403 name, and perform @code{sudo -u root} on that host afterwards. It is
1404 important to know that the given method is applied on the host which
1405 has been reached so far. @code{sudo -u root}, applied on your local
1406 host, wouldn't be useful here.
1408 @var{host}, @var{user} and @var{proxy} can also be Lisp forms. These
1409 forms are evaluated, and must return a string, or @code{nil}. The
1410 previous example could be generalized then: For all hosts except my
1411 local one connect via @command{ssh} first, and apply @command{sudo -u
1415 (add-to-list 'tramp-default-proxies-alist
1416 '(nil "\\`root\\'" "@trampfn{ssh, , %h,}"))
1417 (add-to-list 'tramp-default-proxies-alist
1418 '((regexp-quote (system-name)) nil nil))
1421 This is the recommended configuration to work as @samp{root} on remote
1425 Finally, @code{tramp-default-proxies-alist} can be used to pass
1426 firewalls or proxy servers. Imagine your local network has a host
1427 @samp{proxy.your.domain} which is used on port 3128 as HTTP proxy to
1428 the outer world. Your friendly administrator has granted you access
1429 under your user name to @samp{host.other.domain} on that proxy
1430 server.@footnote{HTTP tunnels are intended for secure SSL/TLS
1431 communication. Therefore, many proxy server restrict the tunnels to
1432 related target ports. You might need to run your ssh server on your
1433 target host @samp{host.other.domain} on such a port, like 443 (https).
1434 See @uref{http://savannah.gnu.org/maintenance/CvsFromBehindFirewall}
1435 for discussion of ethical issues.} You would need to add the
1439 (add-to-list 'tramp-default-proxies-alist
1440 '("\\`host\\.other\\.domain\\'" nil
1441 "@trampfn{tunnel, , proxy.your.domain#3128,}"))
1444 Gateway methods can be declared as first hop only in a multiple hop
1449 Hops to be passed tend to be restricted firewalls and alike.
1450 Sometimes they offer limited features only, like running @command{rbash}
1451 (restricted bash). This must be told to @value{tramp}.
1453 @vindex tramp-restricted-shell-hosts-alist
1454 @defopt tramp-restricted-shell-hosts-alist
1455 This variable keeps a list of regular expressions, which denote hosts
1456 running a registered shell like "rbash". Those hosts can be used as
1459 If the bastion host from the example above runs a restricted shell,
1463 (add-to-list 'tramp-restricted-shell-hosts-alist
1464 "\\`bastion\\.your\\.domain\\'")
1469 @node Customizing Methods
1470 @section Using Non-Standard Methods
1471 @cindex customizing methods
1472 @cindex using non-standard methods
1473 @cindex create your own methods
1475 There is a variable @code{tramp-methods} which you can change if the
1476 predefined methods don't seem right.
1478 For the time being, I'll refer you to the Lisp documentation of that
1479 variable, accessible with @kbd{C-h v tramp-methods @key{RET}}.
1482 @node Customizing Completion
1483 @section Selecting config files for user/host name completion
1484 @cindex customizing completion
1485 @cindex selecting config files
1486 @vindex tramp-completion-function-alist
1488 The variable @code{tramp-completion-function-alist} is intended to
1489 customize which files are taken into account for user and host name
1490 completion (@pxref{Filename completion}). For every method, it keeps
1491 a set of configuration files, accompanied by a Lisp function able to
1492 parse that file. Entries in @code{tramp-completion-function-alist}
1493 have the form (@var{method} @var{pair1} @var{pair2} ...).
1495 Each @var{pair} is composed of (@var{function} @var{file}).
1496 @var{function} is responsible to extract user names and host names
1497 from @var{file} for completion. There are two functions which access
1500 @defun tramp-get-completion-function method
1501 This function returns the list of completion functions for @var{method}.
1505 (tramp-get-completion-function "rsh")
1507 @result{} ((tramp-parse-rhosts "/etc/hosts.equiv")
1508 (tramp-parse-rhosts "~/.rhosts"))
1512 @defun tramp-set-completion-function method function-list
1513 This function sets @var{function-list} as list of completion functions
1518 (tramp-set-completion-function "ssh"
1519 '((tramp-parse-sconfig "/etc/ssh_config")
1520 (tramp-parse-sconfig "~/.ssh/config")))
1522 @result{} ((tramp-parse-sconfig "/etc/ssh_config")
1523 (tramp-parse-sconfig "~/.ssh/config"))
1527 The following predefined functions parsing configuration files exist:
1530 @item @code{tramp-parse-rhosts}
1531 @findex tramp-parse-rhosts
1533 This function parses files which are syntactical equivalent to
1534 @file{~/.rhosts}. It returns both host names and user names, if
1537 @item @code{tramp-parse-shosts}
1538 @findex tramp-parse-shosts
1540 This function parses files which are syntactical equivalent to
1541 @file{~/.ssh/known_hosts}. Since there are no user names specified
1542 in such files, it can return host names only.
1544 @item @code{tramp-parse-sconfig}
1545 @findex tramp-parse-shosts
1547 This function returns the host nicknames defined by @code{Host} entries
1548 in @file{~/.ssh/config} style files.
1550 @item @code{tramp-parse-shostkeys}
1551 @findex tramp-parse-shostkeys
1553 SSH2 parsing of directories @file{/etc/ssh2/hostkeys/*} and
1554 @file{~/ssh2/hostkeys/*}. Hosts are coded in file names
1555 @file{hostkey_@var{portnumber}_@var{host-name}.pub}. User names
1556 are always @code{nil}.
1558 @item @code{tramp-parse-sknownhosts}
1559 @findex tramp-parse-shostkeys
1561 Another SSH2 style parsing of directories like
1562 @file{/etc/ssh2/knownhosts/*} and @file{~/ssh2/knownhosts/*}. This
1563 case, hosts names are coded in file names
1564 @file{@var{host-name}.@var{algorithm}.pub}. User names are always @code{nil}.
1566 @item @code{tramp-parse-hosts}
1567 @findex tramp-parse-hosts
1569 A function dedicated to @file{/etc/hosts} style files. It returns
1572 @item @code{tramp-parse-passwd}
1573 @findex tramp-parse-passwd
1575 A function which parses @file{/etc/passwd} like files. Obviously, it
1576 can return user names only.
1578 @item @code{tramp-parse-netrc}
1579 @findex tramp-parse-netrc
1581 Finally, a function which parses @file{~/.netrc} like files. This
1582 includes also @file{~/.authinfo}-style files.
1585 If you want to keep your own data in a file, with your own structure,
1586 you might provide such a function as well. This function must meet
1587 the following conventions:
1589 @defun my-tramp-parse file
1590 @var{file} must be either a file name on your host, or @code{nil}.
1591 The function must return a list of (@var{user} @var{host}), which are
1592 taken as candidates for user and host name completion.
1596 (my-tramp-parse "~/.my-tramp-hosts")
1598 @result{} ((nil "toto") ("daniel" "melancholia"))
1603 @node Password handling
1604 @section Reusing passwords for several connections
1607 Sometimes it is necessary to connect to the same remote host several
1608 times. Reentering passwords again and again would be annoying, when
1609 the chosen method does not support access without password prompt
1610 through own configuration.
1612 The best recommendation is to use the method's own mechanism for
1613 password handling. Consider @command{ssh-agent} for @option{ssh}-like
1614 methods, or @command{pageant} for @option{plink}-like methods.
1616 However, if you cannot apply such native password handling,
1617 @value{tramp} offers alternatives.
1620 @anchor{Using an authentication file}
1621 @subsection Using an authentication file
1623 @vindex auth-sources
1624 The package @file{auth-source.el}, originally developed in No Gnus,
1625 offers the possibility to read passwords from a file, like FTP does it
1626 from @file{~/.netrc}. The default authentication file is
1627 @file{~/.authinfo.gpg}, this can be changed via the variable
1628 @code{auth-sources}.
1631 A typical entry in the authentication file would be
1634 machine melancholia port scp login daniel password geheim
1637 The port can be any @value{tramp} method (@pxref{Inline methods},
1638 @pxref{External methods}), to match only this method. When you omit
1639 the port, you match all @value{tramp} methods.
1641 In case of problems, setting @code{auth-source-debug} to @code{t}
1642 gives useful debug messages.
1645 @anchor{Caching passwords}
1646 @subsection Caching passwords
1648 If there is no authentication file, @value{tramp} caches the passwords
1649 entered by you. They will be reused next time if a connection needs
1650 them for the same user name and host name, independently of the
1653 @vindex password-cache-expiry
1654 Passwords are not saved permanently, that means the password caching
1655 is limited to the lifetime of your @value{emacsname} session. You
1656 can influence the lifetime of password caching by customizing the
1657 variable @code{password-cache-expiry}. The value is the number of
1658 seconds how long passwords are cached. Setting it to @code{nil}
1659 disables the expiration.
1661 @vindex password-cache
1662 If you don't like this feature for security reasons, password caching
1663 can be disabled totally by customizing the variable
1664 @code{password-cache} (setting it to @code{nil}).
1666 Implementation Note: password caching is based on the package
1667 @file{password-cache.el}. For the time being, it is activated only
1668 when this package is seen in the @code{load-path} while loading
1670 @ifset installchapter
1671 If you don't use No Gnus, you can take @file{password.el} from the
1672 @value{tramp} @file{contrib} directory, see @ref{Installation
1677 @node Connection caching
1678 @section Reusing connection related information
1681 @vindex tramp-persistency-file-name
1682 In order to reduce initial connection time, @value{tramp} stores
1683 connection related information persistently. The variable
1684 @code{tramp-persistency-file-name} keeps the file name where these
1685 information are written. Its default value is
1687 @file{~/.emacs.d/tramp}.
1690 @file{~/.xemacs/tramp}.
1692 It is recommended to choose a local file name.
1694 @value{tramp} reads this file during startup, and writes it when
1695 exiting @value{emacsname}. You can simply remove this file if
1696 @value{tramp} shall be urged to recompute these information next
1697 @value{emacsname} startup time.
1699 Using such persistent information can be disabled by setting
1700 @code{tramp-persistency-file-name} to @code{nil}.
1702 Once consequence of reusing connection related information is that
1703 @var{tramp} needs to distinguish hosts. If you, for example, run a
1704 local @code{sshd} on port 3001, which tunnels @command{ssh} to another
1705 host, you could access both @file{@trampfn{ssh, , localhost,}} and
1706 @file{@trampfn{ssh, , localhost#3001,}}. @var{tramp} would use the
1707 same host related information (like paths, Perl variants, etc) for
1708 both connections, although the information is valid only for one of
1711 In order to avoid trouble, you must use another host name for one of
1712 the connections, like introducing a @option{Host} section in
1713 @file{~/.ssh/config} (@pxref{Frequently Asked Questions}) or applying
1714 multiple hops (@pxref{Multi-hops}).
1716 When @value{tramp} detects a changed operating system version on a
1717 remote host (via the command @command{uname -sr}), it flushes all
1718 connection related information for this host, and opens the
1722 @node Remote Programs
1723 @section How @value{tramp} finds and uses programs on the remote machine
1725 @value{tramp} depends on a number of programs on the remote host in order to
1726 function, including @command{ls}, @command{test}, @command{find} and
1729 In addition to these required tools, there are various tools that may be
1730 required based on the connection method. See @ref{Inline methods} and
1731 @ref{External methods} for details on these.
1733 Certain other tools, such as @command{perl} (or @command{perl5}) and
1734 @command{grep} will be used if they can be found. When they are
1735 available, they are used to improve the performance and accuracy of
1738 @vindex tramp-remote-path
1739 @vindex tramp-default-remote-path
1740 @vindex tramp-own-remote-path
1741 @defopt tramp-remote-path
1742 When @value{tramp} connects to the remote machine, it searches for the
1743 programs that it can use. The variable @code{tramp-remote-path}
1744 controls the directories searched on the remote machine.
1746 By default, this is set to a reasonable set of defaults for most
1747 machines. The symbol @code{tramp-default-remote-path} is a place
1748 holder, it is replaced by the list of directories received via the
1749 command @command{getconf PATH} on your remote machine. For example,
1750 on Debian GNU/Linux this is @file{/bin:/usr/bin}, whereas on Solaris
1751 this is @file{/usr/xpg4/bin:/usr/ccs/bin:/usr/bin:/opt/SUNWspro/bin}.
1752 It is recommended to apply this symbol on top of
1753 @code{tramp-remote-path}.
1755 It is possible, however, that your local (or remote ;) system
1756 administrator has put the tools you want in some obscure local
1759 In this case, you can still use them with @value{tramp}. You simply
1760 need to add code to your @file{.emacs} to add the directory to the
1761 remote path. This will then be searched by @value{tramp} when you
1762 connect and the software found.
1764 To add a directory to the remote search path, you could use code such
1768 @i{;; We load @value{tramp} to define the variable.}
1770 @i{;; We have @command{perl} in "/usr/local/perl/bin"}
1771 (add-to-list 'tramp-remote-path "/usr/local/perl/bin")
1774 Another possibility is to reuse the path settings of your remote
1775 account when you log in. Usually, these settings are overwritten,
1776 because they might not be useful for @value{tramp}. The place holder
1777 @code{tramp-own-remote-path} preserves these settings. You can
1781 (add-to-list 'tramp-remote-path 'tramp-own-remote-path)
1785 @value{tramp} caches several information, like the Perl binary
1786 location. The changed remote search path wouldn't affect these
1787 settings. In order to force @value{tramp} to recompute these values,
1788 you must exit @value{emacsname}, remove your persistency file
1789 (@pxref{Connection caching}), and restart @value{emacsname}.
1792 @node Remote shell setup
1793 @section Remote shell setup hints
1794 @cindex remote shell setup
1795 @cindex @file{.profile} file
1796 @cindex @file{.login} file
1797 @cindex shell init files
1799 As explained in the @ref{Overview} section, @value{tramp} connects to the
1800 remote host and talks to the shell it finds there. Of course, when you
1801 log in, the shell executes its init files. Suppose your init file
1802 requires you to enter the birth date of your mother; clearly @value{tramp}
1803 does not know this and hence fails to log you in to that host.
1805 There are different possible strategies for pursuing this problem. One
1806 strategy is to enable @value{tramp} to deal with all possible situations.
1807 This is a losing battle, since it is not possible to deal with
1808 @emph{all} situations. The other strategy is to require you to set up
1809 the remote host such that it behaves like @value{tramp} expects. This might
1810 be inconvenient because you have to invest a lot of effort into shell
1811 setup before you can begin to use @value{tramp}.
1813 The package, therefore, pursues a combined approach. It tries to
1814 figure out some of the more common setups, and only requires you to
1815 avoid really exotic stuff. For example, it looks through a list of
1816 directories to find some programs on the remote host. And also, it
1817 knows that it is not obvious how to check whether a file exists, and
1818 therefore it tries different possibilities. (On some hosts and
1819 shells, the command @command{test -e} does the trick, on some hosts
1820 the shell builtin doesn't work but the program @command{/usr/bin/test
1821 -e} or @command{/bin/test -e} works. And on still other hosts,
1822 @command{ls -d} is the right way to do this.)
1824 Below you find a discussion of a few things that @value{tramp} does not deal
1825 with, and that you therefore have to set up correctly.
1828 @item @var{shell-prompt-pattern}
1829 @vindex shell-prompt-pattern
1831 After logging in to the remote host, @value{tramp} has to wait for the remote
1832 shell startup to finish before it can send commands to the remote
1833 shell. The strategy here is to wait for the shell prompt. In order to
1834 recognize the shell prompt, the variable @code{shell-prompt-pattern} has
1835 to be set correctly to recognize the shell prompt on the remote host.
1837 Note that @value{tramp} requires the match for @code{shell-prompt-pattern}
1838 to be at the end of the buffer. Many people have something like the
1839 following as the value for the variable: @code{"^[^>$][>$] *"}. Now
1840 suppose your shell prompt is @code{a <b> c $ }. In this case,
1841 @value{tramp} recognizes the @code{>} character as the end of the prompt,
1842 but it is not at the end of the buffer.
1844 @item @var{tramp-shell-prompt-pattern}
1845 @vindex tramp-shell-prompt-pattern
1847 This regular expression is used by @value{tramp} in the same way as
1848 @code{shell-prompt-pattern}, to match prompts from the remote shell.
1849 This second variable exists because the prompt from the remote shell
1850 might be different from the prompt from a local shell --- after all,
1851 the whole point of @value{tramp} is to log in to remote hosts as a
1852 different user. The default value of
1853 @code{tramp-shell-prompt-pattern} is the same as the default value of
1854 @code{shell-prompt-pattern}, which is reported to work well in many
1857 @item @var{tramp-password-prompt-regexp}
1858 @vindex tramp-password-prompt-regexp
1859 @vindex tramp-wrong-passwd-regexp
1861 During login, @value{tramp} might be forced to enter a password or a
1862 passphrase. The difference between both is that a password is
1863 requested from the shell on the remote host, while a passphrase is
1864 needed for accessing local authentication information, like your ssh
1867 @var{tramp-password-prompt-regexp} handles the detection of such
1868 requests for English environments. When you use another localization
1869 of your (local or remote) host, you might need to adapt this. Example:
1873 tramp-password-prompt-regexp
1877 '("passphrase" "Passphrase"
1879 "password" "Password"
1881 "passwort" "Passwort"
1883 "mot de passe" "Mot de passe") t)