\input texinfo @c -*- mode: texinfo; coding: utf-8 -*- @setfilename ../../info/tramp.info @c %**start of header @settitle TRAMP User Manual @include docstyle.texi @c %**end of header @c This is *so* much nicer :) @footnotestyle end @c In the Tramp repository, the version number is auto-frobbed from @c configure.ac, so you should edit that file and run @c "autoconf && ./configure" to change the version number. @include trampver.texi @c Macro for formatting a file name according to the respective @c syntax. Macro arguments should not have any leading or trailing @c whitespace. Not very elegant, but I don't know it better. @macro trampfn {method, userhost, localname} @value{prefix}@c \method\@value{postfixhop}@c \userhost\@value{postfix}\localname\ @end macro @c Similar, but without a method prefix. @macro trampf {userhost, localname} @value{prefix}@c \userhost\@value{postfix}\localname\ @end macro @copying Copyright @copyright{} 1999--2016 Free Software Foundation, Inc. @quotation Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, with the Front-Cover Texts being ``A GNU Manual'', and with the Back-Cover Texts as in (a) below. A copy of the license is included in the section entitled ``GNU Free Documentation License''. (a) The FSF's Back-Cover Text is: ``You have the freedom to copy and modify this GNU manual.'' @end quotation @end copying @c Entries for @command{install-info} to use @dircategory Emacs network features @direntry * TRAMP: (tramp). Transparent Remote Access, Multiple Protocol Emacs remote file access via ssh and scp. @end direntry @titlepage @title @value{tramp} version @value{trampver} User Manual @author by Daniel Pittman @author based on documentation by Kai Großjohann @end titlepage @contents @node Top, Overview, (dir), (dir) @top @value{tramp} version @value{trampver} User Manual This file documents @value{tramp} version @value{trampver}, a remote file editing package for Emacs. @value{tramp} stands for ``Transparent Remote (file) Access, Multiple Protocol''. This package provides remote file editing, similar to Ange FTP. The difference is that Ange FTP uses FTP to transfer files between the local and the remote host, whereas @value{tramp} uses a combination of @command{rsh} and @command{rcp} or other work-alike programs, such as @command{ssh}/@command{scp}. You can find the latest version of this document on the web at @uref{http://www.gnu.org/software/tramp/}. @ifhtml The latest release of @value{tramp} is available for @uref{ftp://ftp.gnu.org/gnu/tramp/, download}, or you may see @ref{Obtaining Tramp} for more details, including the Git server details. @value{tramp} also has a @uref{http://savannah.gnu.org/projects/tramp/, Savannah Project Page}. @end ifhtml There is a mailing list for @value{tramp}, available at @email{tramp-devel@@gnu.org}, and archived at @uref{http://lists.gnu.org/archive/html/tramp-devel/, the @value{tramp} Mail Archive}. @page @insertcopying @menu * Overview:: What @value{tramp} can and cannot do. For the end user: * Obtaining Tramp:: How to obtain @value{tramp}. * History:: History of @value{tramp}. @ifset installchapter * Installation:: Installing @value{tramp} with your Emacs. @end ifset * Configuration:: Configuring @value{tramp} for use. * Usage:: An overview of the operation of @value{tramp}. * Bug Reports:: Reporting Bugs and Problems. * Frequently Asked Questions:: Questions and answers from the mailing list. For the developer: * Files directories and localnames:: How file names, directories and localnames are mangled and managed. * Traces and Profiles:: How to Customize Traces. * GNU Free Documentation License:: The license for this documentation. * Function Index:: @value{tramp} functions. * Variable Index:: User options and variables. * Concept Index:: An item for each concept. @detailmenu --- The Detailed Node Listing --- @c @ifset installchapter Installing @value{tramp} with your Emacs * Installation parameters:: Parameters in order to control installation. * Load paths:: How to plug-in @value{tramp} into your environment. @end ifset Configuring @value{tramp} for use * Connection types:: Types of connections to remote hosts. * Inline methods:: Inline methods. * External methods:: External methods. * GVFS based methods:: GVFS based external methods. * Gateway methods:: Gateway methods. * Default Method:: Selecting a default method. * Default User:: Selecting a default user. * Default Host:: Selecting a default host. * Multi-hops:: Connecting to a remote host using multiple hops. * Customizing Methods:: Using Non-Standard Methods. * Customizing Completion:: Selecting config files for user/host name completion. * Password handling:: Reusing passwords for several connections. * Connection caching:: Reusing connection related information. * Predefined connection information:: Setting own connection related information. * Remote programs:: How @value{tramp} finds and uses programs on the remote host. * Remote shell setup:: Remote shell setup hints. * Android shell setup:: Android shell setup hints. * Auto-save and Backup:: Auto-save and Backup. * Windows setup hints:: Issues with Cygwin ssh. Using @value{tramp} * File name Syntax:: @value{tramp} file name conventions. * File name completion:: File name completion. * Ad-hoc multi-hops:: Declaring multiple hops in the file name. * Remote processes:: Integration with other Emacs packages. * Cleanup remote connections:: Cleanup remote connections. How file names, directories and localnames are mangled and managed * Localname deconstruction:: Breaking a localname into its components. * External packages:: Integration with external Lisp packages. @end detailmenu @end menu @node Overview @chapter An overview of @value{tramp} @cindex overview @value{tramp} is for transparently accessing remote files from within Emacs. @value{tramp} enables an easy, convenient, and consistent interface to remote files as if they are local files. @value{tramp}'s transparency extends to editing, version control, and @code{dired}. @value{tramp} can access remote hosts using any number of access methods, such as @command{rsh}, @command{rlogin}, @command{telnet}, and related programs. If these programs can successfully pass @acronym{ASCII]} characters, @value{tramp} can use them. @value{tramp} does not require or mandate 8-bit clean connections. @value{tramp}'s most common access method is through @command{ssh}, a more secure alternative to @command{ftp} and other older access methods. @value{tramp} on Windows operating systems is integrated with the PuTTY package, and uses the @command{plink} program. @value{tramp} mostly operates transparently in the background using the connection programs. As long as these programs enable remote login and can use the terminal, @value{tramp} can adapt them for seamless and transparent access. @value{tramp} temporarily transfers a remote file's contents to the local host editing and related operations. @value{tramp} can also transfer files between hosts using standard Emacs interfaces, a benefit of direct integration of @value{tramp} in Emacs. @value{tramp} can transfer files using any number of available host programs for remote files, such as @command{rcp}, @command{scp}, @command{rsync} or (under Windows) @command{pscp}. @value{tramp} provides easy ways to specify these programs and customize them to specific files, hosts, or access methods. For faster small-size file transfers, @value{tramp} supports encoded transfers directly through the shell using @command{mimencode} or @command{uuencode} provided such tools are available on the remote host. @subsubheading @value{tramp} behind the scenes @cindex behind the scenes @cindex details of operation @cindex how it works Accessing a remote file through @value{tramp} entails a series of actions, many of which are transparent to the user. Yet some actions may require user response (such as entering passwords or completing file names). One typical scenario, opening a file on a remote host, is presented here to illustrate the steps involved: @kbd{C-x C-f} to initiate find-file, enter part of the @value{tramp} file name, then hit @kbd{@key{TAB}} for completion. If this is the first time connection to that host, here's what happens: @itemize @item @value{tramp} invokes @samp{telnet @var{host}} or @samp{rsh @var{host} -l @var{user}} and establishes an external process to connect to the remote host. @value{tramp} communicates with the process through an Emacs buffer, which also shows output from the remote host. @item The remote host may prompt for a login name (for @command{telnet}, for example) in the buffer. If on the other hand, the login name was included in the file name portion, @value{tramp} sends the login name followed by a newline. @item The remote host may then prompt for a password or pass phrase (for @command{rsh} or for @command{telnet}). @value{tramp} displays the password prompt in the minibuffer. @value{tramp} then sends whatever is entered to the remote host, followed by a newline. @item @value{tramp} now waits for either the shell prompt or a failed login message. If @value{tramp} does not receive any messages within a timeout period (a minute, for example), then @value{tramp} responds with an error message about not finding the remote shell prompt. If any messages from the remote host, @value{tramp} displays them in the buffer. For any @samp{login failed} message from the remote host, @value{tramp} aborts the login attempt, and repeats the login steps again. @item Upon successful login and @value{tramp} recognizes the shell prompt from the remote host, @value{tramp} prepares the shell environment by turning off echoing, setting shell prompt, and other housekeeping chores. @strong{Note} that for the remote shell, @value{tramp} invokes @command{/bin/sh}. The remote host must recognize @samp{exec /bin/sh} and execute the appropriate shell. This shell must support Bourne shell syntax. @item @value{tramp} executes @command{cd} and @command{ls} commands to find which files exist on the remote host. @value{tramp} sometimes uses @command{echo} with globbing. @value{tramp} checks if a file or directory is writable with @command{test}. After each command, @value{tramp} parses the output from the remote host for completing the next operation. @item After remote file name completion, @value{tramp} transfers the file contents from the remote host. For inline transfers, @value{tramp} sends a command, such as @samp{mimencode -b /path/to/remote/file}, waits until the output has accumulated in the buffer, decodes that output to produce the file's contents. For external transfers, @value{tramp} sends a command as follows: @example rcp user@@host:/path/to/remote/file /tmp/tramp.4711 @end example @value{tramp} reads the local temporary file @file{/tmp/tramp.4711} into a buffer, and then deletes the temporary file. @item Edit, modify, change the buffer contents as normal, and then save the buffer wth @kbd{C-x C-s}. @item @value{tramp} transfers the buffer contents to the remote host in a reverse of the process using the appropriate inline or external program. @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}. @c For the end user @node Obtaining Tramp @chapter Obtaining @value{tramp} @cindex obtaining Tramp @value{tramp} is included as part of Emacs (since Emacs version 22.1). @value{tramp} is also freely packaged for download on the Internet at @uref{ftp://ftp.gnu.org/gnu/tramp/}. @value{tramp} development versions are available on Git servers. Development versions contain new and incomplete features. One way to obtain from Git server is to visit the Savannah project page at the following URL and then clicking on the Git link in the navigation bar at the top. @noindent @uref{http://savannah.gnu.org/projects/tramp/} @noindent Another way is to follow the terminal session below: @example @group ] @strong{cd ~/emacs} ] @strong{git clone git://git.savannah.gnu.org/tramp.git} @end group @end example @noindent From behind a firewall: @example @group ] @strong{git config --global http.proxy http://user:pwd@@proxy.server.com:8080} ] @strong{git clone http://git.savannah.gnu.org/r/tramp.git} @end group @end example @noindent Tramp developers: @example ] @strong{git clone login@@git.sv.gnu.org:/srv/git/tramp.git} @end example @noindent After one of the above commands, @file{~/emacs/tramp} will containing the latest version of @value{tramp}. @noindent To fetch updates from the repository, use git pull: @example @group ] @strong{cd ~/emacs/tramp} ] @strong{git pull} @end group @end example @noindent Run @command{autoconf} as follows to generate an up-to-date @file{configure} script: @example @group ] @strong{cd ~/emacs/tramp} ] @strong{autoconf} @end group @end example @node History @chapter History of @value{tramp} @cindex history @cindex development history @value{tramp} development started at the end of November 1998 as @file{rssh.el}. It provided only one method of access. It used @command{ssh} for login and @command{scp} to transfer file contents. The name was changed to @file{rcp.el} before it got its preset name @value{tramp}. New methods of remote access were added, so was support for version control. April 2000 was the first time when multi-hop methods were added. In July 2002, @value{tramp} unified file names with Ange-FTP@. In July 2004, proxy hosts replaced multi-hop methods. Running commands on remote hosts was introduced in December 2005. Support for gateways since April 2007. GVFS integration started in February 2009. Remote commands on Windows hosts since September 2011. Ad-hoc multi-hop methods (with a changed syntax) re-enabled in November 2011. In November 2012, added Juergen Hoetzel's @file{tramp-adb.el}. XEmacs support has been stopped in January 2016. @c Installation chapter is necessary only in case of standalone @c installation. Text taken from trampinst.texi. @ifset installchapter @include trampinst.texi @end ifset @node Configuration @chapter Configuring @value{tramp} @cindex configuration @cindex default configuration @value{tramp} is initially configured to use the @command{scp} program to connect to the remote host. Just type @kbd{C-x C-f} and then enter file name @file{@trampf{user@@host,/path/to.file}}. For details, @xref{Default Method}, @xref{Default User}, @xref{Default Host}. For problems related to the behavior of the remote shell, @xref{Remote shell setup}. For changing the connection type and file access method from the defaults to one of several other options, @xref{Connection types}. @strong{Note} that some user options described in these examples are not auto loaded by Emacs. All examples require @value{tramp} is installed and loaded: @lisp @group (custom-set-variables '(tramp-verbose 6 nil (tramp) "Enable remote command traces")) @end group @end lisp @menu * Connection types:: Types of connections to remote hosts. * Inline methods:: Inline methods. * External methods:: External methods. * GVFS based methods:: GVFS based external methods. * Gateway methods:: Gateway methods. * Default Method:: Selecting a default method. Here we also try to help those who don't have the foggiest which method is right for them. * Default User:: Selecting a default user. * Default Host:: Selecting a default host. * Multi-hops:: Connecting to a remote host using multiple hops. * Customizing Methods:: Using Non-Standard Methods. * Customizing Completion:: Selecting config files for user/host name completion. * Password handling:: Reusing passwords for several connections. * Connection caching:: Reusing connection related information. * Predefined connection information:: Setting own connection related information. * Remote programs:: How @value{tramp} finds and uses programs on the remote host. * Remote shell setup:: Remote shell setup hints. * Android shell setup:: Android shell setup hints. * Auto-save and Backup:: Auto-save and Backup. * Windows setup hints:: Issues with Cygwin ssh. @end menu @node Connection types @section Types of connections to remote hosts @cindex connection types, overview @dfn{Inline method} and @dfn{external method} are the two basic types of access methods. While they both use the same remote shell access programs, such as @command{rsh}, @command{ssh}, or @command{telnet}, they differ in the file access methods. Choosing the right method becomes important for editing files, transferring large files, or operating on a large number of files. The performance of the external methods is generally better than that of the inline methods, at least for large files. This is caused by the need to encode and decode the data when transferring inline. The one exception to this rule are the @option{scp}-based access methods. While these methods do see better performance when actually transferring files, the overhead of the cryptographic negotiation at startup may drown out the improvement in file transfer times. External methods should be configured such a way that they don't require a password (with @command{ssh-agent}, or such alike). Modern @command{scp} implementations offer options to reuse existing @command{ssh} connections, which will be enabled by default if available. If it isn't possible, you should consider @ref{Password handling}, otherwise you will be prompted for a password every copy action. @node Inline methods @section Inline methods @cindex inline methods @cindex methods, inline Inline methods use the same login connection to transfer file contents. Inline methods are quick and easy for small files. They depend on the availability of suitable encoding and decoding programs on the remote host. For local source and destination, @value{tramp} may use built-in equivalents of such programs in Emacs. Inline methods can work in situations where an external transfer program is unavailable. Inline methods also work when transferring files between different @emph{user identities} on the same host. @cindex uuencode @cindex mimencode @cindex base-64 encoding @value{tramp} checks the remote host for the availability and usability of @command{mimencode} (part of the @command{metamail} package) or @command{uuencode}. @value{tramp} uses the first reliable command it finds. @value{tramp}'s search path can be customized, see @ref{Remote programs}. In case both @command{mimencode} and @command{uuencode} are unavailable, @value{tramp} first transfers a small Perl program to the remote host, and then tries that program for encoding and decoding. To increase transfer speeds for large text files, use compression before encoding. The user option @option{tramp-inline-compress-start-size} specifies the file size for such optimization. @table @asis @item @option{rsh} @cindex method rsh @cindex rsh method @command{rsh} is an option for connecting to hosts within local networks since @command{rsh} is not as secure as other methods. @item @option{ssh} @cindex method ssh @cindex ssh method @command{ssh} is a more secure option than others to connect to a remote host. @command{ssh} can also take extra parameters as port numbers. For example, a host on port 42 is specified as @file{host#42} (the real host name, a hash sign, then a port number). It is the same as passing @samp{-p 42} to the @command{ssh} command. @item @option{telnet} @cindex method telnet @cindex telnet method Connecting to a remote host with @command{telnet} is as insecure as the @option{rsh} method. @item @option{su} @cindex method su @cindex su method Instead of connecting to a remote host, @command{su} program allows editing as another user. The host can be either @samp{localhost} or the host returned by the function @command{(system-name)}. See @ref{Multi-hops} for an exception to this behavior. @item @option{sudo} @cindex method sudo @cindex sudo method Similar to @option{su} method, @option{sudo} uses @command{sudo}. @command{sudo} must have sufficient rights to start a shell. @item @option{doas} @cindex method doas @cindex doas method This method is used on OpenBSD like the @command{sudo} command. @item @option{sg} @cindex method sg @cindex sg method The @command{sg} program allows editing as different group. The host can be either @samp{localhost} or the host returned by the function @command{(system-name)}. The user name must be specified, but it denotes a group name. See @ref{Multi-hops} for an exception to this behavior. @item @option{sshx} @cindex method sshx @cindex sshx method Works like @option{ssh} but without the extra authentication prompts. @option{sshx} uses @samp{ssh -t -t @var{host} -l @var{user} /bin/sh} to open a connection with a ``standard'' login shell. @strong{Note} that @option{sshx} does not bypass authentication questions. For example, if the host key of the remote host is not known, @option{sshx} will still ask ``Are you sure you want to continue connecting?''. @value{tramp} cannot handle such questions. Connections will have to be setup where logins can proceed without such questions. @option{sshx} is useful for Windows users when @command{ssh} triggers an error about allocating a pseudo tty. This happens due to missing shell prompts that confuses @value{tramp}. @option{sshx} supports the @samp{-p} argument. @item @option{krlogin} @cindex method krlogin @cindex krlogin method @cindex Kerberos (with krlogin method) This method is also similar to @option{ssh}. It uses the @command{krlogin -x} command only for remote host login. @item @option{ksu} @cindex method ksu @cindex ksu method @cindex Kerberos (with ksu method) This is another method from the Kerberos suite. It behaves like @option{su}. @item @option{plink} @cindex method plink @cindex plink method @option{plink} method is for Windows users with the PuTTY implementation of SSH@. It uses @samp{plink -ssh} to log in to the remote host. Check the @samp{Share SSH connections if possible} control for that session. @option{plink} method supports the @samp{-P} argument. @item @option{plinkx} @cindex method plinkx @cindex plinkx method Another method using PuTTY on Windows with session names instead of host names. @option{plinkx} calls @samp{plink -load @var{session} -t}. User names and port numbers must be defined in the session. Check the @samp{Share SSH connections if possible} control for that session. @end table @node External methods @section External methods @cindex methods, external @cindex external methods External methods operate over multiple channels, using the remote shell connection for some actions while delegating file transfers to an external transfer program. External methods save on the overhead of encoding and decoding of inline methods. Since external methods have the overhead of opening a new channel, files smaller than @var{tramp-copy-size-limit} still use inline methods. @table @asis @item @option{rcp} @cindex method rcp @cindex rcp method @cindex rcp (with rcp method) @cindex rsh (with rcp method) This method uses the @command{rsh} and @command{rcp} commands to connect to the remote host and transfer files. This is the fastest access method available. The alternative method @option{remcp} uses the @command{remsh} and @command{rcp} commands. @item @option{scp} @cindex method scp @cindex scp method @cindex scp (with scp method) @cindex ssh (with scp method) Using a combination of @command{ssh} to connect and @command{scp} to transfer is the most secure. While the performance is good, it is slower than the inline methods for smaller files. Though there is no overhead of encoding and decoding of the inline methods, @command{scp}'s cryptographic handshake negates those speed gains. @option{ssh}-based methods support @samp{-p} feature for specifying port numbers. For example, @file{host#42} passes @samp{-p 42} in the argument list to @command{ssh}, and @samp{-P 42} in the argument list to @command{scp}. @item @option{rsync} @cindex method rsync @cindex rsync method @cindex rsync (with rsync method) @cindex ssh (with rsync method) @command{ssh} command to connect in combination with @command{rsync} command to transfer is similar to the @option{scp} method. @command{rsync} performs much better than @command{scp} when transferring files that exist on both hosts. However, this advantage is lost if the file exists only on one side of the connection. This method supports the @samp{-p} argument. @item @option{scpx} @cindex method scpx @cindex scpx method @cindex scp (with scpx method) @cindex ssh (with scpx method) @option{scpx} is useful to avoid login shell questions. It is similar in performance to @option{scp}. @option{scpx} uses @samp{ssh -t -t @var{host} -l @var{user} /bin/sh} to open a connection. @option{scpx} is useful for Windows users when @command{ssh} triggers an error about allocating a pseudo tty. This happens due to missing shell prompts that confuses @value{tramp}. This method supports the @samp{-p} argument. @item @option{pscp} @item @option{psftp} @cindex method pscp @cindex pscp method @cindex pscp (with pscp method) @cindex plink (with pscp method) @cindex PuTTY (with pscp method) @cindex method psftp @cindex psftp method @cindex pscp (with psftp method) @cindex plink (with psftp method) @cindex PuTTY (with psftp method) These methods are similar to @option{scp} or @option{sftp}, but they use the @command{plink} command to connect to the remote host, and they use @command{pscp} or @command{psftp} for transferring the files. These programs are part of PuTTY, an SSH implementation for Windows. Check the @samp{Share SSH connections if possible} control for that session. These methods support the @samp{-P} argument. @item @option{fcp} @cindex method fcp @cindex fcp method @cindex fsh (with fcp method) @cindex fcp (with fcp method) This method is similar to @option{scp}, but uses @command{fsh} to connect and @command{fcp} to transfer files. @command{fsh/fcp}, a front-end for @command{ssh}, reuse @command{ssh} session by submitting several commands. This avoids the startup overhead due to @command{scp}'s secure connection. Inline methods have similar benefits. The command used for this connection is: @samp{fsh @var{host} -l @var{user} /bin/sh -i} @cindex method fsh @cindex fsh method @option{fsh} has no inline method since the multiplexing it offers is not useful for @value{tramp}. @command{fsh} connects to remote host and @value{tramp} keeps that one connection open. @item @option{nc} @cindex method nc @cindex nc method @cindex nc (with nc method) @cindex telnet (with nc method) Using @command{telnet} to connect and @command{nc} to transfer files is sometimes the only combination suitable for accessing routers or NAS hosts. These dumb devices have severely restricted local shells, such as the @command{busybox} and do not host any other encode or decode programs. @item @option{ftp} @cindex method ftp @cindex ftp method When @value{tramp} uses @option{ftp}, it forwards requests to whatever ftp program is specified by Ange FTP. This external program must be capable of servicing requests from @value{tramp}. @item @option{smb} @cindex method smb @cindex smb method This non-native @value{tramp} method connects via the Server Message Block (SMB) networking protocol to hosts running file servers that are typically based on @url{https://www.samba.org/,,Samba} or MS Windows. Using @command{smbclient} requires a few tweaks when working with @value{tramp}: The first directory in the localname must be a share name on the remote host. Since SMB shares end in the @code{$} character, @value{tramp} must use @code{$$} when specifying those shares to avoid environment variable substitutions. When @value{tramp} is not specific about the share name or uses the generic remote directory @file{/}, @command{smbclient} returns all available shares. Since SMB authentication is based on each SMB share, @value{tramp} prompts for a password even when accessing a different share on the same SMB host. This prompting can be suppressed by @ref{Password handling}. To accommodate user name/domain name syntax required by MS Windows authorization, @value{tramp} provides for an extended syntax in @code{user%domain} format (where user is username, @code{%} is the percent symbol, and domain is the windows domain name). An example: @example @trampfn{smb,daniel%BIZARRE@@melancholia,/daniel$$/.emacs} @end example where user @code{daniel} connects as a domain user to the SMB host @code{melancholia} in the windows domain @code{BIZARRE} to edit @file{.emacs} located in the home directory (share @code{daniel$}). Alternatively, for local WINS users (as opposed to domain users), substitute the domain name with the name of the local host in UPPERCASE as shown here: @example @trampfn{smb,daniel%MELANCHOLIA@@melancholia,/daniel$$/.emacs} @end example where user @code{daniel} connects as local user to the SMB host @code{melancholia} in the local domain @code{MELANCHOLIA} to edit @file{.emacs} located in the home directory (share @code{daniel$}). The domain name and user name are optional for @command{smbclient} authentication. When user name is not specified, @command{smbclient} uses the anonymous user (without prompting for password). This behavior is unlike other @value{tramp} methods, where local user name is substituted. The @option{smb} method is unavailable if Emacs is run under a local user authentication context in MS Windows. However such users can still access remote files using UNC file names instead of @value{tramp}: @example //melancholia/daniel$$/.emacs @end example UNC file name specification does not allow the specification of a different user name for authentication like the @command{smbclient} can. @item @option{adb} @cindex method adb @cindex adb method This method uses Android Debug Bridge program for accessing Android devices. The Android Debug Bridge must be installed locally for @value{tramp} to work. Some GNU/Linux distributions provide Android Debug Bridge as an installation package. Alternatively, the program is installed as part of the Android SDK@. @value{tramp} finds the @command{adb} program either via the @env{PATH} environment variable or the absolute path set in the user option @option{tramp-adb-program}. @value{tramp} connects to Android devices with @option{adb} only when the user option @option{tramp-adb-connect-if-not-connected} is not @code{nil}. Otherwise, the connection must be established outside Emacs. @value{tramp} does not require a host name part of the remote file name when a single Android device is connected to @command{adb}. @value{tramp} instead uses @file{@trampfn{adb,,}} as the default name. @command{adb devices} shows available host names. @option{adb} method normally does not need user name to authenticate on the Android device because it runs under the @command{adbd} process. But when a user name is specified, however, @value{tramp} applies an @command{su} in the syntax. When authentication does not succeed, especially on un-rooted Android devices, @value{tramp} displays login errors. For Android devices connected through TCP/IP, a port number can be specified using @file{device#42} host name syntax or @value{tramp} can use the default value as declared in @command{adb} command. Port numbers are not applicable to Android devices connected through USB@. @end table @node GVFS based methods @section GVFS based external methods @cindex methods, gvfs @cindex gvfs based methods @cindex dbus GVFS is the virtual file system for the Gnome Desktop, @uref{http://en.wikipedia.org/wiki/GVFS}. Remote files on GVFS are mounted locally through FUSE and @value{tramp} uses this locally mounted directory internally. Emacs uses the D-Bus mechanism to communicate with GVFS@. Emacs must have the message bus system, D-Bus integration active, @pxref{Top, , D-Bus, dbus}. @table @asis @item @option{afp} @cindex method afp @cindex afp method This method is for connecting to remote hosts with the Apple Filing Protocol for accessing files on Mac OS X volumes. @value{tramp} access syntax requires a leading volume (share) name, for example: @file{@trampfn{afp,user@@host,/volume}}. @item @option{dav} @item @option{davs} @cindex method dav @cindex method davs @cindex dav method @cindex davs method @option{dav} method provides access to WebDAV files and directories based on standard protocols, such as HTTP@. @option{davs} does the same but with SSL encryption. Both methods support the port numbers. @item @option{gdrive} @cindex method gdrive @cindex gdrive method @cindex Google Drive Via the @option{gdrive} method it is possible to access your Google Drive online storage. User and host name of the remote file name are your email address of the Google Drive credentials, like @file{@trampfn{gdrive,john.doe@@gmail.com,/}}. These credentials must be populated in your @command{Online Accounts} application outside Emacs. Since Google Drive uses cryptic blob file names internally, @value{tramp} works with the @code{display-name} of the files. This could produce unexpected behavior in case two files in the same directory have the same @code{display-name}, such a situation must be avoided. @item @option{obex} @cindex method obex @cindex obex method OBEX is an FTP-like access protocol for cell phones and similar simple devices. @value{tramp} supports OBEX over Bluetooth. @item @option{sftp} @cindex method sftp @cindex sftp method This method uses @command{sftp} in order to securely access remote hosts. @command{sftp} is a more secure option for connecting to hosts that for security reasons refuse @command{ssh} connections. @item @option{synce} @cindex method synce @cindex synce method @option{synce} method allows connecting to Windows Mobile devices. It uses GVFS for mounting remote files and directories via FUSE and requires the SYNCE-GVFS plugin. @end table @defopt tramp-gvfs-methods This user option is a list of external methods for GVFS@. By default, this list includes @option{afp}, @option{dav}, @option{davs}, @option{gdrive}, @option{obex}, @option{sftp} and @option{synce}. Other methods to include are: @option{ftp} and @option{smb}. @end defopt @node Gateway methods @section Gateway methods @cindex methods, gateway @cindex gateway methods Gateway methods are for proxy host declarations (@pxref{Multi-hops}) so as to pass through firewalls and proxy servers. They are not like the other methods that declare direct connections to a remote host. A gateway method always comes with a port setting. @value{tramp} targets the port number with the gateway method @file{localhost#random_port} from where the firewall or proxy server is accessed. Gateway methods support user name and password declarations for authenticating the corresponding firewall or proxy server. Such authentication can be passed through only if granted access by system administrators. @table @asis @item @option{tunnel} @cindex method tunnel @cindex tunnel method This method implements an HTTP tunnel via the @command{CONNECT} command (conforming to RFC 2616, 2817 specifications). Proxy servers using HTTP version 1.1 or later protocol support this command. For authentication, this protocol uses only @option{Basic Authentication} (see RFC 2617). When no port number is specified, this protocol defaults to @option{8080}. @item @option{socks} @cindex method socks @cindex socks method The @option{socks} method connects to SOCKSv5 servers (see RFC 1928) and supports @option{Username/Password Authentication}. The default port number for the socks server is @option{1080}, if not specified otherwise. @end table @node Default Method @section Selecting a default method @cindex default method @defopt tramp-default-method Default method is for transferring files. The user option @option{tramp-default-method} sets it. @value{tramp} uses this user option to determine the default method for remote file names that do not have one specified. @lisp (custom-set-variables '(tramp-default-method "ssh" nil (tramp))) @end lisp @end defopt @defopt tramp-default-method-alist Default methods for transferring files can be customized for specific user and host combinations through the user option @option{tramp-default-method-alist}. For example, the following two lines specify to use the @option{ssh} method for all user names matching @samp{john} and the @option{rsync} method for all host names matching @samp{lily}. The third line specifies to use the @option{su} method for the user @samp{root} on the host @samp{localhost}. @lisp @group (add-to-list 'tramp-default-method-alist '("" "john" "ssh")) (add-to-list 'tramp-default-method-alist '("lily" "" "rsync")) (add-to-list 'tramp-default-method-alist '("\\`localhost\\'" "\\`root\\'" "su")) @end group @end lisp @end defopt @noindent External methods performance faster for large files. @pxref{Inline methods}. @pxref{External methods}. Choosing the access method also depends on the security environment. For example, @option{rsh} and @option{telnet} methods that use clear text password transfers are inappropriate for over the Internet connections. Secure remote connections should use @option{ssh} that provide encryption. @subsection Which method to use? @cindex choosing the right method @value{tramp} provides maximum number of choices for maximum flexibility. Choosing which method depends on the hosts, clients, network speeds, and the security context. Start by using an inline method. External methods might be more efficient for large files, but most @value{tramp} users edit small files more often than large files. Enable compression, @var{tramp-inline-compress-start-size}, for a performance boost for large files. Since @command{ssh} has become the most common method of remote host access and it has the most reasonable security protocols, use @option{ssh} method. Typical @option{ssh} usage to edit the @file{/etc/motd} file on the otherhost: @example @kbd{C-x C-f @trampfn{ssh,root@@otherhost,/etc/motd} @key{RET}} @end example If @option{ssh} is unavailable for whatever reason, look for other obvious options. For Windows, try the @option{plink} method. For Kerberos, try @option{krlogin}. For editing local files as @option{su} or @option{sudo} methods, try the shortened syntax of @samp{root}: @example @kbd{C-x C-f @trampfn{su,,/etc/motd} @key{RET}} @end example For editing large files, @option{scp} is faster than @option{ssh}. @option{pscp} is faster than @option{plink}. But this speed improvement is not always true. @node Default User @section Selecting a default user @cindex default user @defopt tramp-default-user @value{tramp} file name can omit the user name part since @value{tramp} substitutes the currently logged-in user name. However this substitution can be overridden with @option{tramp-default-user}. For example: @lisp (custom-set-variables'(tramp-default-user "root" nil (tramp))) @end lisp @end defopt @defopt tramp-default-user-alist Instead of a single default user, @option{tramp-default-user-alist} allows multiple default user values based on access method or host name combinations. The alist can hold multiple values. For example, to use the @samp{john} as the default user for the domain @samp{somewhere.else} only: @lisp @group (add-to-list 'tramp-default-user-alist '("ssh" ".*\\.somewhere\\.else\\'" "john")) @end group @end lisp A Caution: @value{tramp} will override any default user specified in the configuration files outside Emacs, such as @file{~/.ssh/config}. To stop @value{tramp} from applying the default value, set the corresponding alist entry to nil: @lisp @group (add-to-list 'tramp-default-user-alist '("ssh" "\\`here\\.somewhere\\.else\\'" nil)) @end group @end lisp The last entry in @option{tramp-default-user-alist} should be reserved for catch-all or most often used login. @lisp @group (add-to-list 'tramp-default-user-alist '(nil nil "jonas") t) @end group @end lisp @end defopt @node Default Host @section Selecting a default host @cindex default host @defopt tramp-default-host When host name is omitted, @value{tramp} substitutes the value from the @option{tramp-default-host} user option. It is initially populated with the local hostname where Emacs is running. Both the default user and default host can be overridden as follows: @lisp @group (custom-set-variables '(tramp-default-user "john" nil (tramp)) '(tramp-default-host "target" nil (tramp))) @end group @end lisp With both defaults set, @samp{@trampfn{ssh,,}} will connect @value{tramp} to John's home directory on @code{target}. @strong{Note} @samp{/::} won't work, because @samp{/:} is the prefix for quoted file names. @ifinfo @pxref{Quoted File Names, , , emacs}. @end ifinfo @end defopt @defopt tramp-default-host-alist Instead of a single default host, @option{tramp-default-host-alist} allows multiple default host values based on access method or user name combinations. The alist can hold multiple values. While @option{tramp-default-host} is sufficient in most cases, some methods, like @option{adb}, require defaults overwritten. @end defopt @node Multi-hops @section Connecting to a remote host using multiple hops @cindex multi-hop @cindex proxy hosts Multi-hops are methods to reach hosts behind firewalls or to reach the outside world from inside a bastion host. With multi-hops, @value{tramp} can negotiate these hops with the appropriate user/host authentication at each hop. All methods until now have been the single hop kind, where the start and end points of the connection did not have intermediate check points. @defopt tramp-default-proxies-alist @option{tramp-default-proxies-alist} specifies proxy hosts to pass through. This user option is list of triples consisting of @code{(@var{host} @var{user} @var{proxy})}. The first match is the proxy host through which passes the file name and the target host matching @var{user}@@@var{host}. @var{host} and @var{user} are regular expressions or @code{nil}, interpreted as a regular expression which always matches. @var{proxy} is a literal @value{tramp} file name whose local name part is ignored, and the method and user name parts are optional. The method must be an inline or gateway method (@pxref{Inline methods}, @pxref{Gateway methods}). If @var{proxy} is @code{nil}, no additional hop is required reaching @var{user}@@@var{host}. For example, to pass through the host @samp{bastion.your.domain} as user @samp{bird} to reach remote hosts outside the local domain: @lisp @group (add-to-list 'tramp-default-proxies-alist '("\\." nil "@trampfn{ssh,bird@@bastion.your.domain,}")) (add-to-list 'tramp-default-proxies-alist '("\\.your\\.domain\\'" nil nil)) @end group @end lisp @strong{Note}: @code{add-to-list} adds elements at the beginning of a list. Therefore, most relevant rules must come last in the list. Proxy hosts can be cascaded in the alist. If there is another host called @samp{jump.your.domain}, which is the only host allowed to connect to @samp{bastion.your.domain}, then: @lisp @group (add-to-list 'tramp-default-proxies-alist '("\\`bastion\\.your\\.domain\\'" "\\`bird\\'" "@trampfn{ssh,jump.your.domain,}")) @end group @end lisp @var{proxy} can take patterns @code{%h} or @code{%u} for @var{host} or @var{user} respectively. To login as @samp{root} on remote hosts in the domain @samp{your.domain}, but login as @samp{root} is disabled for non-local access, then use this alist entry: @lisp @group (add-to-list 'tramp-default-proxies-alist '("\\.your\\.domain\\'" "\\`root\\'" "@trampfn{ssh,%h,}")) @end group @end lisp Opening @file{@trampfn{sudo,randomhost.your.domain,}} first connects to @samp{randomhost.your.domain} via @code{ssh} under your account name, and then perform @code{sudo -u root} on that host. It is key for the sudo method in the above example to be applied on the host after reaching it and not on the local host. @var{host}, @var{user} and @var{proxy} can also take Lisp forms. These forms when evaluated must return either a string or @code{nil}. To generalize (from the previous example): For all hosts, except my local one, first connect via @command{ssh}, and then apply @command{sudo -u root}: @lisp @group (add-to-list 'tramp-default-proxies-alist '(nil "\\`root\\'" "@trampfn{ssh,%h,}")) (add-to-list 'tramp-default-proxies-alist '((regexp-quote (system-name)) nil nil)) @end group @end lisp The above configuration allows @value{tramp} connection as @samp{root} to remote Ubuntu hosts. @option{tramp-default-proxies-alist} is also used for passing through firewalls or proxy servers. For example, the local host @samp{proxy.your.domain} on port 3128 serves as HTTP proxy to the outer world. User has access rights to another proxy server on @samp{host.other.domain}.@footnote{HTTP tunnels are intended for secure SSL/TLS communication. Therefore, many proxy servers restrict the tunnels to related target ports. You might need to run your ssh server on your target host @samp{host.other.domain} on such a port, like 443 (https). See @uref{http://savannah.gnu.org/maintenance/CvsFromBehindFirewall} for discussion of ethical issues.} Then the configuration is: @lisp @group (add-to-list 'tramp-default-proxies-alist '("\\`host\\.other\\.domain\\'" nil "@trampfn{tunnel,proxy.your.domain#3128,}")) @end group @end lisp Gateway methods in a multiple hop chain can be declared only as the first hop. @end defopt Passing through hops involves dealing with restricted shells, such as @command{rbash}. If @value{tramp} is made aware, then it would use them for proxies only. @defopt tramp-restricted-shell-hosts-alist An alist of regular expressions of hosts running restricted shells, such as @command{rbash}. @value{tramp} will then use them only as proxies. To specify the bastion host from the example above as running a restricted shell: @lisp @group (add-to-list 'tramp-restricted-shell-hosts-alist "\\`bastion\\.your\\.domain\\'") @end group @end lisp @end defopt @node Customizing Methods @section Using Non-Standard Methods @cindex customizing methods @cindex using non-standard methods @cindex create your own methods The @code{tramp-methods} variable currently has an exhaustive list of predefined methods. Any part of this list can be modified with more suitable settings. Refer to the Lisp documentation of that variable, accessible with @kbd{C-h v tramp-methods @key{RET}}. @node Customizing Completion @section Selecting config files for user/host name completion @cindex customizing completion @cindex selecting config files @vindex tramp-completion-function-alist @code{tramp-completion-function-alist} uses predefined files for user and host name completion (@pxref{File name completion}). For each method, it keeps a set of configuration files and a function that can parse that file. Each entry in @code{tramp-completion-function-alist} is of the form (@var{method} @var{pair1} @var{pair2} @dots{}). Each @var{pair} is composed of (@var{function} @var{file}). @var{function} is responsible for extracting user names and host names from @var{file} for completion. There are two functions which access this variable: @defun tramp-get-completion-function method This function returns the list of completion functions for @var{method}. Example: @example @group (tramp-get-completion-function "rsh") @result{} ((tramp-parse-rhosts "/etc/hosts.equiv") (tramp-parse-rhosts "~/.rhosts")) @end group @end example @end defun @defun tramp-set-completion-function method function-list This function sets @var{function-list} as list of completion functions for @var{method}. Example: @example @group (tramp-set-completion-function "ssh" '((tramp-parse-sconfig "/etc/ssh_config") (tramp-parse-sconfig "~/.ssh/config"))) @result{} ((tramp-parse-sconfig "/etc/ssh_config") (tramp-parse-sconfig "~/.ssh/config")) @end group @end example @end defun The following predefined functions parsing configuration files exist: @table @asis @item @code{tramp-parse-rhosts} @findex tramp-parse-rhosts This function parses files which are syntactical equivalent to @file{~/.rhosts}. It returns both host names and user names, if specified. @item @code{tramp-parse-shosts} @findex tramp-parse-shosts This function parses files which are syntactical equivalent to @file{~/.ssh/known_hosts}. Since there are no user names specified in such files, it can return host names only. @item @code{tramp-parse-sconfig} @findex tramp-parse-sconfig This function returns the host nicknames defined by @code{Host} entries in @file{~/.ssh/config} style files. @item @code{tramp-parse-shostkeys} @findex tramp-parse-shostkeys SSH2 parsing of directories @file{/etc/ssh2/hostkeys/*} and @file{~/ssh2/hostkeys/*}. Hosts are coded in file names @file{hostkey_@var{portnumber}_@var{host-name}.pub}. User names are always @code{nil}. @item @code{tramp-parse-sknownhosts} @findex tramp-parse-sknownhosts Another SSH2 style parsing of directories like @file{/etc/ssh2/knownhosts/*} and @file{~/ssh2/knownhosts/*}. This case, hosts names are coded in file names @file{@var{host-name}.@var{algorithm}.pub}. User names are always @code{nil}. @item @code{tramp-parse-hosts} @findex tramp-parse-hosts A function dedicated to @file{/etc/hosts} for host names. @item @code{tramp-parse-passwd} @findex tramp-parse-passwd A function which parses @file{/etc/passwd} files for user names. @item @code{tramp-parse-etc-group} @findex tramp-parse-etc-group A function which parses @file{/etc/group} files for group names. @item @code{tramp-parse-netrc} @findex tramp-parse-netrc A function which parses @file{~/.netrc} and @file{~/.authinfo}-style files. @end table To keep a custom file with custom data in a custom structure, a custom function has to be provided. This function must meet the following conventions: @defun my-tramp-parse file @var{file} must be either a file on the host, or @code{nil}. The function must return a list of (@var{user} @var{host}), which are taken as candidates for completion for user and host names. Example: @example @group (my-tramp-parse "~/.my-tramp-hosts") @result{} ((nil "toto") ("daniel" "melancholia")) @end group @end example @end defun @node Password handling @section Reusing passwords for several connections @cindex passwords To avoid repeated prompts for passwords, consider native caching mechanisms, such as @command{ssh-agent} for @option{ssh}-like methods, or @command{pageant} for @option{plink}-like methods. @value{tramp} offers alternatives when native solutions cannot meet the need. @anchor{Using an authentication file} @subsection Using an authentication file @vindex auth-sources The package @file{auth-source.el}, originally developed for No Gnus, reads passwords from different sources, @xref{Help for users, , auth-source, auth}. The default authentication file is @file{~/.authinfo.gpg}, but this can be changed via the variable @code{auth-sources}. @noindent A typical entry in the authentication file: @example machine melancholia port scp login daniel password geheim @end example The port can take any @value{tramp} method (@pxref{Inline methods}, @pxref{External methods}). Omitting port values matches all @value{tramp} methods. Setting @code{auth-source-debug} to @code{t} to debug messages. @anchor{Caching passwords} @subsection Caching passwords @value{tramp} can cache passwords as entered and reuse when needed for the same user or host name independent of the access method. @vindex password-cache-expiry @code{password-cache-expiry} sets the duration (in seconds) the passwords are remembered. Passwords are never saved permanently nor can they extend beyond the lifetime of the current Emacs session. Set @code{password-cache-expiry} to @code{nil} to disable expiration. @vindex password-cache Set @code{password-cache} to @code{nil} to disable password caching. @node Connection caching @section Reusing connection related information @cindex caching @vindex tramp-persistency-file-name For faster initial connection times, @value{tramp} stores previous connection properties in a file specified by the user option @option{tramp-persistency-file-name}. The default file name for @option{tramp-persistency-file-name} is @file{~/.emacs.d/tramp}. @value{tramp} reads this file during Emacs startup, and writes to it when exiting Emacs. Delete this file for @value{tramp} to recreate a new one on next Emacs startup. Set @option{tramp-persistency-file-name} to @code{nil} to disable storing connections persistently. To reuse connection information from the persistent list, @value{tramp} needs to uniquely identify every host. However in some cases, two different connections may result in the same persistent information. For example, connecting to a host using @command{ssh} and connecting to the same host through @code{sshd} on port 3001. Both access methods result in nearly identical persistent specifications @file{@trampfn{ssh,localhost,}} and @file{@trampfn{ssh,localhost#3001,}}. Changing host names could avoid duplicates. One way is to add a @option{Host} section in @file{~/.ssh/config} (@pxref{Frequently Asked Questions}). Another way is to apply multiple hops (@pxref{Multi-hops}). When @value{tramp} detects a change in the operating system version in a remote host (via the command @command{uname -sr}), it flushes all connection related information for that host and creates a new entry. @node Predefined connection information @section Setting own connection related information @vindex tramp-connection-properties For more precise customization, parameters specified by @code{tramp-methods} can be overwritten manually. Set @option{tramp-connection-properties} to manually override @code{tramp-methods}. Properties in this list are in the form @code{(@var{regexp} @var{property} @var{value})}. @var{regexp} matches remote file names. Use @code{nil} to match all. @var{property} is the property's name, and @var{value} is the property's value. @var{property} is any method specific parameter contained in @code{tramp-methods}. The parameter key in @code{tramp-methods} is a symbol name @code{tramp-}. To overwrite that property, use the string @samp{} for @var{property}. For example, this changes the remote shell: @lisp @group (add-to-list 'tramp-connection-properties (list (regexp-quote "@trampfn{ssh,user@@randomhost.your.domain,}") "remote-shell" "/bin/ksh")) @end group @group (add-to-list 'tramp-connection-properties (list (regexp-quote "@trampfn{ssh,user@@randomhost.your.domain,}") "remote-shell-login" '("-"))) @end group @end lisp The parameters @code{tramp-remote-shell} and @code{tramp-remote-shell-login} in @code{tramp-methods} now have new values for the remote host. @var{property} could also be any property found in @option{tramp-persistency-file-name}. To get around how restricted shells randomly drop connections, set the special property @samp{busybox}. For example: @lisp @group (add-to-list 'tramp-connection-properties (list (regexp-quote "@trampfn{ssh,user@@randomhost.your.domain,}") "busybox" t)) @end group @end lisp @node Remote programs @section How @value{tramp} finds and uses programs on the remote host @value{tramp} requires access to and rights to several commands on remote hosts: @command{ls}, @command{test}, @command{find} and @command{cat}. Besides there are other required programs for @ref{Inline methods} and @ref{External methods} of connection. To improve performance and accuracy of remote file access, @value{tramp} uses @command{perl} (or @command{perl5}) and @command{grep} when available. @defopt tramp-remote-path @option{tramp-remote-path} specifies which remote directory paths @value{tramp} can search for @ref{Remote programs}. @vindex tramp-default-remote-path @value{tramp} uses standard defaults, such as @file{/bin} and @file{/usr/bin}, which are reasonable for most hosts. To accommodate differences in hosts and paths, for example, @file{/bin:/usr/bin} on Debian GNU/Linux or @file{/usr/xpg4/bin:/usr/ccs/bin:/usr/bin:/opt/SUNWspro/bin} on Solaris, @value{tramp} queries the remote host with @command{getconf PATH} and updates the symbol @code{tramp-default-remote-path}. For instances where hosts keep obscure locations for paths for security reasons, manually add such paths to local @file{.emacs} as shown below for @value{tramp} to use when connecting. @lisp (add-to-list 'tramp-remote-path "/usr/local/perl/bin") @end lisp @vindex tramp-own-remote-path Another way to find the remote path is to use the path assigned to the remote user by the remote host. @value{tramp} does not normally retain this remote path after login. However, @code{tramp-own-remote-path} preserves the path value, which can be used to update @option{tramp-remote-path}. @lisp (add-to-list 'tramp-remote-path 'tramp-own-remote-path) @end lisp @strong{Note} that this works only if your remote @command{/bin/sh} shell supports the login argument @samp{-l}. @end defopt When remote search paths are changed, local @value{tramp} caches must be recomputed. To force @value{tramp} to recompute afresh, exit Emacs, remove the persistent file (@pxref{Connection caching}), and restart Emacs. @node Remote shell setup @section Remote shell setup hints @cindex remote shell setup @cindex @file{.profile} file @cindex @file{.login} file @cindex shell init files @value{tramp} checks for the availability of standard programs in the usual locations. Common tactics include successively trying @command{test -e}, @command{/usr/bin/test -e}, and @command{/bin/test -e}. @command{ls -d} is another approach. But these approaches do not help with these new login patterns. When @value{tramp} encounters two-factor logins or additional challenge questions, such as entering birth date or security code or passphrase, @value{tramp} needs a few more configuration steps to accommodate them. The difference between a password prompt and a passphrase prompt is that the password for completing the login while the passphrase is for authorizing access to local authentication information, such as the ssh key. There is no one configuration to accommodate all the variations in login security, especially not the exotic ones. However, @value{tramp} provides a few tweaks to address the most common ones. @table @asis @item @option{tramp-shell-prompt-pattern} @vindex tramp-shell-prompt-pattern @option{tramp-shell-prompt-pattern} is for remote login shell prompt, which may not be the same as the local login shell prompt, @code{shell-prompt-pattern}. Since most hosts use identical prompts, @value{tramp} sets a similar default value for both prompts. @item @option{tramp-password-prompt-regexp} @item @option{tramp-wrong-passwd-regexp} @vindex tramp-password-prompt-regexp @vindex tramp-wrong-passwd-regexp @value{tramp} uses @option{tramp-password-prompt-regexp} to distinguish between prompts for passwords and prompts for passphrases. By default, @option{tramp-password-prompt-regexp} handles the detection in English language environments. See a localization example below: @lisp @group (custom-set-variables '(tramp-password-prompt-regexp (concat "^.*" (regexp-opt '("passphrase" "Passphrase" ;; English "password" "Password" ;; Deutsch "passwort" "Passwort" ;; Français "mot de passe" "Mot de passe") t) ".*:\0? *") nil (tramp))) @end group @end lisp Similar localization may be necessary for handling wrong password prompts, for which @value{tramp} uses @option{tramp-wrong-passwd-regexp}. @item @command{tset} and other questions @cindex Unix command tset @cindex tset Unix command @vindex tramp-terminal-type To suppress inappropriate prompts for terminal type, @value{tramp} sets the @env{TERM} to @code{dumb} before the remote login process begins via the user option @option{tramp-terminal-type}. This will silence common @command{tset} related prompts. @value{tramp}'s strategy for handling such prompts (commonly triggered from login scripts on remote hosts) is to set the environment variables so that no prompts interrupt the shell initialization process. @vindex tramp-actions-before-shell An alternative approach is to configure @value{tramp} with strings that can identify such questions using @code{tramp-actions-before-shell}. Example: @lisp @group (defconst my-tramp-prompt-regexp (concat (regexp-opt '("Enter the birth date of your mother:") t) "\\s-*") "Regular expression matching my login prompt question.") @end group @group (defun my-tramp-action (proc vec) "Enter \"19000101\" in order to give a correct answer." (save-window-excursion (with-current-buffer (tramp-get-connection-buffer vec) (tramp-message vec 6 "\n%s" (buffer-string)) (tramp-send-string vec "19000101")))) @end group @group (add-to-list 'tramp-actions-before-shell '(my-tramp-prompt-regexp my-tramp-action)) @end group @end lisp @item Conflicting names for users and variables in @file{.profile} When a user name is the same as a variable name in a local file, such as @file{.profile}, then @value{tramp} may send incorrect values for environment variables. To avoid incorrect values, change the local variable name to something different from the user name. For example, if the user name is @env{FRUMPLE}, then change the variable name to @env{FRUMPLE_DIR}. @item Non-Bourne commands in @file{.profile} When the remote host's @file{.profile} is also used for shells other than Bourne shell, then some incompatible syntaxes for commands in @file{.profile} may trigger errors in Bourne shell on the host and may not complete client's @value{tramp} connections. One example of a Bourne shell incompatible syntax in @file{.profile}: using @command{export FOO=bar} instead of @command{FOO=bar; export FOO}. After remote login, @value{tramp} will trigger an error during its execution of @command{/bin/sh} on the remote host because Bourne shell does not recognize the export command as entered in @file{.profile}. Likewise, (@code{~}) character in paths will cause errors because Bourne shell does not do (@code{~}) character expansions. One approach to avoiding these incompatibilities is to make all commands in @file{~/.shrc} and @file{~/.profile} Bourne shell compatible so @value{tramp} can complete connections to that remote. To accommodate using non-Bourne shells on that remote, use other shell-specific config files. For example, bash can use @file{~/.bash_profile} and ignore @file{.profile}. @item Interactive shell prompt @value{tramp} redefines the remote shell prompt internally for robust parsing. This redefinition affects the looks of a prompt in an interactive remote shell through commands, such as @kbd{M-x shell}. Such prompts, however, can be reset to something more readable and recognizable using these @value{tramp} variables. @value{tramp} sets the @env{INSIDE_EMACS} variable in the startup script file @file{~/.emacs_SHELLNAME}. @env{SHELLNAME} is @code{bash} or equivalent shell names. Change it by setting the environment variable @env{ESHELL} in the @file{.emacs} as follows: @lisp (setenv "ESHELL" "bash") @end lisp Then re-set the prompt string in @file{~/.emacs_SHELLNAME} as follows: @example @group # Reset the prompt for remote Tramp shells. if [ "$@{INSIDE_EMACS/*tramp*/tramp@}" == "tramp" ] ; then PS1="[\u@@\h \w]$ " fi @end group @end example @ifinfo @xref{Interactive Shell, , , emacs}. @end ifinfo @item @command{busybox} / @command{nc} @cindex Unix command nc @cindex nc Unix command @value{tramp}'s @option{nc} method uses the @command{nc} command to install and execute a listener as follows (see @code{tramp-methods}): @example # nc -l -p 42 @end example The above command-line syntax has changed with @command{busybox} versions. If @command{nc} refuses the @samp{-p} parameter, then overwrite as follows: @lisp @group (add-to-list 'tramp-connection-properties `(,(regexp-quote "192.168.0.1") "remote-copy-args" (("-l") ("%r")))) @end group @end lisp @noindent where @samp{192.168.0.1} is the remote host IP address (@pxref{Predefined connection information}). @end table @node Android shell setup @section Android shell setup hints @cindex android shell setup @value{tramp} uses the @option{adb} method to access Android devices. Android devices provide a restricted shell access through an USB connection. The local host must have the Android SDK installed. Applications such as @code{SSHDroid} that run @command{sshd} process on the Android device can accept any @option{ssh}-based methods provided these settings are adjusted: @command{sh} must be specified for remote shell since Android devices do not provide @command{/bin/sh}. @command{sh} will then invoke whatever shell is installed on the device with this setting: @lisp @group (add-to-list 'tramp-connection-properties (list (regexp-quote "192.168.0.26") "remote-shell" "sh")) @end group @end lisp @noindent where @samp{192.168.0.26} is the Android device's IP address. (@pxref{Predefined connection information}). @value{tramp} requires preserving @env{PATH} environment variable from user settings. Android devices prefer @file{/system/xbin} path over @file{/system/bin}. Both of these are set as follows: @lisp @group (add-to-list 'tramp-remote-path 'tramp-own-remote-path) (add-to-list 'tramp-remote-path "/system/xbin") @end group @end lisp @noindent When the Android device is not @samp{rooted}, specify a writable directory for temporary files: @lisp (add-to-list 'tramp-remote-process-environment "TMPDIR=$HOME") @end lisp @noindent Open a remote connection with the command @kbd{C-x C-f @trampfn{ssh,192.168.0.26#2222,}}, where @command{sshd} is listening on port @samp{2222}. To add a corresponding entry to the @file{~/.ssh/config} file (recommended), use this: @example @group Host android HostName 192.168.0.26 User root Port 2222 @end group @end example @noindent To use the host name @samp{android} instead of the IP address shown in the previous example, fix the connection properties as follows: @lisp @group (add-to-list 'tramp-connection-properties (list (regexp-quote "android") "remote-shell" "sh")) @end group @end lisp @noindent Open a remote connection with a more concise command @kbd{C-x C-f @trampfn{ssh,android,}}. @node Auto-save and Backup @section Auto-save and Backup configuration @cindex auto-save @cindex backup @vindex backup-directory-alist To avoid @value{tramp} from saving backup files owned by @samp{root} to locations accessible to others, default backup settings in @option{backup-directory-alist} have to be altered. Here's a scenario where files could be inadvertently exposed. Emacs by default writes backup files to the same directory as the original files unless changed to another location, such as @file{~/.emacs.d/backups/}. Such a directory will also be used by default by @value{tramp} when using, say, a restricted file @file{@trampfn{su,root@@localhost,/etc/secretfile}}. The backup file of the secretfile is now owned by the user logged in from @value{tramp} and not @samp{root}. When @option{backup-directory-alist} is @code{nil} (the default), such problems do not occur. To ``turn off'' the backup feature for @value{tramp} files and stop @value{tramp} from saving to the backup directory, use this: @lisp @group (add-to-list 'backup-directory-alist (cons tramp-file-name-regexp nil)) @end group @end lisp @noindent Disabling backups can be targeted to just the @option{su} and @option{sudo} methods: @lisp @group (setq backup-enable-predicate (lambda (name) (and (normal-backup-enable-predicate name) (not (let ((method (file-remote-p name 'method))) (when (stringp method) (member method '("su" "sudo")))))))) @end group @end lisp Another option is to create better backup file naming with user and host names prefixed to the file name. For example, transforming @file{/etc/secretfile} to @file{~/.emacs.d/backups/!su:root@@localhost:!etc!secretfile}, set the @value{tramp} user option @option{tramp-backup-directory-alist} from the existing user option @option{backup-directory-alist}. Then @value{tramp} backs up to a file name that is transformed with a prefix consisting of the DIRECTORY name. This file name prefixing happens only when the DIRECTORY is an absolute local file name. @noindent Example: @lisp @group (add-to-list 'backup-directory-alist (cons "." "~/.emacs.d/backups/")) (custom-set-variables '(tramp-backup-directory-alist backup-directory-alist 6 nil (tramp))) @end group @end lisp @noindent The backup file name of @file{@trampfn{su,root@@localhost,/etc/secretfile}} would be @ifset unified @file{@trampfn{su,root@@localhost,~/.emacs.d/backups/!su:root@@localhost:!etc!secretfile~}} @end ifset @ifset separate @file{@trampfn{su,root@@localhost,~/.emacs.d/backups/![su!root@@localhost]!etc!secretfile~}} @end ifset Just as for backup files, similar issues of file naming affect auto-saving @value{tramp} files. Auto-saved files are saved in the directory specified by the user option @option{auto-save-file-name-transforms}. By default this is set to the local temporary directory. But in some versions of Debian GNU/Linux, this points to the source directory where the Emacs was compiled. Reset such values to a valid directory. Set @option{auto-save-file-name-transforms} to @code{nil} to save auto-saved files to the same directory as the original file. Alternatively, set the user option @option{tramp-auto-save-directory} to direct all auto saves to that location. @node Windows setup hints @section Issues with Cygwin ssh @cindex Cygwin, issues This section is incomplete. Please share your solutions. @cindex method sshx with Cygwin @cindex sshx method with Cygwin Cygwin's @command{ssh} works only with a Cygwin version of Emacs. To check for compatibility: type @kbd{M-x eshell}, and start @kbd{ssh test.host}. Incompatibilities trigger this message: @example Pseudo-terminal will not be allocated because stdin is not a terminal. @end example Some older versions of Cygwin's @command{ssh} work with the @option{sshx} access method. Consult Cygwin's FAQ at @uref{http://cygwin.com/faq/} for details. @cindex Cygwin and fakecygpty @cindex fakecygpty and Cygwin On @uref{https://www.emacswiki.org/emacs/SshWithNTEmacs, the Emacs Wiki} it is explained how to use the helper program @code{fakecygpty} to fix this problem. @cindex method scpx with Cygwin @cindex scpx method with Cygwin When using the @option{scpx} access method, Emacs may call @command{scp} with Windows file naming, such as @code{c:/foo}. But the version of @command{scp} that is installed with Cygwin does not know about Windows file naming, which causes it to incorrectly look for a host named @code{c}. A workaround: write a wrapper script for @option{scp} to convert Windows file names to Cygwin file names. @cindex Cygwin and ssh-agent @cindex SSH_AUTH_SOCK and Emacs on Windows When using the @command{ssh-agent} on Windows for password-less interaction, @option{ssh} methods depend on the environment variable @env{SSH_AUTH_SOCK}. But this variable is not set when Emacs is started from a Desktop shortcut and authentication fails. One workaround is to use a Windows based SSH Agent, such as Pageant. It is part of the Putty Suite of tools. The fallback is to start Emacs from a shell. @node Usage @chapter Using @value{tramp} @cindex using @value{tramp} @value{tramp} operates transparently, accessing remote files as if they are local. However, @value{tramp} employs a formalized remote file naming syntax to perform its functions transparently. This syntax consists of many parts specifying access methods, authentication, host names, and file names. Ange FTP uses a similar syntax. @cindex type-ahead Unlike opening local files in Emacs, which are instantaneous, opening remote files in @value{tramp} is slower at first. Sometimes there is a noticeable delay before the prompts for passwords or authentication appear in the minibuffer. Hitting @kbd{@key{RET}} or other keys during this gap will be processed by Emacs. This type-ahead facility is a feature of Emacs that may cause missed prompts when using @value{tramp}. @menu * File name Syntax:: @value{tramp} file name conventions. * File name completion:: File name completion. * Ad-hoc multi-hops:: Declaring multiple hops in the file name. * Remote processes:: Integration with other Emacs packages. * Cleanup remote connections:: Cleanup remote connections. @end menu @node File name Syntax @section @value{tramp} file name conventions @cindex file name syntax @cindex file name examples @file{@trampf{host,localfilename}} opens file @var{localfilename} on the remote host @var{host}, using the default method. @xref{Default Method}. @table @file @item @value{prefix}melancholia@value{postfix}.emacs For the file @file{.emacs} located in the home directory, on the host @code{melancholia}. @item @value{prefix}melancholia.danann.net@value{postfix}.emacs For the file @file{.emacs} specified using the fully qualified domain name of the host. @item @value{prefix}melancholia@value{postfix}~/.emacs For the file @file{.emacs} specified using the @file{~}, which is expanded. @item @value{prefix}melancholia@value{postfix}~daniel/.emacs For the file @file{.emacs} located in @code{daniel}'s home directory on the host, @code{melancholia}. The @file{~} construct is expanded to the home directory of that user on the remote host. @item @value{prefix}melancholia@value{postfix}/etc/squid.conf For the file @file{/etc/squid.conf} on the host @code{melancholia}. @end table @var{host} can take IPv4 or IPv6 address, as in @file{@trampf{127.0.0.1,.emacs}} or @file{@trampf{@value{ipv6prefix}::1@value{ipv6postfix},.emacs}}. @ifset unified For syntactical reasons, IPv6 addresses must be embedded in square brackets @file{@value{ipv6prefix}} and @file{@value{ipv6postfix}}. @end ifset By default, @value{tramp} will use the current local user name as the remote user name for log in to the remote host. Specifying a different name using the proper syntax will override this default behavior: @example @trampf{user@@host,path/to.file} @end example @file{@trampf{daniel@@melancholia,.emacs}} is for file @file{.emacs} in @code{daniel}'s home directory on the host, @code{melancholia}. Specify other file access methods (@pxref{Inline methods}, @pxref{External methods}) as part of the file name. Method name comes before user name, as in @file{@value{prefix}@var{method}@value{postfixhop}} (Note the trailing colon). The syntax specifications for user, host, and file do not change. To connect to the host @code{melancholia} as @code{daniel}, using @option{ssh} method for @file{.emacs} in @code{daniel}'s home directory, the full specification is: @file{@trampfn{ssh,daniel@@melancholia,.emacs}}. A remote file name containing a host name, which is the same string as a method name, is not allowed. For specifying port numbers, affix @file{#} to the host name. For example: @file{@trampfn{ssh,daniel@@melancholia#42,.emacs}}. @node File name completion @section File name completion @cindex file name completion @value{tramp} can complete the following @value{tramp} file name components: method names, user names, host names, and file names located on remote hosts. Enable this by activating partial completion in @file{.emacs}. @ifinfo @xref{Completion Options, , , emacs}. @end ifinfo For example, type @kbd{C-x C-f @value{prefix}t @key{TAB}}, @value{tramp} completion choices show up as @example @group @c @multitable {@trampfn{telnet,melancholia.danann.net,}} {@trampfn{telnet,192.168.0.1,}} @multitable @columnfractions .5 .5 @item @value{prefixhop}telnet@value{postfixhop} @tab tmp/ @item @value{prefixhop}toto@value{postfix} @tab @end multitable @end group @end example @samp{@value{prefixhop}telnet@value{postfixhop}} is a possible completion for the respective method, @samp{tmp/} stands for the directory @file{/tmp} on your local host, and @samp{@value{prefixhop}toto@value{postfix}} might be a host @value{tramp} has detected in your @file{~/.ssh/known_hosts} file (when using @option{ssh} as default method). Type @kbd{e @key{TAB}} for the minibuffer completion to @samp{@value{prefix}telnet@value{postfixhop}}. Typing @kbd{@key{TAB}} shows host names @value{tramp} from @file{/etc/hosts} file, for example. @example @group @multitable @columnfractions .5 .5 @c @multitable {@trampfn{telnet,melancholia.danann.net,}} {@trampfn{telnet,192.168.0.1,}} @item @trampfn{telnet,127.0.0.1,} @tab @trampfn{telnet,192.168.0.1,} @c @item @trampfn{telnet,@value{ipv6prefix}::1@value{ipv6postfix},} @tab @trampfn{telnet,localhost,} @item @value{prefix}telnet@value{postfixhop}@value{ipv6prefix}::1@value{ipv6postfix}@value{postfix} @tab @trampfn{telnet,localhost,} @item @trampfn{telnet,melancholia.danann.net,} @tab @trampfn{telnet,melancholia,} @end multitable @end group @end example Choose a host from the above list and then continue to complete file names on that host. When the configuration (@pxref{Customizing Completion}) includes user names, then the completion lists will account for the user names as well. Remote hosts previously visited or hosts whose connections are kept persistently (@pxref{Connection caching}) will be included in the completion lists. After remote host name completion comes completion of file names on the remote host. It works the same as with local host file completion except that killing with double-slash @file{//} kills only the file name part of the @value{tramp} file name syntax. A triple-slash stands for the default behavior. @ifinfo @xref{Minibuffer File, , , emacs}. @end ifinfo @noindent Example: @example @group @kbd{C-x C-f @trampfn{telnet,melancholia,/usr/local/bin//etc} @key{TAB}} @print{} @trampfn{telnet,melancholia,/etc} @kbd{C-x C-f @trampfn{telnet,melancholia,//etc} @key{TAB}} @print{} /etc @kbd{C-x C-f @trampfn{telnet,melancholia,/usr/local/bin///etc} @key{TAB}} @print{} /etc @end group @end example During file name completion, remote directory contents are re-read regularly to account for any changes in the filesystem that may affect the completion candidates. Such re-reads can account for changes to the file system by applications outside Emacs (@pxref{Connection caching}). @defopt tramp-completion-reread-directory-timeout The timeout is number of seconds since last remote command for rereading remote directory contents. A value of 0 re-reads immediately during file name completion, @code{nil} uses cached directory contents. @end defopt @node Ad-hoc multi-hops @section Declaring multiple hops in the file name @cindex multi-hop, ad-hoc @cindex proxy hosts, ad-hoc @value{tramp} file name syntax can accommodate ad hoc specification of multiple proxies without using @option{tramp-default-proxies-alist} configuration setup(@pxref{Multi-hops}). Each proxy is specified using the same syntax as the remote host specification minus the file name part. Each hop is separated by a @samp{|}. Chain the proxies from the starting host to the destination remote host name and file name. For example, hopping over a single proxy @samp{bird@@bastion} to a remote file on @samp{you@@remotehost}: @example @c @kbd{C-x C-f @trampfn{ssh@value{postfixhop}bird@@bastion|ssh,you,remotehost,/path}} @kbd{C-x C-f @value{prefix}ssh@value{postfixhop}bird@@bastion|ssh@value{postfixhop}you@@remotehost@value{postfix}/path} @end example Proxies can take patterns @code{%h} or @code{%u}. @value{tramp} adds the ad-hoc definitions on the fly to @option{tramp-default-proxies-alist} and is available for re-use during that Emacs session. Subsequent @value{tramp} connections to the same remote host can then use the shortcut form: @samp{@trampfn{ssh,you@@remotehost,/path}}. @defopt tramp-save-ad-hoc-proxies For ad-hoc definitions to be saved automatically in @option{tramp-default-proxies-alist} for future Emacs sessions, set @option{tramp-save-ad-hoc-proxies} to non-@code{nil}. @lisp (custom-set-variables '(tramp-save-ad-hoc-proxies t nil (tramp))) @end lisp @end defopt @node Remote processes @section Integration with other Emacs packages @cindex compile @cindex recompile @value{tramp} supports starting new running processes on the remote host for discovering remote file names. Emacs packages on the remote host need no specific modifications for @value{tramp}'s use. This type of integration does not work with the @option{ftp} method, and does not support the pty association as specified in @code{start-file-process}. @code{process-file} and @code{start-file-process} work on the remote host when the variable @code{default-directory} is remote: @lisp @group (let ((default-directory "/ssh:remote.host:")) (start-file-process "grep" (get-buffer-create "*grep*") "/bin/sh" "-c" "grep -e tramp *")) @end group @end lisp Remote processes do not apply to GVFS (see @ref{GVFS based methods}) because the remote file system is mounted on the local host and @value{tramp} just accesses by changing the @code{default-directory}. @value{tramp} starts a remote process when a command is executed in a remote file or directory buffer. As of now, these packages have been integrated to work with @value{tramp}: @file{compile.el} (commands like @code{compile} and @code{grep}) and @file{gud.el} (@code{gdb} or @code{perldb}). For @value{tramp} to find the command on the remote, it must be accessible through the default search path as setup by @value{tramp} upon first connection. Alternatively, use an absolute path or extend @option{tramp-remote-path} (see @ref{Remote programs}): @lisp @group (add-to-list 'tramp-remote-path "~/bin") (add-to-list 'tramp-remote-path "/appli/pub/bin") @end group @end lisp Customize user option @option{tramp-remote-process-environment} to suit the remote program's environment for the remote host. @option{tramp-remote-process-environment} is a list of strings structured similar to @code{process-environment}, where each element is a string of the form @samp{ENVVARNAME=VALUE}. To avoid any conflicts with local host environment variables set through local configuration files, such as @file{~/.profile}, use @samp{ENVVARNAME=} to unset them for the remote environment. @noindent Use @code{add-to-list} to add entries: @lisp (add-to-list 'tramp-remote-process-environment "JAVA_HOME=/opt/java") @end lisp Modifying or deleting already existing values in the @option{tramp-remote-process-environment} list may not be feasible on restricted remote hosts. For example, some system administrators disallow changing @env{HISTORY} environment variable. To accommodate such restrictions when using @value{tramp}, fix the @option{tramp-remote-process-environment} by the following code in the local @file{.emacs} file: @lisp @group (let ((process-environment tramp-remote-process-environment)) (setenv "HISTORY" nil) (setq tramp-remote-process-environment process-environment)) @end group @end lisp Setting the @env{ENV} environment variable instructs some shells to read an initialization file. Per default, @value{tramp} has disabled this. You could overwrite this behavior by evaluating @lisp @group (let ((process-environment tramp-remote-process-environment)) (setenv "ENV" "$HOME/.profile") (setq tramp-remote-process-environment process-environment)) @end group @end lisp @value{tramp} does not use the defaults specified in @code{process-environment} for running @code{process-file} or @code{start-file-process} on remote hosts. When values from @code{process-environment} are needed for remote processes, then set them as follows: @lisp @group (let ((process-environment (cons "HGPLAIN=1" process-environment))) (process-file @dots{})) @end group @end lisp This works only for environment variables not already set in the @code{process-environment}. For integrating other Emacs packages so @value{tramp} can execute remotely, please file a bug report. @xref{Bug Reports}. @subsection Running remote programs that create local X11 windows To allow a remote program to create an X11 window on the local host, set the @env{DISPLAY} environment variable for the remote host as follows in the local @file{.emacs} file: @lisp @group (add-to-list 'tramp-remote-process-environment (format "DISPLAY=%s" (getenv "DISPLAY"))) @end group @end lisp @noindent @code{(getenv "DISPLAY")} should return a recognizable name for the local host that the remote host can redirect X11 window interactions. If querying for a recognizable name is not possible for whatever reason, then replace @code{(getenv "DISPLAY")} with a hard-coded, fixed name. Note that using @code{:0} for X11 display name here will not work as expected. An alternate approach is specify @code{ForwardX11 yes} or @code{ForwardX11Trusted yes} in the file @file{~/.ssh/config} on the local host. @subsection Running @code{shell} on a remote host @cindex shell Set @option{explicit-shell-file-name} to the appropriate shell name when using @value{tramp} between two hosts with different operating systems, such as @samp{windows-nt} and @samp{gnu/linux}. This option ensures the correct name of the remote shell program. Starting with Emacs 24, when @option{explicit-shell-file-name} is equal to @code{nil}, calling @code{shell} interactively will prompt for a shell name. @subsection Running @code{shell-command} on a remote host @cindex shell-command @code{shell-command} executes commands synchronously or asynchronously on remote hosts and displays output in buffers on the local host. Example: @example @group @kbd{C-x C-f @trampfn{sudo,,} @key{RET}} @kbd{M-! tail -f /var/log/syslog.log & @key{RET}} @end group @end example @command{tail} command outputs continuously to the local buffer, @file{*Async Shell Command*} @kbd{M-x auto-revert-tail-mode} runs similarly showing continuous output. @subsection Running @code{eshell} on a remote host @cindex eshell @value{tramp} is integrated into @file{eshell.el}, which enables interactive eshell sessions on remote hosts at the command prompt. You must add the module @code{eshell-tramp} to @code{eshell-modules-list}. Here's a sample interaction after opening @kbd{M-x eshell} on a remote host: @example @group @b{~ $} cd @trampfn{sudo,,/etc} @key{RET} @b{@trampfn{sudo,root@@host,/etc} $} hostname @key{RET} host @b{@trampfn{sudo,root@@host,/etc} $} id @key{RET} uid=0(root) gid=0(root) groups=0(root) @b{@trampfn{sudo,root@@host,/etc} $} find-file shadow @key{RET} # @b{@trampfn{sudo,root@@host,/etc} $} @end group @end example @code{eshell} in Emacs 23.2 added custom @code{su} and @code{sudo} commands that set the default directory correctly for the @file{*eshell*} buffer. @value{tramp} silently updates @option{tramp-default-proxies-alist} with an entry for this directory (@pxref{Multi-hops}): @example @group @b{~ $} cd @trampfn{ssh,user@@remotehost,/etc} @key{RET} @b{@trampfn{ssh,user@@remotehost,/etc} $} find-file shadow @key{RET} File is not readable: @trampfn{ssh,user@@remotehost,/etc/shadow} @b{@trampfn{ssh,user@@remotehost,/etc} $} sudo find-file shadow @key{RET} # @end group @group @b{@trampfn{ssh,user@@remotehost,/etc} $} su - @key{RET} @b{@trampfn{su,root@@remotehost,/root} $} id @key{RET} uid=0(root) gid=0(root) groups=0(root) @b{@trampfn{su,root@@remotehost,/root} $} @end group @end example @anchor{Running a debugger on a remote host} @subsection Running a debugger on a remote host @cindex gud @cindex gdb @cindex perldb @file{gud.el} provides a unified interface to symbolic debuggers @ifinfo (@ref{Debuggers, , , emacs}). @end ifinfo @value{tramp} can run debug on remote hosts by calling @code{gdb} with a remote file name: @example @group @kbd{M-x gdb @key{RET}} @b{Run gdb (like this):} gdb -i=mi @trampfn{ssh,host,~/myprog} @key{RET} @end group @end example Since the remote @code{gdb} and @code{gdb-inferior} processes do not belong to the same process group on the remote host, there will be a warning, which can be ignored: @example &"warning: GDB: Failed to set controlling terminal: Operation not permitted\n" @end example Relative file names are based on the remote default directory. When @file{myprog.pl} exists in @file{@trampfn{ssh,host,/home/user}}, valid calls include: @example @group @kbd{M-x perldb @key{RET}} @b{Run perldb (like this):} perl -d myprog.pl @key{RET} @end group @end example Just the local part of a remote file name, such as @kbd{perl -d /home/user/myprog.pl}, is not possible. Arguments of the program to be debugged must be literal, can take relative or absolute paths, but not remote paths. @subsection Running remote processes on Windows hosts @cindex winexe @cindex powershell @command{winexe} runs processes on a remote Windows host, and @value{tramp} can use it for @code{process-file} and @code{start-file-process}. @code{tramp-smb-winexe-program} specifies the local @command{winexe} command. Powershell V2.0 on the remote host is required to run processes triggered from @value{tramp}. @option{explicit-shell-file-name} and @option{explicit-*-args} have to be set properly so @kbd{M-x shell} can open a proper remote shell on a Windows host. To open @command{cmd}, set it as follows: @lisp @group (setq explicit-shell-file-name "cmd" explicit-cmd-args '("/q")) @end group @end lisp @noindent To open @command{powershell} as a remote shell, use this: @lisp @group (setq explicit-shell-file-name "powershell" explicit-powershell-args '("-file" "-")) @end group @end lisp @node Cleanup remote connections @section Cleanup remote connections @cindex cleanup @value{tramp} provides several ways to flush remote connections. @deffn Command tramp-cleanup-connection vec This command flushes all connection related objects. @option{vec} is the internal representation of a remote connection. When called interactively, this command lists active remote connections in the minibuffer. Each connection is of the format @file{@trampfn{method,user@@host,}}. Flushing remote connections also cleans the password cache (@pxref{Password handling}), file cache, connection cache (@pxref{Connection caching}), and connection buffers. @end deffn @deffn Command tramp-cleanup-this-connection Flushes only the current buffer's remote connection objects, the same as in @code{tramp-cleanup-connection}. @end deffn @deffn Command tramp-cleanup-all-connections Flushes all active remote connection objects, the same as in @code{tramp-cleanup-connection}. @end deffn @deffn Command tramp-cleanup-all-buffers Just as for @code{tramp-cleanup-all-connections}, all remote connections are cleaned up in addition to killing buffers related to that remote connection. @end deffn @node Bug Reports @chapter Reporting Bugs and Problems @cindex bug reports @value{tramp}'s development team is actively engaged in solving bugs and problems and looks to feature requests and suggestions. @value{tramp}'s mailing list is the place for more advice and information on working with @value{tramp}, solving problems, discussing, and general discussions about @value{tramp}. @value{tramp}'s mailing list is moderated but even non-subscribers can post for moderator approval. Sometimes this approval step may take as long as 48 hours due to public holidays. @email{tramp-devel@@gnu.org} is the mailing list. Messages sent to this address go to all the subscribers. This is @emph{not} the address to send subscription requests to. To subscribe to the mailing list, visit: @uref{http://lists.gnu.org/mailman/listinfo/tramp-devel/, the @value{tramp} Mail Subscription Page}. @ifset installchapter Before sending a bug report, run the test suite first @ref{Testing}. @end ifset @findex tramp-bug Check if the bug or problem is already addressed in @xref{Frequently Asked Questions}. Run @kbd{M-x tramp-bug} to generate a buffer with details of the system along with the details of the @value{tramp} installation. Please include these details with the bug report. The bug report must describe in as excruciating detail as possible the steps required to reproduce the problem. These details must include the setup of the remote host and any special or unique conditions that exist. Include a minimal test case that reproduces the problem. This will help the development team find the best solution and avoid unrelated detours. To exclude cache-related problems, flush all caches before running the test, @ref{Cleanup remote connections}. When including @value{tramp}'s messages in the bug report, increase the verbosity level to 6 (@pxref{Traces and Profiles, Traces}) in the @file{~/.emacs} file before repeating steps to the bug. Include the contents of the @file{*tramp/foo*} and @file{*debug tramp/foo*} buffers with the bug report. @strong{Note} that a verbosity level greater than 6 is not necessary at this stage. Also note that a verbosity level of 6 or greater, the contents of files and directories will be included in the debug buffer. Passwords typed in @value{tramp} will never be included there. @node Frequently Asked Questions @chapter Frequently Asked Questions @cindex frequently asked questions @cindex FAQ @itemize @bullet @item Where is the latest @value{tramp}? @value{tramp} is available at the GNU URL: @noindent @uref{ftp://ftp.gnu.org/gnu/tramp/} @noindent @value{tramp}'s GNU project page is located here: @noindent @uref{http://savannah.gnu.org/projects/tramp/} @item Which systems does it work on? The package works successfully on Emacs 23, Emacs 24, and Emacs 25. While Unix and Unix-like systems are the primary remote targets, @value{tramp} has equal success connecting to other platforms, such as MS Windows XP/Vista/7. @item How to speed up @value{tramp}? @value{tramp} does many things in the background, some of which depends on network speeds, response speeds of remote hosts, and authentication delays. During these operations, @value{tramp}'s responsiveness slows down. Some suggestions within the scope of @value{tramp}'s settings include: Use an external method, such as @option{scp}, which are faster than internal methods. Keep the file @option{tramp-persistency-file-name}, which is where @value{tramp} caches remote information about hosts and files. Caching is enabled by default. Don't disable it. Set @code{remote-file-name-inhibit-cache} to @code{nil} if remote files are not independently updated outside @value{tramp}'s control. That cache cleanup will be necessary if the remote directories or files are updated independent of @value{tramp}. Set @option{tramp-completion-reread-directory-timeout} to @code{nil} to speed up completions, @ref{File name completion}. Disable version control to avoid delays: @lisp @group (setq vc-ignore-dir-regexp (format "\\(%s\\)\\|\\(%s\\)" vc-ignore-dir-regexp tramp-file-name-regexp)) @end group @end lisp Disable excessive traces. Set @code{tramp-verbose} to 3 or lower, default being 3. Increase trace levels temporarily when hunting for bugs. @item @value{tramp} does not connect to the remote host Three main reasons for why @value{tramp} does not connect to the remote host: @itemize @minus @item Unknown characters in the prompt @value{tramp} needs a clean recognizable prompt on the remote host for accurate parsing. Shell prompts that contain escape sequences for coloring cause parsing problems. @ref{Remote shell setup} for customizing prompt detection using regular expressions. To check if the remote host's prompt is being recognized, use this test: switch to @value{tramp} connection buffer @file{*tramp/foo*}, put the cursor at the top of the buffer, and then apply the following expression: @example @kbd{M-: (re-search-forward (concat tramp-shell-prompt-pattern "$"))} @end example If the cursor has not moved to the prompt at the bottom of the buffer, then @value{tramp} has failed to recognize the prompt. When using zsh on remote hosts, disable zsh line editor because zsh uses left-hand side and right-hand side prompts in parallel. Add the following line to @file{~/.zshrc}: @example [ $TERM = "dumb" ] && unsetopt zle && PS1='$ ' @end example When using fish shell on remote hosts, disable fancy formatting by adding the following to @file{~/.config/fish/config.fish}: @example @group function fish_prompt if test $TERM = "dumb" echo "\$ " else @dots{} end end @end group @end example When using WinSSHD on remote hosts, @value{tramp} does not recognize the strange prompt settings. A similar problem exist with the iTerm2 shell integration, which sends proprietary escape codes when starting a shell. This can be suppressed by changing the respective integration snippet in your @file{~/.profile} like this: @example @group [ $TERM = "dumb" ] || \ test -e "$@{HOME@}/.iterm2_shell_integration.bash" && \ source "$@{HOME@}/.iterm2_shell_integration.bash" @end group @end example @item Echoed characters after login @value{tramp} suppresses echos from remote hosts with the @command{stty -echo} command. But sometimes it is too late to suppress welcome messages from the remote host containing harmful control characters. Using @option{sshx} or @option{scpx} methods can avoid this problem because they allocate a pseudo tty. @xref{Inline methods}. @item @value{tramp} stops transferring strings longer than 500 characters Set @code{tramp-chunksize} to 500 to get around this problem, which is related to faulty implementation of @code{process-send-string} on HP-UX, FreeBSD and Tru64 Unix systems. Consult the documentation for @code{tramp-chunksize} to see when this is necessary. Set @code{file-precious-flag} to @code{t} for files accessed by @value{tramp} so the file contents are checked using checksum by first saving to a temporary file. @ifinfo @pxref{Saving Buffers, , , elisp} @end ifinfo @lisp @group (add-hook 'find-file-hook (lambda () (when (file-remote-p default-directory) (set (make-local-variable 'file-precious-flag) t)))) @end group @end lisp @end itemize @item @value{tramp} does not recognize if a @command{ssh} session hangs @command{ssh} sessions on the local host hang when the network is down. @value{tramp} cannot safely detect such hangs. The network configuration for @command{ssh} can be configured to kill such hangs with the following command in the @file{~/.ssh/config}: @example @group Host * ServerAliveInterval 5 @end group @end example @item @value{tramp} does not use default @command{ssh} @code{ControlPath} @value{tramp} overwrites @code{ControlPath} settings when initiating @command{ssh} sessions. @value{tramp} does this to fend off a stall if a master session opened outside the Emacs session is no longer open. That is why @value{tramp} prompts for the password again even if there is an @command{ssh} already open. @vindex tramp-ssh-controlmaster-options Some @command{ssh} versions support a @code{ControlPersist} option, which allows you to set the @code{ControlPath} provided the variable @code{tramp-ssh-controlmaster-options} is customized as follows: @lisp @group (setq tramp-ssh-controlmaster-options (concat "-o ControlPath=/tmp/ssh-ControlPath-%%r@@%%h:%%p " "-o ControlMaster=auto -o ControlPersist=yes")) @end group @end lisp Note how "%r", "%h" and "%p" must be encoded as "%%r", "%%h" and "%%p". @vindex tramp-use-ssh-controlmaster-options If the @file{~/.ssh/config} is configured appropriately for the above behavior, then any changes to @command{ssh} can be suppressed with this @code{nil} setting: @lisp @group (custom-set-variables '(tramp-use-ssh-controlmaster-options nil nil (tramp))) @end group @end lisp @item File name completion does not work with @value{tramp} @acronym{ANSI} escape sequences from the remote shell may cause errors in @value{tramp}'s parsing of remote buffers. To test if this is the case, open a remote shell and check if the output of @command{ls} is in color. To disable @acronym{ANSI} escape sequences from the remote hosts, disable @samp{--color=yes} or @samp{--color=auto} in the remote host's @file{.bashrc} or @file{.profile}. Turn this alias on and off to see if file name completion works. @item File name completion does not work in directories with large number of files This may be related to globbing, which is the use of shell's ability to expand wild card specifications, such as @samp{*.c}. For directories with large number of files, globbing might exceed the shell's limit on length of command lines and hang. @value{tramp} uses globbing. To test if globbing hangs, open a shell on the remote host and then run @command{ls -d * ..?* > /dev/null}. When testing, ensure the remote shell is the same shell (@command{/bin/sh}, @command{ksh} or @command{bash}), that @value{tramp} uses when connecting to that host. @item How to get notified after @value{tramp} completes file transfers? Make Emacs beep after reading from or writing to the remote host with the following code in @file{~/.emacs} file. @lisp @group (defadvice tramp-handle-write-region (after tramp-write-beep-advice activate) "Make tramp beep after writing a file." (interactive) (beep)) @end group @group (defadvice tramp-handle-do-copy-or-rename-file (after tramp-copy-beep-advice activate) "Make tramp beep after copying a file." (interactive) (beep)) @end group @group (defadvice tramp-handle-insert-file-contents (after tramp-insert-beep-advice activate) "Make tramp beep after inserting a file." (interactive) (beep)) @end group @end lisp @item How to get a Visual Warning when working with @samp{root} privileges? Host indication in the mode line? Install @file{tramp-theme} from GNU ELPA via Emacs' Package Manager. Enable it via @kbd{M-x load-theme @key{RET} tramp}. Further customization is explained in user option @option{tramp-theme-face-remapping-alist}. @item Remote host does not understand default options for directory listing Emacs computes the @command{dired} options based on the local host but if the remote host cannot understand the same @command{ls} command, then set them with a hook as follows: @lisp @group (add-hook 'dired-before-readin-hook (lambda () (when (file-remote-p default-directory) (setq dired-actual-switches "-al")))) @end group @end lisp @item Why is @file{~/.sh_history} file on the remote host growing? Due to @command{ksh} saving tilde expansions triggered by @value{tramp}, the history file is probably growing rapidly. To fix, turn off saving history by putting this shell code in the @file{.kshrc} file: @example @group if [ -f $HOME/.sh_history ] ; then /bin/rm $HOME/.sh_history fi if [ "$@{HISTFILE-unset@}" != "unset" ] ; then unset HISTFILE fi if [ "$@{HISTSIZE-unset@}" != "unset" ] ; then unset HISTSIZE fi @end group @end example For @option{ssh}-based method, add the following line to your @file{~/.ssh/environment} file: @example HISTFILE=/dev/null @end example @item How to shorten long file names when typing in @value{tramp}? Adapt several of these approaches to reduce typing. If the full name is @file{@trampfn{ssh,news@@news.my.domain,/opt/news/etc}}, then: @enumerate @item Use default values for method name and user name: You can define default methods and user names for hosts, (@pxref{Default Method}, @pxref{Default User}): @lisp @group (custom-set-variables '(tramp-default-method "ssh" nil (tramp)) '(tramp-default-user "news" nil (tramp))) @end group @end lisp The reduced typing: @kbd{C-x C-f @trampf{news.my.domain,/opt/news/etc}}. @strong{Note} that there are some useful shortcuts already. Accessing your local host as @samp{root} user, is possible just by @kbd{C-x C-f @trampfn{su,,}}. @item Use configuration options of the access method: Programs used for access methods already offer powerful configurations (@pxref{Customizing Completion}). For @option{ssh}, configure the file @file{~/.ssh/config}: @example @group Host xy HostName news.my.domain User news @end group @end example The reduced typing: @kbd{C-x C-f @trampfn{ssh,xy,/opt/news/etc}}. Depending on the number of files in the directories, host names completion can further reduce key strokes: @kbd{C-x C-f @value{prefix}ssh@value{postfixhop}x @key{TAB}}. @item Use environment variables to expand long strings For long file names, set up environment variables that are expanded in the minibuffer. Environment variables are set either outside Emacs or inside Emacs with Lisp: @lisp (setenv "xy" "@trampfn{ssh,news@@news.my.domain,/opt/news/etc/}") @end lisp The reduced typing: @kbd{C-x C-f $xy @key{RET}}. @strong{Note} that file name cannot be edited here because the environment variables are not expanded during editing in the minibuffer. @item Define own keys: Redefine another key sequence in Emacs for @kbd{C-x C-f}: @lisp @group (global-set-key [(control x) (control y)] (lambda () (interactive) (find-file (read-file-name "Find Tramp file: " "@trampfn{ssh,news@@news.my.domain,/opt/news/etc/}")))) @end group @end lisp Simply typing @kbd{C-x C-y} would prepare minibuffer editing of file name. See @uref{https://www.emacswiki.org/emacs/TrampMode, the Emacs Wiki} for a more comprehensive example. @item Define own abbreviation (1): Abbreviation list expansion can be used to reduce typing long file names: @lisp @group (add-to-list 'directory-abbrev-alist '("^/xy" . "@trampfn{ssh,news@@news.my.domain,/opt/news/etc/}")) @end group @end lisp The reduced typing: @kbd{C-x C-f /xy @key{RET}}. @strong{Note} that file name cannot be edited here because the environment variables are not expanded during editing in the minibuffer. @item Define own abbreviation (2): The @code{abbrev-mode} gives additional flexibility for editing in the minibuffer: @lisp @group (define-abbrev-table 'my-tramp-abbrev-table '(("xy" "@trampfn{ssh,news@@news.my.domain,/opt/news/etc/}"))) @end group @group (add-hook 'minibuffer-setup-hook (lambda () (abbrev-mode 1) (setq local-abbrev-table my-tramp-abbrev-table))) @end group @group (defadvice minibuffer-complete (before my-minibuffer-complete activate) (expand-abbrev)) @end group @group ;; If you use partial-completion-mode (defadvice PC-do-completion (before my-PC-do-completion activate) (expand-abbrev)) @end group @end lisp The reduced typing: @kbd{C-x C-f xy @key{TAB}}. The minibuffer expands for further editing. @item Use bookmarks: Use bookmarks to save Tramp file names. @ifinfo @pxref{Bookmarks, , , emacs}. @end ifinfo Upon visiting a location with @value{tramp}, save it as a bookmark with @kbd{@key{menu-bar} @key{edit} @key{bookmarks} @key{set}}. To revisit that bookmark: @kbd{@key{menu-bar} @key{edit} @key{bookmarks} @key{jump}}. @item Use recent files: @file{recentf} remembers visited places. @ifinfo @pxref{File Conveniences, , , emacs}. @end ifinfo Keep remote file names in the recent list without have to check for their accessibility through remote access: @lisp (recentf-mode 1) @end lisp Reaching recently opened files: @kbd{@key{menu-bar} @key{file} @key{Open Recent}}. @item Use filecache: Since @file{filecache} remembers visited places, add the remote directory to the cache: @lisp @group @c `with-eval-after-load' has been introduced with Emacs 24.4. Shall @c be used when appropriate. (eval-after-load "filecache" '(file-cache-add-directory "@trampfn{ssh,news@@news.my.domain,/opt/news/etc/}")) @end group @end lisp Then use directory completion in the minibuffer with @kbd{C-x C-f C-@key{TAB}}. @item Use bbdb: @file{bbdb} has a built-in feature for Ange FTP files, which also works for @value{tramp} file names. @ifinfo @pxref{bbdb-ftp, Storing FTP sites in the BBDB, , bbdb}. @end ifinfo Load @file{bbdb} in Emacs: @lisp @group (require 'bbdb) (bbdb-initialize) @end group @end lisp Create a BBDB entry with @kbd{M-x bbdb-create-ftp-site}. Then specify a method and user name where needed. Examples: @example @group @kbd{M-x bbdb-create-ftp-site @key{RET}} @b{Ftp Site:} news.my.domain @key{RET} @b{Ftp Directory:} /opt/news/etc/ @key{RET} @b{Ftp Username:} ssh@value{postfixhop}news @key{RET} @b{Company:} @key{RET} @b{Additional Comments:} @key{RET} @end group @end example In BBDB buffer, access an entry by pressing the key @key{F}. @end enumerate Thanks to @value{tramp} users for contributing to these recipes. @item Why saved multi-hop file names do not work in a new Emacs session? When saving ad-hoc multi-hop @value{tramp} file names (@pxref{Ad-hoc multi-hops}) via bookmarks, recent files, filecache, bbdb, or another package, use the full ad-hoc file name including all hops, like @file{@trampfn{ssh,bird@@bastion|ssh@value{postfixhop}news.my.domain,/opt/news/etc}}. Alternatively, when saving abbreviated multi-hop file names @file{@trampfn{ssh,news@@news.my.domain,/opt/news/etc}}, the user option @option{tramp-save-ad-hoc-proxies} must be set non-@code{nil} value. @item How to connect to a remote Emacs session using @value{tramp}? Configure Emacs Client @ifinfo (@pxref{Emacs Server, , , emacs}). @end ifinfo Then on the remote host, start the Emacs Server: @lisp @group (require 'server) (setq server-host (system-name) server-use-tcp t) (server-start) @end group @end lisp If @code{(system-name)} of the remote host cannot be resolved on the local host, use IP address instead. Copy from the remote host the resulting file @file{~/.emacs.d/server/server} to the local host, to the same location. Then start Emacs Client from the command line: @example emacsclient @trampfn{ssh,user@@host,/file/to/edit} @end example @code{user} and @code{host} refer to the local host. To make Emacs Client an editor for other programs, use a wrapper script @file{emacsclient.sh}: @example @group #!/bin/sh emacsclient @trampfn{ssh,$(whoami)@@$(hostname --fqdn),$1} @end group @end example Then change the environment variable @env{EDITOR} to point to the wrapper script: @example export EDITOR=/path/to/emacsclient.sh @end example @item How to disable other packages from calling @value{tramp}? There are packages that call @value{tramp} without the user ever entering a remote file name. Even without applying a remote file syntax, some packages enable @value{tramp} on their own. How can users disable such features. @itemize @minus @item @file{ido.el} Disable @value{tramp} file name completion: @lisp (custom-set-variables '(ido-enable-tramp-completion nil)) @end lisp @item @file{rlogin.el} Disable remote directory tracking mode: @lisp (rlogin-directory-tracking-mode -1) @end lisp @end itemize @item How to disable @value{tramp}? @itemize @minus @item To keep Ange FTP as default the remote files access package, set this in @file{.emacs}: @lisp (custom-set-variables '(tramp-default-method "ftp" nil (tramp))) @end lisp @item To disable both @value{tramp} (and Ange FTP), set @code{tramp-mode} to @code{nil} in @file{.emacs}. @lisp (custom-set-variables '(tramp-mode nil nil (tramp))) @end lisp @item To unload @value{tramp}, type @kbd{M-x tramp-unload-tramp}. Unloading @value{tramp} resets Ange FTP plugins also. @end itemize @end itemize @c For the developer @node Files directories and localnames @chapter How file names, directories and localnames are mangled and managed. @menu * Localname deconstruction:: Splitting a localname into its component parts. * External packages:: Integrating with external Lisp packages. @end menu @node Localname deconstruction @section Splitting a localname into its component parts @value{tramp} package redefines lisp functions @code{file-name-directory} and @code{file-name-nondirectory} to accommodate the unique file naming syntax that @value{tramp} requires. The replacements dissect the file name, use the original handler for the localname, take that result, and then re-build the @value{tramp} file name. By relying on the original handlers for localnames, @value{tramp} benefits from platform specific hacks to the original handlers. @node External packages @section Integrating with external Lisp packages @subsection File name completion. For name completions in the minibuffer, @value{tramp} depends on the last input character to decide whether to look for method name completion or host name completion. For example, @kbd{C-x C-f @value{prefix}ssh@value{postfixhop} @key{TAB}} is not entirely clear if @option{ssh} is a method or a host name. But if the last input character was either @key{TAB}, @key{SPACE} or @kbd{?}, then @value{tramp} favors file name completion over host name completion. What about external packages using other characters to trigger file name completions? They must somehow signal this to @value{tramp}. Use the variable @code{non-essential} temporarily and bind it to non-@code{nil} value. @lisp @group (let ((non-essential t)) @dots{}) @end group @end lisp @subsection File attributes cache. Keeping a local cache of remote file attributes in sync with the remote host is a time-consuming operation. Flushing and re-querying these attributes can tax @value{tramp} to a grinding halt on busy remote servers. To get around these types of slow-downs in @value{tramp}'s responsiveness, set the @code{process-file-side-effects} to @code{nil} to stop @value{tramp} from flushing the cache. This is helpful in situations where callers to @code{process-file} know there are no file attribute changes. The let-bind form to accomplish this: @lisp @group (let (process-file-side-effects) @dots{}) @end group @end lisp For asynchronous processes, @value{tramp} uses a process sentinel to flush file attributes cache. When callers to @code{start-file-process} know beforehand no file attribute changes are expected, then the process sentinel should be set to the default state. In cases where the caller defines its own process sentinel, @value{tramp}'s process sentinel is overwritten. The caller can still flush the file attributes cache in its process sentinel with this code: @lisp @group (unless (memq (process-status proc) '(run open)) (dired-uncache remote-directory)) @end group @end lisp Since @value{tramp} traverses subdirectories starting with the root-directory, it is most likely sufficient to make the @code{default-directory} of the process buffer as the root directory. @node Traces and Profiles @chapter How to Customize Traces @value{tramp} messages are raised with verbosity levels ranging from 0 to 10. @value{tramp} does not display all messages; only those with a verbosity level less than or equal to @code{tramp-verbose}. The verbosity levels are @w{ 0} silent (no @value{tramp} messages at all) @*@indent @w{ 1} errors @*@indent @w{ 2} warnings @*@indent @w{ 3} connection to remote hosts (default verbosity) @*@indent @w{ 4} activities @*@indent @w{ 5} internal @*@indent @w{ 6} sent and received strings @*@indent @w{ 7} file caching @*@indent @w{ 8} connection properties @*@indent @w{ 9} test commands @*@indent @w{10} traces (huge) With @code{tramp-verbose} greater than or equal to 4, messages are also written to a @value{tramp} debug buffer. Such debug buffers are essential to bug and problem analyses. For @value{tramp} bug reports, set the @code{tramp-verbose} level to 6 (@pxref{Bug Reports}). The debug buffer is in @ifinfo @ref{Outline Mode, , , emacs}. @end ifinfo @ifnotinfo Outline Mode. @end ifnotinfo In this buffer, messages can be filtered by their level. To see messages up to verbosity level 5, enter @kbd{C-u 6 C-c C-q}. @ifinfo Other navigation keys are described in @ref{Outline Visibility, , , emacs}. @end ifinfo @value{tramp} handles errors internally. But to get a Lisp backtrace, both the error and the signal have to be set as follows: @lisp @group (setq debug-on-error t debug-on-signal t) @end group @end lisp To enable stepping through @value{tramp} function call traces, they have to be specifically enabled as shown in this code: @lisp @group (require 'trace) (dolist (elt (all-completions "tramp-" obarray 'functionp)) (trace-function-background (intern elt))) (untrace-function 'tramp-read-passwd) (untrace-function 'tramp-gw-basic-authentication) @end group @end lisp The buffer @file{*trace-output*} contains the output from the function call traces. Disable @code{tramp-read-passwd} and @code{tramp-gw-basic-authentication} to stop password strings from being written to @file{*trace-output*}. @node GNU Free Documentation License @appendix GNU Free Documentation License @include doclicense.texi @node Function Index @unnumbered Function Index @printindex fn @node Variable Index @unnumbered Variable Index @printindex vr @node Concept Index @unnumbered Concept Index @printindex cp @bye @c TODO @c @c * Say something about the .login and .profile files of the remote @c shells. @c * Explain how tramp.el works in principle: open a shell on a remote @c host and then send commands to it. @c * Consistent small or capitalized words especially in menus. @c * Make a unique declaration of @trampfn.