1 \input texinfo @c -*-texinfo-*-
8 @include tincinclude.texi
11 @dircategory Networking tools
13 * tinc: (tinc). The tinc Manual.
16 This is the info manual for @value{PACKAGE} version @value{VERSION}, a Virtual Private Network daemon.
18 Copyright @copyright{} 1998-2018 Ivo Timmermans,
19 Guus Sliepen <guus@@tinc-vpn.org> and
20 Wessel Dankers <wsl@@tinc-vpn.org>.
22 Permission is granted to make and distribute verbatim copies of this
23 manual provided the copyright notice and this permission notice are
24 preserved on all copies.
26 Permission is granted to copy and distribute modified versions of this
27 manual under the conditions for verbatim copying, provided that the
28 entire resulting derived work is distributed under the terms of a
29 permission notice identical to this one.
39 @subtitle Setting up a Virtual Private Network with tinc
40 @author Ivo Timmermans and Guus Sliepen
43 @vskip 0pt plus 1filll
44 This is the info manual for @value{PACKAGE} version @value{VERSION}, a Virtual Private Network daemon.
46 Copyright @copyright{} 1998-2018 Ivo Timmermans,
47 Guus Sliepen <guus@@tinc-vpn.org> and
48 Wessel Dankers <wsl@@tinc-vpn.org>.
50 Permission is granted to make and distribute verbatim copies of this
51 manual provided the copyright notice and this permission notice are
52 preserved on all copies.
54 Permission is granted to copy and distribute modified versions of this
55 manual under the conditions for verbatim copying, provided that the
56 entire resulting derived work is distributed under the terms of a
57 permission notice identical to this one.
62 @c ==================================================================
74 * Technical information::
75 * Platform specific information::
77 * Concept Index:: All used terms explained
81 @c ==================================================================
86 Tinc is a Virtual Private Network (VPN) daemon that uses tunneling and
87 encryption to create a secure private network between hosts on the
90 Because the tunnel appears to the IP level network code as a normal
91 network device, there is no need to adapt any existing software.
92 The encrypted tunnels allows VPN sites to share information with each other
93 over the Internet without exposing any information to others.
95 This document is the manual for tinc. Included are chapters on how to
96 configure your computer to use tinc, as well as the configuration
97 process of tinc itself.
100 * Virtual Private Networks::
102 * Supported platforms::
105 @c ==================================================================
106 @node Virtual Private Networks
107 @section Virtual Private Networks
110 A Virtual Private Network or VPN is a network that can only be accessed
111 by a few elected computers that participate. This goal is achievable in
112 more than just one way.
115 Private networks can consist of a single stand-alone Ethernet LAN. Or
116 even two computers hooked up using a null-modem cable. In these cases,
118 obvious that the network is @emph{private}, no one can access it from the
119 outside. But if your computers are linked to the Internet, the network
120 is not private anymore, unless one uses firewalls to block all private
121 traffic. But then, there is no way to send private data to trusted
122 computers on the other end of the Internet.
125 This problem can be solved by using @emph{virtual} networks. Virtual
126 networks can live on top of other networks, but they use encapsulation to
127 keep using their private address space so they do not interfere with
128 the Internet. Mostly, virtual networks appear like a single LAN, even though
129 they can span the entire world. But virtual networks can't be secured
130 by using firewalls, because the traffic that flows through it has to go
131 through the Internet, where other people can look at it.
133 As is the case with either type of VPN, anybody could eavesdrop. Or
134 worse, alter data. Hence it's probably advisable to encrypt the data
135 that flows over the network.
137 When one introduces encryption, we can form a true VPN. Other people may
138 see encrypted traffic, but if they don't know how to decipher it (they
139 need to know the key for that), they cannot read the information that flows
140 through the VPN. This is what tinc was made for.
143 @c ==================================================================
148 I really don't quite remember what got us started, but it must have been
149 Guus' idea. He wrote a simple implementation (about 50 lines of C) that
150 used the ethertap device that Linux knows of since somewhere
151 about kernel 2.1.60. It didn't work immediately and he improved it a
152 bit. At this stage, the project was still simply called "vpnd".
154 Since then, a lot has changed---to say the least.
157 Tinc now supports encryption, it consists of a single daemon (tincd) for
158 both the receiving and sending end, it has become largely
159 runtime-configurable---in short, it has become a full-fledged
160 professional package.
162 @cindex traditional VPNs
164 Tinc also allows more than two sites to connect to each other and form a single VPN.
165 Traditionally VPNs are created by making tunnels, which only have two endpoints.
166 Larger VPNs with more sites are created by adding more tunnels.
167 Tinc takes another approach: only endpoints are specified,
168 the software itself will take care of creating the tunnels.
169 This allows for easier configuration and improved scalability.
171 A lot can---and will be---changed. We have a number of things that we would like to
172 see in the future releases of tinc. Not everything will be available in
173 the near future. Our first objective is to make tinc work perfectly as
174 it stands, and then add more advanced features.
176 Meanwhile, we're always open-minded towards new ideas. And we're
180 @c ==================================================================
181 @node Supported platforms
182 @section Supported platforms
185 Tinc has been verified to work under Linux, FreeBSD, OpenBSD, NetBSD, MacOS/X (Darwin), Solaris, and Windows (both natively and in a Cygwin environment),
186 with various hardware architectures. These are some of the platforms
187 that are supported by the universal tun/tap device driver or other virtual network device drivers.
188 Without such a driver, tinc will most
189 likely compile and run, but it will not be able to send or receive data
193 For an up to date list of supported platforms, please check the list on
195 @uref{https://www.tinc-vpn.org/platforms/}.
203 @c Preparing your system
210 @c ==================================================================
212 @chapter Preparations
214 This chapter contains information on how to prepare your system to
218 * Configuring the kernel::
223 @c ==================================================================
224 @node Configuring the kernel
225 @section Configuring the kernel
228 * Configuration of Linux kernels::
229 * Configuration of FreeBSD kernels::
230 * Configuration of OpenBSD kernels::
231 * Configuration of NetBSD kernels::
232 * Configuration of Solaris kernels::
233 * Configuration of Darwin (MacOS/X) kernels::
234 * Configuration of Windows::
238 @c ==================================================================
239 @node Configuration of Linux kernels
240 @subsection Configuration of Linux kernels
242 @cindex Universal tun/tap
243 For tinc to work, you need a kernel that supports the Universal tun/tap device.
244 Most distributions come with kernels that already support this.
245 Here are the options you have to turn on when configuring a new kernel:
248 Code maturity level options
249 [*] Prompt for development and/or incomplete code/drivers
250 Network device support
251 <M> Universal tun/tap device driver support
254 It's not necessary to compile this driver as a module, even if you are going to
255 run more than one instance of tinc.
257 If you decide to build the tun/tap driver as a kernel module, add these lines
258 to @file{/etc/modules.conf}:
261 alias char-major-10-200 tun
265 @c ==================================================================
266 @node Configuration of FreeBSD kernels
267 @subsection Configuration of FreeBSD kernels
269 For FreeBSD version 4.1 and higher, tun and tap drivers are included in the default kernel configuration.
270 The tap driver can be loaded with @code{kldload if_tap}, or by adding @code{if_tap_load="YES"} to @file{/boot/loader.conf}.
273 @c ==================================================================
274 @node Configuration of OpenBSD kernels
275 @subsection Configuration of OpenBSD kernels
277 Recent versions of OpenBSD come with both tun and tap devices enabled in the default kernel configuration.
280 @c ==================================================================
281 @node Configuration of NetBSD kernels
282 @subsection Configuration of NetBSD kernels
284 For NetBSD version 1.5.2 and higher,
285 the tun driver is included in the default kernel configuration.
287 Tunneling IPv6 may not work on NetBSD's tun device.
290 @c ==================================================================
291 @node Configuration of Solaris kernels
292 @subsection Configuration of Solaris kernels
294 For Solaris 8 (SunOS 5.8) and higher,
295 the tun driver may or may not be included in the default kernel configuration.
296 If it isn't, the source can be downloaded from @uref{http://vtun.sourceforge.net/tun/}.
297 For x86 and sparc64 architectures, precompiled versions can be found at @uref{https://www.monkey.org/~dugsong/fragroute/}.
298 If the @file{net/if_tun.h} header file is missing, install it from the source package.
301 @c ==================================================================
302 @node Configuration of Darwin (MacOS/X) kernels
303 @subsection Configuration of Darwin (MacOS/X) kernels
305 Tinc on Darwin relies on a tunnel driver for its data acquisition from the kernel.
306 OS X version 10.6.8 and later have a built-in tun driver called "utun".
307 Tinc also supports the driver from @uref{http://tuntaposx.sourceforge.net/},
308 which supports both tun and tap style devices,
310 By default, tinc expects the tuntaposx driver to be installed.
311 To use the utun driver, set add @code{Device = utunX} to @file{tinc.conf},
312 where X is the desired number for the utun interface.
313 You can also omit the number, in which case the first free number will be chosen.
316 @c ==================================================================
317 @node Configuration of Windows
318 @subsection Configuration of Windows
320 You will need to install the latest TAP-Win32 driver from OpenVPN.
321 You can download it from @uref{https://openvpn.net/index.php/open-source/downloads.html}.
322 Using the Network Connections control panel,
323 configure the TAP-Win32 network interface in the same way as you would do from the tinc-up script,
324 as explained in the rest of the documentation.
327 @c ==================================================================
333 Before you can configure or build tinc, you need to have the LibreSSL or OpenSSL, zlib,
334 LZO, curses and readline libraries installed on your system. If you try to
335 configure tinc without having them installed, configure will give you an error
347 @c ==================================================================
348 @node LibreSSL/OpenSSL
349 @subsection LibreSSL/OpenSSL
353 For all cryptography-related functions, tinc uses the functions provided
354 by the LibreSSL or the OpenSSL library.
356 If this library is not installed, you will get an error when configuring
357 tinc for build. Support for running tinc with other cryptographic libraries
358 installed @emph{may} be added in the future.
360 You can use your operating system's package manager to install this if
361 available. Make sure you install the development AND runtime versions
364 If your operating system comes neither with LibreSSL or OpenSSL, you have to
365 install one manually. It is recommended that you get the latest version of
366 LibreSSL from @url{https://www.libressl.org/}. Instructions on how to
367 configure, build and install this package are included within the package.
368 Please make sure you build development and runtime libraries (which is the
371 If you installed the LibreSSL or OpenSSL libraries from source, it may be necessary
372 to let configure know where they are, by passing configure one of the
373 --with-openssl-* parameters. Note that you even have to use --with-openssl-* if you
377 --with-openssl=DIR LibreSSL/OpenSSL library and headers prefix
378 --with-openssl-include=DIR LibreSSL/OpenSSL headers directory
379 (Default is OPENSSL_DIR/include)
380 --with-openssl-lib=DIR LibreSSL/OpenSSL library directory
381 (Default is OPENSSL_DIR/lib)
385 @subsubheading License
388 The complete source code of tinc is covered by the GNU GPL version 2.
389 Since the license under which OpenSSL is distributed is not directly
390 compatible with the terms of the GNU GPL
391 @uref{https://www.openssl.org/support/faq.html#LEGAL2}, we
392 include an exemption to the GPL (see also the file COPYING.README) to allow
393 everyone to create a statically or dynamically linked executable:
396 This program is released under the GPL with the additional exemption
397 that compiling, linking, and/or using OpenSSL is allowed. You may
398 provide binary packages linked to the OpenSSL libraries, provided that
399 all other requirements of the GPL are met.
402 Since the LZO library used by tinc is also covered by the GPL,
403 we also present the following exemption:
406 Hereby I grant a special exception to the tinc VPN project
407 (https://www.tinc-vpn.org/) to link the LZO library with the OpenSSL library
408 (https://www.openssl.org).
410 Markus F.X.J. Oberhumer
414 @c ==================================================================
419 For the optional compression of UDP packets, tinc uses the functions provided
422 If this library is not installed, you will get an error when running the
423 configure script. You can either install the zlib library, or disable support
424 for zlib compression by using the "--disable-zlib" option when running the
425 configure script. Note that if you disable support for zlib, the resulting
426 binary will not work correctly on VPNs where zlib compression is used.
428 You can use your operating system's package manager to install this if
429 available. Make sure you install the development AND runtime versions
432 If you have to install zlib manually, you can get the source code
433 from @url{https://zlib.net/}. Instructions on how to configure,
434 build and install this package are included within the package. Please
435 make sure you build development and runtime libraries (which is the
439 @c ==================================================================
444 Another form of compression is offered using the LZO library.
446 If this library is not installed, you will get an error when running the
447 configure script. You can either install the LZO library, or disable support
448 for LZO compression by using the "--disable-lzo" option when running the
449 configure script. Note that if you disable support for LZO, the resulting
450 binary will not work correctly on VPNs where LZO compression is used.
452 You can use your operating system's package manager to install this if
453 available. Make sure you install the development AND runtime versions
456 If you have to install LZO manually, you can get the source code
457 from @url{https://www.oberhumer.com/opensource/lzo/}. Instructions on how to configure,
458 build and install this package are included within the package. Please
459 make sure you build development and runtime libraries (which is the
463 @c ==================================================================
465 @subsection libcurses
468 For the "tinc top" command, tinc requires a curses library.
470 If this library is not installed, you will get an error when running the
471 configure script. You can either install a suitable curses library, or disable
472 all functionality that depends on a curses library by using the
473 "--disable-curses" option when running the configure script.
475 There are several curses libraries. It is recommended that you install
476 "ncurses" (@url{https://invisible-island.net/ncurses/}),
477 however other curses libraries should also work.
478 In particular, "PDCurses" (@url{https://pdcurses.sourceforge.io/})
479 is recommended if you want to compile tinc for Windows.
481 You can use your operating system's package manager to install this if
482 available. Make sure you install the development AND runtime versions
486 @c ==================================================================
488 @subsection libreadline
491 For the "tinc" command's shell functionality, tinc uses the readline library.
493 If this library is not installed, you will get an error when running the
494 configure script. You can either install a suitable readline library, or
495 disable all functionality that depends on a readline library by using the
496 "--disable-readline" option when running the configure script.
498 You can use your operating system's package manager to install this if
499 available. Make sure you install the development AND runtime versions
502 If you have to install libreadline manually, you can get the source code from
503 @url{https://www.gnu.org/software/readline/}. Instructions on how to configure,
504 build and install this package are included within the package. Please make
505 sure you build development and runtime libraries (which is the default).
517 @c ==================================================================
519 @chapter Installation
521 If you use Debian, you may want to install one of the
522 precompiled packages for your system. These packages are equipped with
523 system startup scripts and sample configurations.
525 If you cannot use one of the precompiled packages, or you want to compile tinc
526 for yourself, you can use the source. The source is distributed under
527 the GNU General Public License (GPL). Download the source from the
528 @uref{https://www.tinc-vpn.org/download/, download page}.
530 Tinc comes in a convenient autoconf/automake package, which you can just
531 treat the same as any other package. Which is just untar it, type
532 `./configure' and then `make'.
533 More detailed instructions are in the file @file{INSTALL}, which is
534 included in the source distribution.
537 * Building and installing tinc::
542 @c ==================================================================
543 @node Building and installing tinc
544 @section Building and installing tinc
546 Detailed instructions on configuring the source, building tinc and installing tinc
547 can be found in the file called @file{INSTALL}.
549 @cindex binary package
550 If you happen to have a binary package for tinc for your distribution,
551 you can use the package management tools of that distribution to install tinc.
552 The documentation that comes along with your distribution will tell you how to do that.
555 * Darwin (MacOS/X) build environment::
556 * Cygwin (Windows) build environment::
557 * MinGW (Windows) build environment::
561 @c ==================================================================
562 @node Darwin (MacOS/X) build environment
563 @subsection Darwin (MacOS/X) build environment
565 In order to build tinc on Darwin, you need to install Xcode from @uref{https://developer.apple.com/xcode/}.
566 It might also help to install a recent version of Fink from @uref{http://www.finkproject.org/}.
568 You need to download and install LibreSSL (or OpenSSL) and LZO,
569 either directly from their websites (see @ref{Libraries}) or using Fink.
571 @c ==================================================================
572 @node Cygwin (Windows) build environment
573 @subsection Cygwin (Windows) build environment
575 If Cygwin hasn't already been installed, install it directly from
576 @uref{https://www.cygwin.com/}.
578 When tinc is compiled in a Cygwin environment, it can only be run in this environment,
579 but all programs, including those started outside the Cygwin environment, will be able to use the VPN.
580 It will also support all features.
582 @c ==================================================================
583 @node MinGW (Windows) build environment
584 @subsection MinGW (Windows) build environment
586 You will need to install the MinGW environment from @uref{http://www.mingw.org}.
587 You also need to download and install LibreSSL (or OpenSSL) and LZO.
589 When tinc is compiled using MinGW it runs natively under Windows,
590 it is not necessary to keep MinGW installed.
592 When detaching, tinc will install itself as a service,
593 which will be restarted automatically after reboots.
596 @c ==================================================================
598 @section System files
600 Before you can run tinc, you must make sure you have all the needed
601 files on your system.
609 @c ==================================================================
611 @subsection Device files
614 Most operating systems nowadays come with the necessary device files by default,
615 or they have a mechanism to create them on demand.
617 If you use Linux and do not have udev installed,
618 you may need to create the following device file if it does not exist:
621 mknod -m 600 /dev/net/tun c 10 200
625 @c ==================================================================
627 @subsection Other files
629 @subsubheading @file{/etc/networks}
631 You may add a line to @file{/etc/networks} so that your VPN will get a
632 symbolic name. For example:
638 @subsubheading @file{/etc/services}
641 You may add this line to @file{/etc/services}. The effect is that you
642 may supply a @samp{tinc} as a valid port number to some programs. The
643 number 655 is registered with the IANA.
648 # Ivo Timmermans <ivo@@tinc-vpn.org>
663 @c ==================================================================
665 @chapter Configuration
668 * Configuration introduction::
669 * Multiple networks::
670 * How connections work::
671 * Configuration files::
672 * Network interfaces::
673 * Example configuration::
676 @c ==================================================================
677 @node Configuration introduction
678 @section Configuration introduction
680 Before actually starting to configure tinc and editing files,
681 make sure you have read this entire section so you know what to expect.
682 Then, make it clear to yourself how you want to organize your VPN:
683 What are the nodes (computers running tinc)?
684 What IP addresses/subnets do they have?
685 What is the network mask of the entire VPN?
686 Do you need special firewall rules?
687 Do you have to set up masquerading or forwarding rules?
688 Do you want to run tinc in router mode or switch mode?
689 These questions can only be answered by yourself,
690 you will not find the answers in this documentation.
691 Make sure you have an adequate understanding of networks in general.
692 @cindex Network Administrators Guide
693 A good resource on networking is the
694 @uref{https://www.tldp.org/LDP/nag2/, Linux Network Administrators Guide}.
696 If you have everything clearly pictured in your mind,
697 proceed in the following order:
698 First, create the initial configuration files and public/private keypairs using the following command:
700 tinc -n @var{NETNAME} init @var{NAME}
702 Second, use @samp{tinc -n @var{NETNAME} add ...} to further configure tinc.
703 Finally, export your host configuration file using @samp{tinc -n @var{NETNAME} export} and send it to those
704 people or computers you want tinc to connect to.
705 They should send you their host configuration file back, which you can import using @samp{tinc -n @var{NETNAME} import}.
707 These steps are described in the subsections below.
710 @c ==================================================================
711 @node Multiple networks
712 @section Multiple networks
714 @cindex multiple networks
717 In order to allow you to run more than one tinc daemon on one computer,
718 for instance if your computer is part of more than one VPN,
719 you can assign a @var{netname} to your VPN.
720 It is not required if you only run one tinc daemon,
721 it doesn't even have to be the same on all the nodes of your VPN,
722 but it is recommended that you choose one anyway.
724 We will assume you use a netname throughout this document.
725 This means that you call tinc with the -n argument,
726 which will specify the netname.
728 The effect of this option is that tinc will set its configuration
729 root to @file{@value{sysconfdir}/tinc/@var{netname}/}, where @var{netname} is your argument to the -n option.
730 You will also notice that log messages it appears in syslog as coming from @file{tinc.@var{netname}},
731 and on Linux, unless specified otherwise, the name of the virtual network interface will be the same as the network name.
733 However, it is not strictly necessary that you call tinc with the -n
734 option. If you do not use it, the network name will just be empty, and
735 tinc will look for files in @file{@value{sysconfdir}/tinc/} instead of
736 @file{@value{sysconfdir}/tinc/@var{netname}/};
737 the configuration file will then be @file{@value{sysconfdir}/tinc/tinc.conf},
738 and the host configuration files are expected to be in @file{@value{sysconfdir}/tinc/hosts/}.
741 @c ==================================================================
742 @node How connections work
743 @section How connections work
745 When tinc starts up, it parses the command-line options and then
746 reads in the configuration file tinc.conf.
747 It will then start listening for incoming connection from other daemons,
748 and will by default also automatically try to connect to known peers.
749 By default, tinc will try to keep at least 3 working meta-connections alive at all times.
753 There is no real distinction between a server and a client in tinc.
754 If you wish, you can view a tinc daemon without a `ConnectTo' statement in tinc.conf and `AutoConnect = no' as a server,
755 and one which does have one or more `ConnectTo' statements or `Autoconnect = yes' (which is the default) as a client.
756 It does not matter if two tinc daemons have a `ConnectTo' value pointing to each other however.
758 Connections specified using `ConnectTo' are so-called meta-connections.
759 Tinc daemons exchange information about all other daemon they know about via these meta-connections.
760 After learning about all the daemons in the VPN,
761 tinc will create other connections as necessary in order to communicate with them.
762 For example, if there are three daemons named A, B and C, and A has @samp{ConnectTo = B} in its tinc.conf file,
763 and C has @samp{ConnectTo = B} in its tinc.conf file, then A will learn about C from B,
764 and will be able to exchange VPN packets with C without the need to have @samp{ConnectTo = C} in its tinc.conf file.
766 It could be that some daemons are located behind a Network Address Translation (NAT) device, or behind a firewall.
767 In the above scenario with three daemons, if A and C are behind a NAT,
768 B will automatically help A and C punch holes through their NAT,
769 in a way similar to the STUN protocol, so that A and C can still communicate with each other directly.
770 It is not always possible to do this however, and firewalls might also prevent direct communication.
771 In that case, VPN packets between A and C will be forwarded by B.
773 In effect, all nodes in the VPN will be able to talk to each other, as long as
774 there is a path of meta-connections between them, and whenever possible, two
775 nodes will communicate with each other directly.
778 @c ==================================================================
779 @node Configuration files
780 @section Configuration files
782 The actual configuration of the daemon is done in the file
783 @file{@value{sysconfdir}/tinc/@var{netname}/tinc.conf} and at least one other file in the directory
784 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/}.
786 An optional directory @file{@value{sysconfdir}/tinc/@var{netname}/conf.d} can be added from which
787 any .conf file will be read.
789 These file consists of comments (lines started with a #) or assignments
796 The variable names are case insensitive, and any spaces, tabs, newlines
797 and carriage returns are ignored. Note: it is not required that you put
798 in the `=' sign, but doing so improves readability. If you leave it
799 out, remember to replace it with at least one space character.
801 The server configuration is complemented with host specific configuration (see
802 the next section). Although all host configuration options for the local node
803 listed in this document can also be put in
804 @file{@value{sysconfdir}/tinc/@var{netname}/tinc.conf}, it is recommended to
805 put host specific configuration options in the host configuration file, as this
806 makes it easy to exchange with other nodes.
808 You can edit the config file manually, but it is recommended that you use
809 the tinc command to change configuration variables for you.
811 In the following two subsections all valid variables are listed in alphabetical order.
812 The default value is given between parentheses,
813 other comments are between square brackets.
816 * Main configuration variables::
817 * Host configuration variables::
823 @c ==================================================================
824 @node Main configuration variables
825 @subsection Main configuration variables
828 @cindex AddressFamily
829 @item AddressFamily = <ipv4|ipv6|any> (any)
830 This option affects the address family of listening and outgoing sockets.
831 If any is selected, then depending on the operating system
832 both IPv4 and IPv6 or just IPv6 listening sockets will be created.
835 @item AutoConnect = <yes|no> (yes)
836 If set to yes, tinc will automatically set up meta connections to other nodes,
837 without requiring @var{ConnectTo} variables.
839 @cindex BindToAddress
840 @item BindToAddress = <@var{address}> [<@var{port}>]
841 This is the same as ListenAddress, however the address given with the BindToAddress option
842 will also be used for outgoing connections.
843 This is useful if your computer has more than one IPv4 or IPv6 address,
844 and you want tinc to only use a specific one for outgoing packets.
846 @cindex BindToInterface
847 @item BindToInterface = <@var{interface}> [experimental]
848 If you have more than one network interface in your computer, tinc will
849 by default listen on all of them for incoming connections. It is
850 possible to bind tinc to a single interface like eth0 or ppp0 with this
853 This option may not work on all platforms.
854 Also, on some platforms it will not actually bind to an interface,
855 but rather to the address that the interface has at the moment a socket is created.
858 @item Broadcast = <no | mst | direct> (mst) [experimental]
859 This option selects the way broadcast packets are sent to other daemons.
860 @emph{NOTE: all nodes in a VPN must use the same Broadcast mode, otherwise routing loops can form.}
864 Broadcast packets are never sent to other nodes.
867 Broadcast packets are sent and forwarded via the VPN's Minimum Spanning Tree.
868 This ensures broadcast packets reach all nodes.
871 Broadcast packets are sent directly to all nodes that can be reached directly.
872 Broadcast packets received from other nodes are never forwarded.
873 If the IndirectData option is also set, broadcast packets will only be sent to nodes which we have a meta connection to.
876 @cindex BroadcastSubnet
877 @item BroadcastSubnet = @var{address}[/@var{prefixlength}]
878 Declares a broadcast subnet.
879 Any packet with a destination address falling into such a subnet will be routed as a broadcast
880 (provided all nodes have it declared).
881 This is most useful to declare subnet broadcast addresses (e.g. 10.42.255.255),
882 otherwise tinc won't know what to do with them.
884 Note that global broadcast addresses (MAC ff:ff:ff:ff:ff:ff, IPv4 255.255.255.255),
885 as well as multicast space (IPv4 224.0.0.0/4, IPv6 ff00::/8)
886 are always considered broadcast addresses and don't need to be declared.
889 @item ConnectTo = <@var{name}>
890 Specifies which other tinc daemon to connect to on startup.
891 Multiple ConnectTo variables may be specified,
892 in which case outgoing connections to each specified tinc daemon are made.
893 The names should be known to this tinc daemon
894 (i.e., there should be a host configuration file for the name on the ConnectTo line).
896 If you don't specify a host with ConnectTo and have disabled AutoConnect,
897 tinc won't try to connect to other daemons at all,
898 and will instead just listen for incoming connections.
901 @item DecrementTTL = <yes | no> (no) [experimental]
902 When enabled, tinc will decrement the Time To Live field in IPv4 packets, or the Hop Limit field in IPv6 packets,
903 before forwarding a received packet to the virtual network device or to another node,
904 and will drop packets that have a TTL value of zero,
905 in which case it will send an ICMP Time Exceeded packet back.
907 Do not use this option if you use switch mode and want to use IPv6.
910 @item Device = <@var{device}> (@file{/dev/tap0}, @file{/dev/net/tun} or other depending on platform)
911 The virtual network device to use.
912 Tinc will automatically detect what kind of device it is.
913 Note that you can only use one device per daemon.
914 Under Windows, use @var{Interface} instead of @var{Device}.
915 Note that you can only use one device per daemon.
916 See also @ref{Device files}.
918 @cindex DeviceStandby
919 @item DeviceStandby = <yes | no> (no)
920 When disabled, tinc calls @file{tinc-up} on startup, and @file{tinc-down} on shutdown.
921 When enabled, tinc will only call @file{tinc-up} when at least one node is reachable,
922 and will call @file{tinc-down} as soon as no nodes are reachable.
923 On Windows, this also determines when the virtual network interface "cable" is "plugged".
926 @item DeviceType = <@var{type}> (platform dependent)
927 The type of the virtual network device.
928 Tinc will normally automatically select the right type of tun/tap interface, and this option should not be used.
929 However, this option can be used to select one of the special interface types, if support for them is compiled in.
934 Use a dummy interface.
935 No packets are ever read or written to a virtual network device.
936 Useful for testing, or when setting up a node that only forwards packets for other nodes.
940 Open a raw socket, and bind it to a pre-existing
941 @var{Interface} (eth0 by default).
942 All packets are read from this interface.
943 Packets received for the local node are written to the raw socket.
944 However, at least on Linux, the operating system does not process IP packets destined for the local host.
948 Open a multicast UDP socket and bind it to the address and port (separated by spaces) and optionally a TTL value specified using @var{Device}.
949 Packets are read from and written to this multicast socket.
950 This can be used to connect to UML, QEMU or KVM instances listening on the same multicast address.
951 Do NOT connect multiple tinc daemons to the same multicast address, this will very likely cause routing loops.
952 Also note that this can cause decrypted VPN packets to be sent out on a real network if misconfigured.
956 Use a file descriptor.
957 All packets are read from this interface.
958 Packets received for the local node are written to it.
961 @item uml (not compiled in by default)
962 Create a UNIX socket with the filename specified by
963 @var{Device}, or @file{@value{runstatedir}/@var{netname}.umlsocket}
965 Tinc will wait for a User Mode Linux instance to connect to this socket.
968 @item vde (not compiled in by default)
969 Uses the libvdeplug library to connect to a Virtual Distributed Ethernet switch,
970 using the UNIX socket specified by
971 @var{Device}, or @file{@value{runstatedir}/vde.ctl}
975 Also, in case tinc does not seem to correctly interpret packets received from the virtual network device,
976 it can be used to change the way packets are interpreted:
979 @item tun (BSD and Linux)
981 Depending on the platform, this can either be with or without an address family header (see below).
984 @item tunnohead (BSD)
985 Set type to tun without an address family header.
986 Tinc will expect packets read from the virtual network device to start with an IP header.
987 On some platforms IPv6 packets cannot be read from or written to the device in this mode.
990 @item tunifhead (BSD)
991 Set type to tun with an address family header.
992 Tinc will expect packets read from the virtual network device
993 to start with a four byte header containing the address family,
994 followed by an IP header.
995 This mode should support both IPv4 and IPv6 packets.
1000 This is only supported on OS X version 10.6.8 and higher, but doesn't require the tuntaposx module.
1001 This mode should support both IPv4 and IPv6 packets.
1003 @item tap (BSD and Linux)
1005 Tinc will expect packets read from the virtual network device
1006 to start with an Ethernet header.
1010 @item DirectOnly = <yes|no> (no) [experimental]
1011 When this option is enabled, packets that cannot be sent directly to the destination node,
1012 but which would have to be forwarded by an intermediate node, are dropped instead.
1013 When combined with the IndirectData option,
1014 packets for nodes for which we do not have a meta connection with are also dropped.
1016 @cindex Ed25519PrivateKeyFile
1017 @item Ed25519PrivateKeyFile = <@var{path}> (@file{@value{sysconfdir}/tinc/@var{netname}/ed25519_key.priv})
1018 The file in which the private Ed25519 key of this tinc daemon resides.
1019 This is only used if ExperimentalProtocol is enabled.
1021 @cindex ExperimentalProtocol
1022 @item ExperimentalProtocol = <yes|no> (yes)
1023 When this option is enabled, the SPTPS protocol will be used when connecting to nodes that also support it.
1024 Ephemeral ECDH will be used for key exchanges,
1025 and Ed25519 will be used instead of RSA for authentication.
1026 When enabled, an Ed25519 key must have been generated before with
1027 @samp{tinc generate-ed25519-keys}.
1030 @item Forwarding = <off|internal|kernel> (internal) [experimental]
1031 This option selects the way indirect packets are forwarded.
1035 Incoming packets that are not meant for the local node,
1036 but which should be forwarded to another node, are dropped.
1039 Incoming packets that are meant for another node are forwarded by tinc internally.
1041 This is the default mode, and unless you really know you need another forwarding mode, don't change it.
1044 Incoming packets using the legacy protocol are always sent to the TUN/TAP device,
1045 even if the packets are not for the local node.
1046 This is less efficient, but allows the kernel to apply its routing and firewall rules on them,
1047 and can also help debugging.
1048 Incoming packets using the SPTPS protocol are dropped, since they are end-to-end encrypted.
1052 @item FWMark = <@var{value}> (0) [experimental]
1053 When set to a non-zero value, all TCP and UDP sockets created by tinc will use the given value as the firewall mark.
1054 This can be used for mark-based routing or for packet filtering.
1055 This option is currently only supported on Linux.
1058 @item Hostnames = <yes|no> (no)
1059 This option selects whether IP addresses (both real and on the VPN)
1060 should be resolved. Since DNS lookups are blocking, it might affect
1061 tinc's efficiency, even stopping the daemon for a few seconds every time
1062 it does a lookup if your DNS server is not responding.
1064 This does not affect resolving hostnames to IP addresses from the
1065 configuration file, but whether hostnames should be resolved while logging.
1068 @item Interface = <@var{interface}>
1069 Defines the name of the interface corresponding to the virtual network device.
1070 Depending on the operating system and the type of device this may or may not actually set the name of the interface.
1071 Under Windows, this variable is used to select which network interface will be used.
1072 If you specified a Device, this variable is almost always already correctly set.
1074 @cindex ListenAddress
1075 @item ListenAddress = <@var{address}> [<@var{port}>]
1076 If your computer has more than one IPv4 or IPv6 address, tinc
1077 will by default listen on all of them for incoming connections.
1078 This option can be used to restrict which addresses tinc listens on.
1079 Multiple ListenAddress variables may be specified,
1080 in which case listening sockets for each specified address are made.
1082 If no @var{port} is specified, the socket will listen on the port specified by the Port option,
1083 or to port 655 if neither is given.
1084 To only listen on a specific port but not to a specific address, use "*" for the @var{address}.
1086 @cindex LocalDiscovery
1087 @item LocalDiscovery = <yes | no> (no)
1088 When enabled, tinc will try to detect peers that are on the same local network.
1089 This will allow direct communication using LAN addresses, even if both peers are behind a NAT
1090 and they only ConnectTo a third node outside the NAT,
1091 which normally would prevent the peers from learning each other's LAN address.
1093 Currently, local discovery is implemented by sending some packets to the local address of the node during UDP discovery.
1094 This will not work with old nodes that don't transmit their local address.
1097 @item LogLevel = <@var{level}> (0)
1098 This option controls the verbosity of the logging.
1099 See @ref{Debug levels}.
1102 @item Mode = <router|switch|hub> (router)
1103 This option selects the way packets are routed to other daemons.
1109 variables in the host configuration files will be used to form a routing table.
1110 Only packets of routable protocols (IPv4 and IPv6) are supported in this mode.
1112 This is the default mode, and unless you really know you need another mode, don't change it.
1116 In this mode the MAC addresses of the packets on the VPN will be used to
1117 dynamically create a routing table just like an Ethernet switch does.
1118 Unicast, multicast and broadcast packets of every protocol that runs over Ethernet are supported in this mode
1119 at the cost of frequent broadcast ARP requests and routing table updates.
1121 This mode is primarily useful if you want to bridge Ethernet segments.
1125 This mode is almost the same as the switch mode, but instead
1126 every packet will be broadcast to the other daemons
1127 while no routing table is managed.
1130 @cindex InvitationExpire
1131 @item InvitationExpire = <@var{seconds}> (604800)
1132 This option controls the time invitations are valid.
1135 @item KeyExpire = <@var{seconds}> (3600)
1136 This option controls the time the encryption keys used to encrypt the data
1137 are valid. It is common practice to change keys at regular intervals to
1138 make it even harder for crackers, even though it is thought to be nearly
1139 impossible to crack a single key.
1142 @item MACExpire = <@var{seconds}> (600)
1143 This option controls the amount of time MAC addresses are kept before they are removed.
1144 This only has effect when Mode is set to "switch".
1146 @cindex MaxConnectionBurst
1147 @item MaxConnectionBurst = <@var{count}> (100)
1148 This option controls how many connections tinc accepts in quick succession.
1149 If there are more connections than the given number in a short time interval,
1150 tinc will reduce the number of accepted connections to only one per second,
1151 until the burst has passed.
1154 @item Name = <@var{name}> [required]
1155 This is a symbolic name for this connection.
1156 The name must consist only of alfanumeric and underscore characters (a-z, A-Z, 0-9 and _), and is case sensitive.
1158 If Name starts with a $, then the contents of the environment variable that follows will be used.
1159 In that case, invalid characters will be converted to underscores.
1160 If Name is $HOST, but no such environment variable exist,
1161 the hostname will be read using the gethostname() system call.
1163 @cindex PingInterval
1164 @item PingInterval = <@var{seconds}> (60)
1165 The number of seconds of inactivity that tinc will wait before sending a
1166 probe to the other end.
1169 @item PingTimeout = <@var{seconds}> (5)
1170 The number of seconds to wait for a response to pings or to allow meta
1171 connections to block. If the other end doesn't respond within this time,
1172 the connection is terminated, and the others will be notified of this.
1174 @cindex PriorityInheritance
1175 @item PriorityInheritance = <yes|no> (no) [experimental]
1176 When this option is enabled the value of the TOS field of tunneled IPv4 packets
1177 will be inherited by the UDP packets that are sent out.
1180 @item PrivateKey = <@var{key}> [obsolete]
1181 This is the RSA private key for tinc. However, for safety reasons it is
1182 advised to store private keys of any kind in separate files. This prevents
1183 accidental eavesdropping if you are editing the configuration file.
1185 @cindex PrivateKeyFile
1186 @item PrivateKeyFile = <@var{path}> (@file{@value{sysconfdir}/tinc/@var{netname}/rsa_key.priv})
1187 This is the full path name of the RSA private key file that was
1188 generated by @samp{tinc generate-keys}. It must be a full path, not a
1191 @cindex ProcessPriority
1192 @item ProcessPriority = <low|normal|high>
1193 When this option is used the priority of the tincd process will be adjusted.
1194 Increasing the priority may help to reduce latency and packet loss on the VPN.
1197 @item Proxy = socks4 | socks5 | http | exec @var{...} [experimental]
1198 Use a proxy when making outgoing connections.
1199 The following proxy types are currently supported:
1203 @item socks4 <@var{address}> <@var{port}> [<@var{username}>]
1204 Connects to the proxy using the SOCKS version 4 protocol.
1205 Optionally, a @var{username} can be supplied which will be passed on to the proxy server.
1208 @item socks5 <@var{address}> <@var{port}> [<@var{username}> <@var{password}>]
1209 Connect to the proxy using the SOCKS version 5 protocol.
1210 If a @var{username} and @var{password} are given, basic username/password authentication will be used,
1211 otherwise no authentication will be used.
1214 @item http <@var{address}> <@var{port}>
1215 Connects to the proxy and sends a HTTP CONNECT request.
1218 @item exec <@var{command}>
1219 Executes the given command which should set up the outgoing connection.
1220 The environment variables @env{NAME}, @env{NODE}, @env{REMOTEADDRES} and @env{REMOTEPORT} are available.
1223 @cindex ReplayWindow
1224 @item ReplayWindow = <bytes> (32)
1225 This is the size of the replay tracking window for each remote node, in bytes.
1226 The window is a bitfield which tracks 1 packet per bit, so for example
1227 the default setting of 32 will track up to 256 packets in the window. In high
1228 bandwidth scenarios, setting this to a higher value can reduce packet loss from
1229 the interaction of replay tracking with underlying real packet loss and/or
1230 reordering. Setting this to zero will disable replay tracking completely and
1231 pass all traffic, but leaves tinc vulnerable to replay-based attacks on your
1234 @cindex StrictSubnets
1235 @item StrictSubnets = <yes|no> (no) [experimental]
1236 When this option is enabled tinc will only use Subnet statements which are
1237 present in the host config files in the local
1238 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/} directory.
1239 Subnets learned via connections to other nodes and which are not
1240 present in the local host config files are ignored.
1242 @cindex TunnelServer
1243 @item TunnelServer = <yes|no> (no) [experimental]
1244 When this option is enabled tinc will no longer forward information between other tinc daemons,
1245 and will only allow connections with nodes for which host config files are present in the local
1246 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/} directory.
1247 Setting this options also implicitly sets StrictSubnets.
1250 @item UDPDiscovery = <yes|no> (yes)
1251 When this option is enabled tinc will try to establish UDP connectivity to nodes,
1252 using TCP while it determines if a node is reachable over UDP. If it is disabled,
1253 tinc always assumes a node is reachable over UDP.
1254 Note that tinc will never use UDP with nodes that have TCPOnly enabled.
1256 @cindex UDPDiscoveryKeepaliveInterval
1257 @item UDPDiscoveryKeepaliveInterval = <seconds> (9)
1258 The minimum amount of time between sending UDP ping datagrams to check UDP connectivity once it has been established.
1259 Note that these pings are large, since they are used to verify link MTU as well.
1261 @cindex UDPDiscoveryInterval
1262 @item UDPDiscoveryInterval = <seconds> (2)
1263 The minimum amount of time between sending UDP ping datagrams to try to establish UDP connectivity.
1265 @cindex UDPDiscoveryTimeout
1266 @item UDPDiscoveryTimeout = <seconds> (30)
1267 If tinc doesn't receive any UDP ping replies over the specified interval,
1268 it will assume UDP communication is broken and will fall back to TCP.
1270 @cindex UDPInfoInterval
1271 @item UDPInfoInterval = <seconds> (5)
1272 The minimum amount of time between sending periodic updates about UDP addresses, which are mostly useful for UDP hole punching.
1275 @item UDPRcvBuf = <bytes> (1048576)
1276 Sets the socket receive buffer size for the UDP socket, in bytes.
1277 If set to zero, the default buffer size will be used by the operating system.
1278 Note: this setting can have a significant impact on performance, especially raw throughput.
1281 @item UDPSndBuf = <bytes> (1048576)
1282 Sets the socket send buffer size for the UDP socket, in bytes.
1283 If set to zero, the default buffer size will be used by the operating system.
1284 Note: this setting can have a significant impact on performance, especially raw throughput.
1287 @item UPnP = <yes|udponly|no> (no)
1288 If this option is enabled then tinc will search for UPnP-IGD devices on the local network.
1289 It will then create and maintain port mappings for tinc's listening TCP and UDP ports.
1290 If set to "udponly", tinc will only create a mapping for its UDP (data) port, not for its TCP (metaconnection) port.
1291 Note that tinc must have been built with miniupnpc support for this feature to be available.
1292 Furthermore, be advised that enabling this can have security implications, because the miniupnpc library that
1293 tinc uses might not be well-hardened with regard to malicious UPnP replies.
1295 @cindex UPnPDiscoverWait
1296 @item UPnPDiscoverWait = <seconds> (5)
1297 The amount of time to wait for replies when probing the local network for UPnP devices.
1299 @cindex UPnPRefreshPeriod
1300 @item UPnPRefreshPeriod = <seconds> (5)
1301 How often tinc will re-add the port mapping, in case it gets reset on the UPnP device.
1302 This also controls the duration of the port mapping itself, which will be set to twice that duration.
1307 @c ==================================================================
1308 @node Host configuration variables
1309 @subsection Host configuration variables
1313 @item Address = <@var{IP address}|@var{hostname}> [<port>] [recommended]
1314 This variable is only required if you want to connect to this host. It
1315 must resolve to the external IP address where the host can be reached,
1316 not the one that is internal to the VPN.
1317 If no port is specified, the default Port is used.
1318 Multiple Address variables can be specified, in which case each address will be
1319 tried until a working connection has been established.
1322 @item Cipher = <@var{cipher}> (blowfish)
1323 The symmetric cipher algorithm used to encrypt UDP packets using the legacy protocol.
1324 Any cipher supported by LibreSSL or OpenSSL is recognized.
1325 Furthermore, specifying "none" will turn off packet encryption.
1326 It is best to use only those ciphers which support CBC mode.
1327 This option has no effect for connections using the SPTPS protocol, which always use AES-256-CTR.
1330 @item ClampMSS = <yes|no> (yes)
1331 This option specifies whether tinc should clamp the maximum segment size (MSS)
1332 of TCP packets to the path MTU. This helps in situations where ICMP
1333 Fragmentation Needed or Packet too Big messages are dropped by firewalls.
1336 @item Compression = <@var{level}> (0)
1337 This option sets the level of compression used for UDP packets.
1338 Possible values are 0 (off), 1 (fast zlib) and any integer up to 9 (best zlib),
1339 10 (fast LZO) and 11 (best LZO).
1342 @item Digest = <@var{digest}> (sha1)
1343 The digest algorithm used to authenticate UDP packets using the legacy protocol.
1344 Any digest supported by LibreSSL or OpenSSL is recognized.
1345 Furthermore, specifying "none" will turn off packet authentication.
1346 This option has no effect for connections using the SPTPS protocol, which always use HMAC-SHA-256.
1348 @cindex IndirectData
1349 @item IndirectData = <yes|no> (no)
1350 When set to yes, other nodes which do not already have a meta connection to you
1351 will not try to establish direct communication with you.
1352 It is best to leave this option out or set it to no.
1355 @item MACLength = <@var{bytes}> (4)
1356 The length of the message authentication code used to authenticate UDP packets using the legacy protocol.
1357 Can be anything from 0
1358 up to the length of the digest produced by the digest algorithm.
1359 This option has no effect for connections using the SPTPS protocol, which never truncate MACs.
1362 @item PMTU = <@var{mtu}> (1514)
1363 This option controls the initial path MTU to this node.
1365 @cindex PMTUDiscovery
1366 @item PMTUDiscovery = <yes|no> (yes)
1367 When this option is enabled, tinc will try to discover the path MTU to this node.
1368 After the path MTU has been discovered, it will be enforced on the VPN.
1370 @cindex MTUInfoInterval
1371 @item MTUInfoInterval = <seconds> (5)
1372 The minimum amount of time between sending periodic updates about relay path MTU. Useful for quickly determining MTU to indirect nodes.
1375 @item Port = <@var{port}> (655)
1376 This is the port this tinc daemon listens on.
1377 You can use decimal portnumbers or symbolic names (as listed in @file{/etc/services}).
1380 @item PublicKey = <@var{key}> [obsolete]
1381 This is the RSA public key for this host.
1383 @cindex PublicKeyFile
1384 @item PublicKeyFile = <@var{path}> [obsolete]
1385 This is the full path name of the RSA public key file that was generated
1386 by @samp{tinc generate-keys}. It must be a full path, not a relative
1390 From version 1.0pre4 on tinc will store the public key directly into the
1391 host configuration file in PEM format, the above two options then are not
1392 necessary. Either the PEM format is used, or exactly
1393 @strong{one of the above two options} must be specified
1394 in each host configuration file, if you want to be able to establish a
1395 connection with that host.
1398 @item Subnet = <@var{address}[/@var{prefixlength}[#@var{weight}]]>
1399 The subnet which this tinc daemon will serve.
1400 Tinc tries to look up which other daemon it should send a packet to by searching the appropriate subnet.
1401 If the packet matches a subnet,
1402 it will be sent to the daemon who has this subnet in his host configuration file.
1403 Multiple subnet lines can be specified for each daemon.
1405 Subnets can either be single MAC, IPv4 or IPv6 addresses,
1406 in which case a subnet consisting of only that single address is assumed,
1407 or they can be a IPv4 or IPv6 network address with a prefixlength.
1408 For example, IPv4 subnets must be in a form like 192.168.1.0/24,
1409 where 192.168.1.0 is the network address and 24 is the number of bits set in the netmask.
1410 Note that subnets like 192.168.1.1/24 are invalid!
1411 Read a networking HOWTO/FAQ/guide if you don't understand this.
1412 IPv6 subnets are notated like fec0:0:0:1::/64.
1413 MAC addresses are notated like 0:1a:2b:3c:4d:5e.
1415 @cindex CIDR notation
1416 Prefixlength is the number of bits set to 1 in the netmask part; for
1417 example: netmask 255.255.255.0 would become /24, 255.255.252.0 becomes
1418 /22. This conforms to standard CIDR notation as described in
1419 @uref{https://www.ietf.org/rfc/rfc1519.txt, RFC1519}
1421 A Subnet can be given a weight to indicate its priority over identical Subnets
1422 owned by different nodes. The default weight is 10. Lower values indicate
1423 higher priority. Packets will be sent to the node with the highest priority,
1424 unless that node is not reachable, in which case the node with the next highest
1425 priority will be tried, and so on.
1428 @item TCPonly = <yes|no> (no)
1429 If this variable is set to yes, then the packets are tunnelled over a
1430 TCP connection instead of a UDP connection. This is especially useful
1431 for those who want to run a tinc daemon from behind a masquerading
1432 firewall, or if UDP packet routing is disabled somehow.
1433 Setting this options also implicitly sets IndirectData.
1436 @item Weight = <weight>
1437 If this variable is set, it overrides the weight given to connections made with
1438 another host. A higher weight means a lower priority is given to this
1439 connection when broadcasting or forwarding packets.
1443 @c ==================================================================
1448 Apart from reading the server and host configuration files,
1449 tinc can also run scripts at certain moments.
1450 Below is a list of filenames of scripts and a description of when they are run.
1451 A script is only run if it exists and if it is executable.
1453 Scripts are run synchronously;
1454 this means that tinc will temporarily stop processing packets until the called script finishes executing.
1455 This guarantees that scripts will execute in the exact same order as the events that trigger them.
1456 If you need to run commands asynchronously, you have to ensure yourself that they are being run in the background.
1458 Under Windows (not Cygwin), the scripts should have the extension @file{.bat} or @file{.cmd}.
1462 @item @value{sysconfdir}/tinc/@var{netname}/tinc-up
1463 This is the most important script.
1464 If it is present it will be executed right after the tinc daemon has been
1465 started and has connected to the virtual network device.
1466 It should be used to set up the corresponding network interface,
1467 but can also be used to start other things.
1469 Under Windows you can use the Network Connections control panel instead of creating this script.
1472 @item @value{sysconfdir}/tinc/@var{netname}/tinc-down
1473 This script is started right before the tinc daemon quits.
1475 @item @value{sysconfdir}/tinc/@var{netname}/hosts/@var{host}-up
1476 This script is started when the tinc daemon with name @var{host} becomes reachable.
1478 @item @value{sysconfdir}/tinc/@var{netname}/hosts/@var{host}-down
1479 This script is started when the tinc daemon with name @var{host} becomes unreachable.
1481 @item @value{sysconfdir}/tinc/@var{netname}/host-up
1482 This script is started when any host becomes reachable.
1484 @item @value{sysconfdir}/tinc/@var{netname}/host-down
1485 This script is started when any host becomes unreachable.
1487 @item @value{sysconfdir}/tinc/@var{netname}/subnet-up
1488 This script is started when a Subnet becomes reachable.
1489 The Subnet and the node it belongs to are passed in environment variables.
1491 @item @value{sysconfdir}/tinc/@var{netname}/subnet-down
1492 This script is started when a Subnet becomes unreachable.
1494 @item @value{sysconfdir}/tinc/@var{netname}/invitation-created
1495 This script is started when a new invitation has been created.
1497 @item @value{sysconfdir}/tinc/@var{netname}/invitation-accepted
1498 This script is started when an invitation has been used.
1502 @cindex environment variables
1503 The scripts are started without command line arguments,
1504 but can make use of certain environment variables.
1505 Under UNIX like operating systems the names of environment variables must be preceded by a $ in scripts.
1506 Under Windows, in @file{.bat} or @file{.cmd} files, they have to be put between % signs.
1511 If a netname was specified, this environment variable contains it.
1515 Contains the name of this tinc daemon.
1519 Contains the name of the virtual network device that tinc uses.
1523 Contains the name of the virtual network interface that tinc uses.
1524 This should be used for commands like ifconfig.
1528 When a host becomes (un)reachable, this is set to its name.
1529 If a subnet becomes (un)reachable, this is set to the owner of that subnet.
1531 @cindex REMOTEADDRESS
1533 When a host becomes (un)reachable, this is set to its real address.
1537 When a host becomes (un)reachable,
1538 this is set to the port number it uses for communication with other tinc daemons.
1542 When a subnet becomes (un)reachable, this is set to the subnet.
1546 When a subnet becomes (un)reachable, this is set to the subnet weight.
1548 @cindex INVITATION_FILE
1549 @item INVITATION_FILE
1550 When the @file{invitation-created} script is called,
1551 this is set to the file where the invitation details will be stored.
1553 @cindex INVITATION_URL
1554 @item INVITATION_URL
1555 When the @file{invitation-created} script is called,
1556 this is set to the invitation URL that has been created.
1559 Do not forget that under UNIX operating systems,
1560 you have to make the scripts executable, using the command @samp{chmod a+x script}.
1563 @c ==================================================================
1564 @node How to configure
1565 @subsection How to configure
1567 @subsubheading Step 1. Creating initial configuration files.
1569 The initial directory structure, configuration files and public/private keypairs are created using the following command:
1572 tinc -n @var{netname} init @var{name}
1575 (You will need to run this as root, or use "sudo".)
1576 This will create the configuration directory @file{@value{sysconfdir}/tinc/@var{netname}.},
1577 and inside it will create another directory named @file{hosts/}.
1578 In the configuration directory, it will create the file @file{tinc.conf} with the following contents:
1584 It will also create private RSA and Ed25519 keys, which will be stored in the files @file{rsa_key.priv} and @file{ed25519_key.priv}.
1585 It will also create a host configuration file @file{hosts/@var{name}},
1586 which will contain the corresponding public RSA and Ed25519 keys.
1588 Finally, on UNIX operating systems, it will create an executable script @file{tinc-up},
1589 which will initially not do anything except warning that you should edit it.
1591 @subsubheading Step 2. Modifying the initial configuration.
1593 Unless you want to use tinc in switch mode,
1594 you should now configure which range of addresses you will use on the VPN.
1595 Let's assume you will be part of a VPN which uses the address range 192.168.0.0/16,
1596 and you yourself have a smaller portion of that range: 192.168.2.0/24.
1597 Then you should run the following command:
1600 tinc -n @var{netname} add subnet 192.168.2.0/24
1603 This will add a Subnet statement to your host configuration file.
1604 Try opening the file @file{@value{sysconfdir}/tinc/@var{netname}/hosts/@var{name}} in an editor.
1605 You should now see a file containing the public RSA and Ed25519 keys (which looks like a bunch of random characters),
1606 and the following line at the bottom:
1609 Subnet = 192.168.2.0/24
1612 If you will use more than one address range, you can add more Subnets.
1613 For example, if you also use the IPv6 subnet fec0:0:0:2::/64, you can add it as well:
1616 tinc -n @var{netname} add subnet fec0:0:0:2::/24
1619 This will add another line to the file @file{hosts/@var{name}}.
1620 If you make a mistake, you can undo it by simply using @samp{del} instead of @samp{add}.
1622 If you want other tinc daemons to create meta-connections to your daemon,
1623 you should add your public IP address or hostname to your host configuration file.
1624 For example, if your hostname is foo.example.org, run:
1627 tinc -n @var{netname} add address foo.example.org
1630 @subsubheading Step 2. Exchanging configuration files.
1632 In order for two tinc daemons to be able to connect to each other,
1633 they each need the other's host configuration files.
1634 So if you want foo to be able to connect with bar,
1635 You should send @file{hosts/@var{name}} to bar, and bar should send you his file which you should move to @file{hosts/bar}.
1636 If you are on a UNIX platform, you can easily send an email containing the necessary information using the following command
1637 (assuming the owner of bar has the email address bar@@example.org):
1640 tinc -n @var{netname} export | mail -s "My config file" bar@@example.org
1643 If the owner of bar does the same to send his host configuration file to you,
1644 you can probably pipe his email through the following command,
1645 or you can just start this command in a terminal and copy&paste the email:
1648 tinc -n @var{netname} import
1651 If you are the owner of bar yourself, and you have SSH access to that computer,
1652 you can also swap the host configuration files using the following command:
1655 tinc -n @var{netname} export \
1656 | ssh bar.example.org tinc -n @var{netname} exchange \
1657 | tinc -n @var{netname} import
1660 You can repeat this for a few other nodes as well.
1661 It is not necessary to manually exchange host config files between all nodes;
1662 after the initial connections are made tinc will learn about all the other nodes in the VPN,
1663 and will automatically make other connections as necessary.
1666 @c ==================================================================
1667 @node Network interfaces
1668 @section Network interfaces
1670 Before tinc can start transmitting data over the tunnel, it must
1671 set up the virtual network interface.
1673 First, decide which IP addresses you want to have associated with these
1674 devices, and what network mask they must have.
1676 Tinc will open a virtual network device (@file{/dev/tun}, @file{/dev/tap0} or similar),
1677 which will also create a network interface called something like @samp{tun0}, @samp{tap0}.
1678 If you are using the Linux tun/tap driver, the network interface will by default have the same name as the @var{netname}.
1679 Under Windows you can change the name of the network interface from the Network Connections control panel.
1682 You can configure the network interface by putting ordinary ifconfig, route, and other commands
1683 to a script named @file{@value{sysconfdir}/tinc/@var{netname}/tinc-up}.
1684 When tinc starts, this script will be executed. When tinc exits, it will execute the script named
1685 @file{@value{sysconfdir}/tinc/@var{netname}/tinc-down}, but normally you don't need to create that script.
1686 You can manually open the script in an editor, or use the following command:
1689 tinc -n @var{netname} edit tinc-up
1692 An example @file{tinc-up} script, that would be appropriate for the scenario in the previous section, is:
1696 ifconfig $INTERFACE 192.168.2.1 netmask 255.255.0.0
1697 ip addr add fec0:0:0:2::/48 dev $INTERFACE
1700 The first command gives the interface an IPv4 address and a netmask.
1701 The kernel will also automatically add an IPv4 route to this interface, so normally you don't need
1702 to add route commands to the @file{tinc-up} script.
1703 The kernel will also bring the interface up after this command.
1705 The netmask is the mask of the @emph{entire} VPN network, not just your
1707 The second command gives the interface an IPv6 address and netmask,
1708 which will also automatically add an IPv6 route.
1709 If you only want to use "ip addr" commands on Linux, don't forget that it doesn't bring the interface up, unlike ifconfig,
1710 so you need to add @samp{ip link set $INTERFACE up} in that case.
1712 The exact syntax of the ifconfig and route commands differs from platform to platform.
1713 You can look up the commands for setting addresses and adding routes in @ref{Platform specific information},
1714 but it is best to consult the manpages of those utilities on your platform.
1717 @c ==================================================================
1718 @node Example configuration
1719 @section Example configuration
1723 Imagine the following situation. Branch A of our example `company' wants to connect
1724 three branch offices in B, C and D using the Internet. All four offices
1725 have a 24/7 connection to the Internet.
1727 A is going to serve as the center of the network. B and C will connect
1728 to A, and D will connect to C. Each office will be assigned their own IP
1732 A: net 10.1.0.0 mask 255.255.0.0 gateway 10.1.54.1 internet IP 1.2.3.4
1733 B: net 10.2.0.0 mask 255.255.0.0 gateway 10.2.1.12 internet IP 2.3.4.5
1734 C: net 10.3.0.0 mask 255.255.0.0 gateway 10.3.69.254 internet IP 3.4.5.6
1735 D: net 10.4.0.0 mask 255.255.0.0 gateway 10.4.3.32 internet IP 4.5.6.7
1738 Here, ``gateway'' is the VPN IP address of the machine that is running the
1739 tincd, and ``internet IP'' is the IP address of the firewall, which does not
1740 need to run tincd, but it must do a port forwarding of TCP and UDP on port
1741 655 (unless otherwise configured).
1743 In this example, it is assumed that eth0 is the interface that points to
1744 the inner (physical) LAN of the office, although this could also be the
1745 same as the interface that leads to the Internet. The configuration of
1746 the real interface is also shown as a comment, to give you an idea of
1747 how these example host is set up. All branches use the netname `company'
1748 for this particular VPN.
1750 Each branch is set up using the @samp{tinc init} and @samp{tinc config} commands,
1751 here we just show the end results:
1753 @subsubheading For Branch A
1755 @emph{BranchA} would be configured like this:
1757 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1762 # Real interface of internal network:
1763 # ifconfig eth0 10.1.54.1 netmask 255.255.0.0
1765 ifconfig $INTERFACE 10.1.54.1 netmask 255.0.0.0
1768 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1774 On all hosts, @file{@value{sysconfdir}/tinc/company/hosts/BranchA} contains:
1777 Subnet = 10.1.0.0/16
1780 -----BEGIN RSA PUBLIC KEY-----
1782 -----END RSA PUBLIC KEY-----
1785 Note that the IP addresses of eth0 and the VPN interface are the same.
1786 This is quite possible, if you make sure that the netmasks of the interfaces are different.
1787 It is in fact recommended to give both real internal network interfaces and VPN interfaces the same IP address,
1788 since that will make things a lot easier to remember and set up.
1791 @subsubheading For Branch B
1793 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1798 # Real interface of internal network:
1799 # ifconfig eth0 10.2.43.8 netmask 255.255.0.0
1801 ifconfig $INTERFACE 10.2.1.12 netmask 255.0.0.0
1804 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1810 Note here that the internal address (on eth0) doesn't have to be the
1811 same as on the VPN interface.
1813 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchB}:
1816 Subnet = 10.2.0.0/16
1819 -----BEGIN RSA PUBLIC KEY-----
1821 -----END RSA PUBLIC KEY-----
1825 @subsubheading For Branch C
1827 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1832 # Real interface of internal network:
1833 # ifconfig eth0 10.3.69.254 netmask 255.255.0.0
1835 ifconfig $INTERFACE 10.3.69.254 netmask 255.0.0.0
1838 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1844 C already has another daemon that runs on port 655, so they have to
1845 reserve another port for tinc. It knows the portnumber it has to listen on
1846 from it's own host configuration file.
1848 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchC}:
1852 Subnet = 10.3.0.0/16
1855 -----BEGIN RSA PUBLIC KEY-----
1857 -----END RSA PUBLIC KEY-----
1861 @subsubheading For Branch D
1863 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1868 # Real interface of internal network:
1869 # ifconfig eth0 10.4.3.32 netmask 255.255.0.0
1871 ifconfig $INTERFACE 10.4.3.32 netmask 255.0.0.0
1874 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1880 D will be connecting to C, which has a tincd running for this network on
1881 port 2000. It knows the port number from the host configuration file.
1883 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchD}:
1886 Subnet = 10.4.0.0/16
1889 -----BEGIN RSA PUBLIC KEY-----
1891 -----END RSA PUBLIC KEY-----
1894 @subsubheading Key files
1896 A, B, C and D all have their own public/private keypairs:
1898 The private RSA key is stored in @file{@value{sysconfdir}/tinc/company/rsa_key.priv},
1899 the private Ed25519 key is stored in @file{@value{sysconfdir}/tinc/company/ed25519_key.priv},
1900 and the public RSA and Ed25519 keys are put into the host configuration file in the @file{@value{sysconfdir}/tinc/company/hosts/} directory.
1902 @subsubheading Starting
1904 After each branch has finished configuration and they have distributed
1905 the host configuration files amongst them, they can start their tinc daemons.
1906 They don't necessarily have to wait for the other branches to have started
1907 their daemons, tinc will try connecting until they are available.
1910 @c ==================================================================
1912 @chapter Running tinc
1914 If everything else is done, you can start tinc by typing the following command:
1917 tinc -n @var{netname} start
1921 Tinc will detach from the terminal and continue to run in the background like a good daemon.
1922 If there are any problems however you can try to increase the debug level
1923 and look in the syslog to find out what the problems are.
1929 * Solving problems::
1931 * Sending bug reports::
1935 @c ==================================================================
1936 @node Runtime options
1937 @section Runtime options
1939 Besides the settings in the configuration file, tinc also accepts some
1940 command line options.
1942 @cindex command line
1943 @cindex runtime options
1947 @item -c, --config=@var{path}
1948 Read configuration options from the directory @var{path}. The default is
1949 @file{@value{sysconfdir}/tinc/@var{netname}/}.
1951 @item -D, --no-detach
1952 Don't fork and detach.
1953 This will also disable the automatic restart mechanism for fatal errors.
1956 @item -d, --debug=@var{level}
1957 Set debug level to @var{level}. The higher the debug level, the more gets
1958 logged. Everything goes via syslog.
1960 @item -n, --net=@var{netname}
1961 Use configuration for net @var{netname}.
1962 This will let tinc read all configuration files from
1963 @file{@value{sysconfdir}/tinc/@var{netname}/}.
1964 Specifying . for @var{netname} is the same as not specifying any @var{netname}.
1965 @xref{Multiple networks}.
1967 @item --pidfile=@var{filename}
1968 Store a cookie in @var{filename} which allows tinc to authenticate.
1969 If unspecified, the default is
1970 @file{@value{runstatedir}/tinc.@var{netname}.pid}.
1972 @item -o, --option=[@var{HOST}.]@var{KEY}=@var{VALUE}
1973 Without specifying a @var{HOST}, this will set server configuration variable @var{KEY} to @var{VALUE}.
1974 If specified as @var{HOST}.@var{KEY}=@var{VALUE},
1975 this will set the host configuration variable @var{KEY} of the host named @var{HOST} to @var{VALUE}.
1976 This option can be used more than once to specify multiple configuration variables.
1979 Lock tinc into main memory.
1980 This will prevent sensitive data like shared private keys to be written to the system swap files/partitions.
1982 This option is not supported on all platforms.
1984 @item --logfile[=@var{file}]
1985 Write log entries to a file instead of to the system logging facility.
1986 If @var{file} is omitted, the default is @file{@value{localstatedir}/log/tinc.@var{netname}.log}.
1988 @item --pidfile=@var{file}
1989 Write PID to @var{file} instead of @file{@value{runstatedir}/tinc.@var{netname}.pid}.
1991 @item --bypass-security
1992 Disables encryption and authentication.
1993 Only useful for debugging.
1996 Change process root directory to the directory where the config file is
1997 located (@file{@value{sysconfdir}/tinc/@var{netname}/} as determined by
1998 -n/--net option or as given by -c/--config option), for added security.
1999 The chroot is performed after all the initialization is done, after
2000 writing pid files and opening network sockets.
2002 This option is best used in combination with the -U/--user option described below.
2004 You will need to ensure the chroot environment contains all the files necessary
2005 for tinc to run correctly.
2006 Most importantly, for tinc to be able to resolve hostnames inside the chroot environment,
2007 you must copy @file{/etc/resolv.conf} into the chroot directory.
2008 If you want to be able to run scripts other than @file{tinc-up} in the chroot,
2009 you must ensure the appropriate shell is also installed in the chroot, along with all its dependencies.
2011 This option is not supported on all platforms.
2012 @item -U, --user=@var{user}
2013 Switch to the given @var{user} after initialization, at the same time as
2014 chroot is performed (see --chroot above). With this option tinc drops
2015 privileges, for added security.
2017 This option is not supported on all platforms.
2020 Display a short reminder of these runtime options and terminate.
2023 Output version information and exit.
2027 @c ==================================================================
2032 You can also send the following signals to a running tincd process:
2038 Forces tinc to try to connect to all uplinks immediately.
2039 Usually tinc attempts to do this itself,
2040 but increases the time it waits between the attempts each time it failed,
2041 and if tinc didn't succeed to connect to an uplink the first time after it started,
2042 it defaults to the maximum time of 15 minutes.
2045 Partially rereads configuration files.
2046 Connections to hosts whose host config file are removed are closed.
2047 New outgoing connections specified in @file{tinc.conf} will be made.
2048 If the --logfile option is used, this will also close and reopen the log file,
2049 useful when log rotation is used.
2053 @c ==================================================================
2055 @section Debug levels
2057 @cindex debug levels
2058 The tinc daemon can send a lot of messages to the syslog.
2059 The higher the debug level, the more messages it will log.
2060 Each level inherits all messages of the previous level:
2066 This will log a message indicating tinc has started along with a version number.
2067 It will also log any serious error.
2070 This will log all connections that are made with other tinc daemons.
2073 This will log status and error messages from scripts and other tinc daemons.
2076 This will log all requests that are exchanged with other tinc daemons. These include
2077 authentication, key exchange and connection list updates.
2080 This will log a copy of everything received on the meta socket.
2083 This will log all network traffic over the virtual private network.
2087 @c ==================================================================
2088 @node Solving problems
2089 @section Solving problems
2091 If tinc starts without problems, but if the VPN doesn't work, you will have to find the cause of the problem.
2092 The first thing to do is to start tinc with a high debug level in the foreground,
2093 so you can directly see everything tinc logs:
2096 tincd -n @var{netname} -d5 -D
2099 If tinc does not log any error messages, then you might want to check the following things:
2102 @item @file{tinc-up} script
2103 Does this script contain the right commands?
2104 Normally you must give the interface the address of this host on the VPN, and the netmask must be big enough so that the entire VPN is covered.
2107 Does the Subnet (or Subnets) in the host configuration file of this host match the portion of the VPN that belongs to this host?
2109 @item Firewalls and NATs
2110 Do you have a firewall or a NAT device (a masquerading firewall or perhaps an ADSL router that performs masquerading)?
2111 If so, check that it allows TCP and UDP traffic on port 655.
2112 If it masquerades and the host running tinc is behind it, make sure that it forwards TCP and UDP traffic to port 655 to the host running tinc.
2113 You can add @samp{TCPOnly = yes} to your host config file to force tinc to only use a single TCP connection,
2114 this works through most firewalls and NATs.
2119 @c ==================================================================
2120 @node Error messages
2121 @section Error messages
2123 What follows is a list of the most common error messages you might find in the logs.
2124 Some of them will only be visible if the debug level is high enough.
2127 @item Could not open /dev/tap0: No such device
2130 @item You forgot to `modprobe netlink_dev' or `modprobe ethertap'.
2131 @item You forgot to compile `Netlink device emulation' in the kernel.
2134 @item Can't write to /dev/net/tun: No such device
2137 @item You forgot to `modprobe tun'.
2138 @item You forgot to compile `Universal TUN/TAP driver' in the kernel.
2139 @item The tun device is located somewhere else in @file{/dev/}.
2142 @item Network address and prefix length do not match!
2145 @item The Subnet field must contain a @emph{network} address, trailing bits should be 0.
2146 @item If you only want to use one IP address, set the netmask to /32.
2149 @item Error reading RSA key file `rsa_key.priv': No such file or directory
2152 @item You forgot to create a public/private keypair.
2153 @item Specify the complete pathname to the private key file with the @samp{PrivateKeyFile} option.
2156 @item Warning: insecure file permissions for RSA private key file `rsa_key.priv'!
2159 @item The private key file is readable by users other than root.
2160 Use chmod to correct the file permissions.
2163 @item Creating metasocket failed: Address family not supported
2166 @item By default tinc tries to create both IPv4 and IPv6 sockets.
2167 On some platforms this might not be implemented.
2168 If the logs show @samp{Ready} later on, then at least one metasocket was created,
2169 and you can ignore this message.
2170 You can add @samp{AddressFamily = ipv4} to @file{tinc.conf} to prevent this from happening.
2173 @item Cannot route packet: unknown IPv4 destination 1.2.3.4
2176 @item You try to send traffic to a host on the VPN for which no Subnet is known.
2177 @item If it is a broadcast address (ending in .255), it probably is a samba server or a Windows host sending broadcast packets.
2181 @item Cannot route packet: ARP request for unknown address 1.2.3.4
2184 @item You try to send traffic to a host on the VPN for which no Subnet is known.
2187 @item Packet with destination 1.2.3.4 is looping back to us!
2190 @item Something is not configured right. Packets are being sent out to the
2191 virtual network device, but according to the Subnet directives in your host configuration
2192 file, those packets should go to your own host. Most common mistake is that
2193 you have a Subnet line in your host configuration file with a prefix length which is
2194 just as large as the prefix of the virtual network interface. The latter should in almost all
2195 cases be larger. Rethink your configuration.
2196 Note that you will only see this message if you specified a debug
2197 level of 5 or higher!
2198 @item Chances are that a @samp{Subnet = ...} line in the host configuration file of this tinc daemon is wrong.
2199 Change it to a subnet that is accepted locally by another interface,
2200 or if that is not the case, try changing the prefix length into /32.
2203 @item Node foo (1.2.3.4) is not reachable
2206 @item Node foo does not have a connection anymore, its tinc daemon is not running or its connection to the Internet is broken.
2209 @item Received UDP packet from unknown source 1.2.3.4 (port 12345)
2212 @item If you see this only sporadically, it is harmless and caused by a node sending packets using an old key.
2213 @item If you see this often and another node is not reachable anymore, then a NAT (masquerading firewall) is changing the source address of UDP packets.
2214 You can add @samp{TCPOnly = yes} to host configuration files to force all VPN traffic to go over a TCP connection.
2217 @item Got bad/bogus/unauthorized REQUEST from foo (1.2.3.4 port 12345)
2220 @item Node foo does not have the right public/private keypair.
2221 Generate new keypairs and distribute them again.
2222 @item An attacker tries to gain access to your VPN.
2223 @item A network error caused corruption of metadata sent from foo.
2228 @c ==================================================================
2229 @node Sending bug reports
2230 @section Sending bug reports
2232 If you really can't find the cause of a problem, or if you suspect tinc is not working right,
2233 you can send us a bugreport, see @ref{Contact information}.
2234 Be sure to include the following information in your bugreport:
2237 @item A clear description of what you are trying to achieve and what the problem is.
2238 @item What platform (operating system, version, hardware architecture) and which version of tinc you use.
2239 @item If compiling tinc fails, a copy of @file{config.log} and the error messages you get.
2240 @item Otherwise, a copy of @file{tinc.conf}, @file{tinc-up} and all files in the @file{hosts/} directory.
2241 @item The output of the commands @samp{ifconfig -a} and @samp{route -n} (or @samp{netstat -rn} if that doesn't work).
2242 @item The output of any command that fails to work as it should (like ping or traceroute).
2245 @c ==================================================================
2246 @node Controlling tinc
2247 @chapter Controlling tinc
2249 @cindex command line interface
2250 You can start, stop, control and inspect a running tincd through the tinc
2251 command. A quick example:
2254 tinc -n @var{netname} reload
2258 If tinc is started without a command, it will act as a shell; it will display a
2259 prompt, and commands can be entered on the prompt. If tinc is compiled with
2260 libreadline, history and command completion are available on the prompt. One
2261 can also pipe a script containing commands through tinc. In that case, lines
2262 starting with a # symbol will be ignored.
2265 * tinc runtime options::
2266 * tinc environment variables::
2273 @c ==================================================================
2274 @node tinc runtime options
2275 @section tinc runtime options
2279 @item -c, --config=@var{path}
2280 Read configuration options from the directory @var{path}. The default is
2281 @file{@value{sysconfdir}/tinc/@var{netname}/}.
2283 @item -n, --net=@var{netname}
2284 Use configuration for net @var{netname}. @xref{Multiple networks}.
2286 @item --pidfile=@var{filename}
2287 Use the cookie from @var{filename} to authenticate with a running tinc daemon.
2288 If unspecified, the default is
2289 @file{@value{runstatedir}/tinc.@var{netname}.pid}.
2293 Don't ask for anything (non-interactive mode).
2296 Force some commands to work despite warnings.
2299 Display a short reminder of runtime options and commands, then terminate.
2302 Output version information and exit.
2306 @c ==================================================================
2307 @node tinc environment variables
2308 @section tinc environment variables
2313 If no netname is specified on the command line with the @option{-n} option,
2314 the value of this environment variable is used.
2317 @c ==================================================================
2319 @section tinc commands
2325 @item init [@var{name}]
2326 Create initial configuration files and RSA and Ed25519 keypairs with default length.
2327 If no @var{name} for this node is given, it will be asked for.
2330 @item get @var{variable}
2331 Print the current value of configuration variable @var{variable}.
2332 If more than one variable with the same name exists,
2333 the value of each of them will be printed on a separate line.
2336 @item set @var{variable} @var{value}
2337 Set configuration variable @var{variable} to the given @var{value}.
2338 All previously existing configuration variables with the same name are removed.
2339 To set a variable for a specific host, use the notation @var{host}.@var{variable}.
2342 @item add @var{variable} @var{value}
2343 As above, but without removing any previously existing configuration variables.
2344 If the variable already exists with the given value, nothing happens.
2347 @item del @var{variable} [@var{value}]
2348 Remove configuration variables with the same name and @var{value}.
2349 If no @var{value} is given, all configuration variables with the same name will be removed.
2352 @item edit @var{filename}
2353 Start an editor for the given configuration file.
2354 You do not need to specify the full path to the file.
2358 Export the host configuration file of the local node to standard output.
2362 Export all host configuration files to standard output.
2366 Import host configuration file(s) generated by the tinc export command from standard input.
2367 Already existing host configuration files are not overwritten unless the option --force is used.
2371 The same as export followed by import.
2373 @cindex exchange-all
2375 The same as export-all followed by import.
2378 @item invite @var{name}
2379 Prepares an invitation for a new node with the given @var{name},
2380 and prints a short invitation URL that can be used with the join command.
2383 @item join [@var{URL}]
2384 Join an existing VPN using an invitation URL created using the invite command.
2385 If no @var{URL} is given, it will be read from standard input.
2388 @item start [tincd options]
2389 Start @samp{tincd}, optionally with the given extra options.
2396 @item restart [tincd options]
2397 Restart @samp{tincd}, optionally with the given extra options.
2401 Partially rereads configuration files. Connections to hosts whose host
2402 config files are removed are closed. New outgoing connections specified
2403 in @file{tinc.conf} will be made.
2407 Shows the PID of the currently running @samp{tincd}.
2409 @cindex generate-keys
2410 @item generate-keys [@var{bits}]
2411 Generate both RSA and Ed25519 keypairs (see below) and exit.
2412 tinc will ask where you want to store the files, but will default to the
2413 configuration directory (you can use the -c or -n option).
2415 @cindex generate-ed25519-keys
2416 @item generate-ed25519-keys
2417 Generate public/private Ed25519 keypair and exit.
2419 @cindex generate-rsa-keys
2420 @item generate-rsa-keys [@var{bits}]
2421 Generate public/private RSA keypair and exit. If @var{bits} is omitted, the
2422 default length will be 2048 bits. When saving keys to existing files, tinc
2423 will not delete the old keys; you have to remove them manually.
2426 @item dump [reachable] nodes
2427 Dump a list of all known nodes in the VPN.
2428 If the reachable keyword is used, only lists reachable nodes.
2431 Dump a list of all known connections in the VPN.
2434 Dump a list of all known subnets in the VPN.
2436 @item dump connections
2437 Dump a list of all meta connections with ourself.
2440 @item dump graph | digraph
2441 Dump a graph of the VPN in dotty format.
2442 Nodes are colored according to their reachability:
2443 red nodes are unreachable, orange nodes are indirectly reachable, green nodes are directly reachable.
2444 Black nodes are either directly or indirectly reachable, but direct reachability has not been tried yet.
2446 @item dump invitations
2447 Dump a list of outstanding invitations.
2448 The filename of the invitation, as well as the name of the node that is being invited is shown for each invitation.
2451 @item info @var{node} | @var{subnet} | @var{address}
2452 Show information about a particular @var{node}, @var{subnet} or @var{address}.
2453 If an @var{address} is given, any matching subnet will be shown.
2457 Purges all information remembered about unreachable nodes.
2460 @item debug @var{level}
2461 Sets debug level to @var{level}.
2464 @item log [@var{level}]
2465 Capture log messages from a running tinc daemon.
2466 An optional debug level can be given that will be applied only for log messages sent to tinc.
2470 Forces tinc to try to connect to all uplinks immediately.
2471 Usually tinc attempts to do this itself,
2472 but increases the time it waits between the attempts each time it failed,
2473 and if tinc didn't succeed to connect to an uplink the first time after it started,
2474 it defaults to the maximum time of 15 minutes.
2477 @item disconnect @var{node}
2478 Closes the meta connection with the given @var{node}.
2482 If tinc is compiled with libcurses support, this will display live traffic statistics for all the known nodes,
2483 similar to the UNIX top command.
2484 See below for more information.
2488 Dump VPN traffic going through the local tinc node in pcap-savefile format to standard output,
2489 from where it can be redirected to a file or piped through a program that can parse it directly,
2493 @item network [@var{netname}]
2494 If @var{netname} is given, switch to that network.
2495 Otherwise, display a list of all networks for which configuration files exist.
2499 This will check the configuration files for possible problems,
2500 such as unsafe file permissions, missing executable bit on script,
2501 unknown and obsolete configuration variables, wrong public and/or private keys, and so on.
2503 When problems are found, this will be printed on a line with WARNING or ERROR in front of it.
2504 Most problems must be corrected by the user itself, however in some cases (like file permissions and missing public keys),
2505 tinc will ask if it should fix the problem.
2508 @item sign [@var{filename}]
2509 Sign a file with the local node's private key.
2510 If no @var{filename} is given, the file is read from standard input.
2511 The signed file is written to standard output.
2514 @item verify @var{name} [@var{filename}]
2516 Check the signature of a file against a node's public key.
2517 The @var{name} of the node must be given,
2518 or can be "." to check against the local node's public key,
2519 or "*" to allow a signature from any node whose public key is known.
2520 If no @var{filename} is given, the file is read from standard input.
2521 If the verification is successful, a copy of the input with the signature removed is written to standard output, and the exit code will be zero.
2522 If the verification failed, nothing will be written to standard output, and the exit code will be non-zero.
2526 @c ==================================================================
2528 @section tinc examples
2530 Examples of some commands:
2533 tinc -n vpn dump graph | circo -Txlib
2534 tinc -n vpn pcap | tcpdump -r -
2538 Examples of changing the configuration using tinc:
2541 tinc -n vpn init foo
2542 tinc -n vpn add Subnet 192.168.1.0/24
2543 tinc -n vpn add bar.Address bar.example.com
2544 tinc -n vpn set Mode switch
2545 tinc -n vpn export | gpg --clearsign | mail -s "My config" vpnmaster@@example.com
2548 @c ==================================================================
2553 The top command connects to a running tinc daemon and repeatedly queries its per-node traffic counters.
2554 It displays a list of all the known nodes in the left-most column,
2555 and the amount of bytes and packets read from and sent to each node in the other columns.
2556 By default, the information is updated every second.
2557 The behaviour of the top command can be changed using the following keys:
2562 Change the interval between updates.
2563 After pressing the @key{s} key, enter the desired interval in seconds, followed by enter.
2564 Fractional seconds are honored.
2565 Intervals lower than 0.1 seconds are not allowed.
2568 Toggle between displaying current traffic rates (in packets and bytes per second)
2569 and cumulative traffic (total packets and bytes since the tinc daemon started).
2572 Sort the list of nodes by name.
2575 Sort the list of nodes by incoming amount of bytes.
2578 Sort the list of nodes by incoming amount of packets.
2581 Sort the list of nodes by outgoing amount of bytes.
2584 Sort the list of nodes by outgoing amount of packets.
2587 Sort the list of nodes by sum of incoming and outgoing amount of bytes.
2590 Sort the list of nodes by sum of incoming and outgoing amount of packets.
2593 Show amount of traffic in bytes.
2596 Show amount of traffic in kilobytes.
2599 Show amount of traffic in megabytes.
2602 Show amount of traffic in gigabytes.
2610 @c ==================================================================
2612 @chapter Invitations
2614 Invitations are an easy way to add new nodes to an existing VPN. Invitations
2615 can be created on an existing node using the @code{tinc invite} command, which
2616 generates a relatively short URL which can be given to someone else, who uses
2617 the @code{tinc join} command to automatically set up tinc so it can connect to
2618 the inviting node. The next sections describe how invitations actually work,
2619 and how to further automate the invitations.
2622 * How invitations work::
2623 * Invitation file format::
2624 * Writing an invitation-created script::
2628 @c ==================================================================
2629 @node How invitations work
2630 @section How invitations work
2632 When an invitation is created on a node (which from now on we will call the
2633 server) using the @code{tinc invite} command, an invitation file is created
2634 that contains all the information necessary for the invitee (which we will call
2635 the client) to create its configuration files. The invitation file is stays on
2636 the server, but a URL is generated that has enough information for the client
2637 to contact the server and to retrieve the invitation file. The whole URL is
2638 around 80 characters long and looks like this:
2641 server.example.org:12345/cW1NhLHS-1WPFlcFio8ztYHvewTTKYZp8BjEKg3vbMtDz7w4
2644 It is composed of four parts:
2647 hostname : port / keyhash cookie
2650 The hostname and port tell the client how to reach the tinc daemon on the server.
2651 The part after the slash looks like one blob, but is composed of two parts.
2652 The keyhash is the hash of the public key of the server.
2653 The cookie is a shared secret that identifies the client to the server.
2655 When the client connects to the server in order to join the VPN, the client and
2656 server will exchange temporary public keys. The client verifies that the hash
2657 of the server's public key matches the keyhash from the invitation URL. If
2658 not, it will immediately exit with an error. Otherwise, an ECDH exchange will
2659 happen so the client and server can communicate privately with each other. The
2660 client will then present the cookie to the server. The server uses this to
2661 look up the corresponding invitation file it generated earlier. If it exists,
2662 it will send the invitation file to the client. The client will also create a
2663 permanent public key, and send it to the server. After the exchange is
2664 completed, the connection is broken. The server creates a host config file for
2665 the client containing the client's permanent public key, and the client creates
2666 tinc.conf, host config files and possibly a tinc-up script based on the
2667 information in the invitation file.
2669 It is important that the invitation URL is kept secret until it is used; if
2670 another person gets a copy of the invitation URL before the real client runs
2671 the @code{tinc join} command, then that other person can try to join the VPN.
2674 @c ==================================================================
2675 @node Invitation file format
2676 @section Invitation file format
2678 The contents of an invitation file that is generated by the @code{tinc invite}
2679 command looks like this:
2685 #-------------------------------------#
2687 Ed25519PublicKey = augbnwegoij123587...
2688 Address = server.example.com
2691 The file is basically a concatenation of several host config blocks. Each host
2692 config block starts with @code{Name = ...}. Lines that look like @code{#---#}
2693 are not important, it just makes it easier for humans to read the file.
2694 However, the first line of an invitation file @emph{must} always start with
2697 The first host config block is always the one representing the invitee. So the
2698 first Name statement determines the name that the invitee will get. From the
2699 first block, the @file{tinc.conf} and @file{hosts/client} files will be
2700 generated; the @code{tinc join} command on the client will automatically
2701 separate statements based on whether they should be in @file{tinc.conf} or in a
2702 host config file. Some statements are special and are treated differently:
2705 @item Netname = <@var{netname}>
2706 This is a hint to the invitee which netname to use for the VPN. It is used if
2707 the invitee did not already specify a netname, and if there is no pre-existing
2708 configuration with the same netname.
2711 @item Ifconfig = <@var{address}[/@var{netmask}] | dhcp | dhcp6 | slaac>
2712 This is a hint for generating a @file{tinc-up} script.
2713 If an address is specified, a command will be added to @file{tinc-up} so the VPN interface will be configured to have the given address.
2714 If it is the word "dhcp", a command will be added to start a DHCP client on the VPN interface.
2715 If it is the word dhcpv6, it will be a DHCPv6 client.
2716 If it is "slaac", then it will add commands to enable IPv6 stateless address autoconfiguration.
2717 It is also possible to specify a MAC address, in which case a command will be added to set the MAC address of the VPN interface.
2719 The exact commands added to the @file{tinc-up} script depends on the operating system the client is using.
2720 Multiple Ifconfig statements can be specified, however one should only use one Ifconfig statement per address family.
2723 @item Route = <@var{address}[/@var{netmask}]> [<@var{gateway}>]
2724 This is a hint for generating a @file{tinc-up} script.
2725 Route statements are similar to Ifconfig statements, but add routes instead of addresses.
2726 These only allow IPv4 and IPv6 routes.
2727 If no gateway address is specified, the route is directed to the VPN interface.
2728 In general, a gateway is only necessary when running tinc in switch mode.
2731 Subsequent host config blocks are copied verbatim into their respective files
2732 in @file{hosts/}. The invitation file generated by @code{tinc invite} will
2733 normally only contain two blocks; one for the client and one for the server.
2736 @c ==================================================================
2737 @node Writing an invitation-created script
2738 @section Writing an invitation-created script
2740 When an invitation is generated, the "invitation-created" script is called (if
2741 it exists) right after the invitation file is written, but before the URL has
2742 been written to stdout. This allows one to change the invitation file
2743 automatically before the invitation URL is passed to the invitee. Here is an
2744 example shell script that approximately recreates the default invitation file:
2749 cat >$INVITATION_FILE <<EOF
2756 tinc export >>$INVITATION_FILE
2759 You can add more ConnectTo statements, and change `tinc export` to `tinc
2760 export-all` for example. But you can also use the script to automatically hand
2761 out a Subnet to the invitee. Note that the script doesn't have to be a shell script,
2762 you can use any language, it just has to be executable.
2765 @c ==================================================================
2766 @node Technical information
2767 @chapter Technical information
2772 * The meta-protocol::
2777 @c ==================================================================
2778 @node The connection
2779 @section The connection
2782 Tinc is a daemon that takes VPN data and transmit that to another host
2783 computer over the existing Internet infrastructure.
2787 * The meta-connection::
2791 @c ==================================================================
2792 @node The UDP tunnel
2793 @subsection The UDP tunnel
2795 @cindex virtual network device
2797 The data itself is read from a character device file, the so-called
2798 @emph{virtual network device}. This device is associated with a network
2799 interface. Any data sent to this interface can be read from the device,
2800 and any data written to the device gets sent from the interface.
2801 There are two possible types of virtual network devices:
2802 `tun' style, which are point-to-point devices which can only handle IPv4 and/or IPv6 packets,
2803 and `tap' style, which are Ethernet devices and handle complete Ethernet frames.
2805 So when tinc reads an Ethernet frame from the device, it determines its
2806 type. When tinc is in it's default routing mode, it can handle IPv4 and IPv6
2807 packets. Depending on the Subnet lines, it will send the packets off to their destination IP address.
2808 In the `switch' and `hub' mode, tinc will use broadcasts and MAC address discovery
2809 to deduce the destination of the packets.
2810 Since the latter modes only depend on the link layer information,
2811 any protocol that runs over Ethernet is supported (for instance IPX and Appletalk).
2812 However, only `tap' style devices provide this information.
2814 After the destination has been determined,
2815 the packet will be compressed (optionally),
2816 a sequence number will be added to the packet,
2817 the packet will then be encrypted
2818 and a message authentication code will be appended.
2820 @cindex encapsulating
2822 When that is done, time has come to actually transport the
2823 packet to the destination computer. We do this by sending the packet
2824 over an UDP connection to the destination host. This is called
2825 @emph{encapsulating}, the VPN packet (though now encrypted) is
2826 encapsulated in another IP datagram.
2828 When the destination receives this packet, the same thing happens, only
2829 in reverse. So it checks the message authentication code, decrypts the contents of the UDP datagram,
2830 checks the sequence number
2831 and writes the decrypted information to its own virtual network device.
2833 If the virtual network device is a `tun' device (a point-to-point tunnel),
2834 there is no problem for the kernel to accept a packet.
2835 However, if it is a `tap' device (this is the only available type on FreeBSD),
2836 the destination MAC address must match that of the virtual network interface.
2837 If tinc is in it's default routing mode, ARP does not work, so the correct destination MAC
2838 can not be known by the sending host.
2839 Tinc solves this by letting the receiving end detect the MAC address of its own virtual network interface
2840 and overwriting the destination MAC address of the received packet.
2842 In switch or hub modes ARP does work so the sender already knows the correct destination MAC address.
2843 In those modes every interface should have a unique MAC address, so make sure they are not the same.
2844 Because switch and hub modes rely on MAC addresses to function correctly,
2845 these modes cannot be used on the following operating systems which don't have a `tap' style virtual network device:
2846 NetBSD, Darwin and Solaris.
2849 @c ==================================================================
2850 @node The meta-connection
2851 @subsection The meta-connection
2853 Having only a UDP connection available is not enough. Though suitable
2854 for transmitting data, we want to be able to reliably send other
2855 information, such as routing and session key information to somebody.
2858 TCP is a better alternative, because it already contains protection
2859 against information being lost, unlike UDP.
2861 So we establish two connections. One for the encrypted VPN data, and one
2862 for other information, the meta-data. Hence, we call the second
2863 connection the meta-connection. We can now be sure that the
2864 meta-information doesn't get lost on the way to another computer.
2866 @cindex data-protocol
2867 @cindex meta-protocol
2868 Like with any communication, we must have a protocol, so that everybody
2869 knows what everything stands for, and how she should react. Because we
2870 have two connections, we also have two protocols. The protocol used for
2871 the UDP data is the ``data-protocol,'' the other one is the
2874 The reason we don't use TCP for both protocols is that UDP is much
2875 better for encapsulation, even while it is less reliable. The real
2876 problem is that when TCP would be used to encapsulate a TCP stream
2877 that's on the private network, for every packet sent there would be
2878 three ACKs sent instead of just one. Furthermore, if there would be
2879 a timeout, both TCP streams would sense the timeout, and both would
2880 start re-sending packets.
2883 @c ==================================================================
2884 @node The meta-protocol
2885 @section The meta-protocol
2887 The meta protocol is used to tie all tinc daemons together, and
2888 exchange information about which tinc daemon serves which virtual
2891 The meta protocol consists of requests that can be sent to the other
2892 side. Each request has a unique number and several parameters. All
2893 requests are represented in the standard ASCII character set. It is
2894 possible to use tools such as telnet or netcat to connect to a tinc
2895 daemon started with the --bypass-security option
2896 and to read and write requests by hand, provided that one
2897 understands the numeric codes sent.
2899 The authentication scheme is described in @ref{Security}. After a
2900 successful authentication, the server and the client will exchange all the
2901 information about other tinc daemons and subnets they know of, so that both
2902 sides (and all the other tinc daemons behind them) have their information
2909 ------------------------------------------------------------------
2910 ADD_EDGE node1 node2 21.32.43.54 655 222 0
2911 | | | | | +-> options
2912 | | | | +----> weight
2913 | | | +--------> UDP port of node2
2914 | | +----------------> real address of node2
2915 | +-------------------------> name of destination node
2916 +-------------------------------> name of source node
2918 ADD_SUBNET node 192.168.1.0/24
2919 | | +--> prefixlength
2920 | +--------> network address
2921 +------------------> owner of this subnet
2922 ------------------------------------------------------------------
2925 The ADD_EDGE messages are to inform other tinc daemons that a connection between
2926 two nodes exist. The address of the destination node is available so that
2927 VPN packets can be sent directly to that node.
2929 The ADD_SUBNET messages inform other tinc daemons that certain subnets belong
2930 to certain nodes. tinc will use it to determine to which node a VPN packet has
2937 ------------------------------------------------------------------
2938 DEL_EDGE node1 node2
2939 | +----> name of destination node
2940 +----------> name of source node
2942 DEL_SUBNET node 192.168.1.0/24
2943 | | +--> prefixlength
2944 | +--------> network address
2945 +------------------> owner of this subnet
2946 ------------------------------------------------------------------
2949 In case a connection between two daemons is closed or broken, DEL_EDGE messages
2950 are sent to inform the other daemons of that fact. Each daemon will calculate a
2951 new route to the the daemons, or mark them unreachable if there isn't any.
2958 ------------------------------------------------------------------
2959 REQ_KEY origin destination
2960 | +--> name of the tinc daemon it wants the key from
2961 +----------> name of the daemon that wants the key
2963 ANS_KEY origin destination 4ae0b0a82d6e0078 91 64 4
2964 | | \______________/ | | +--> MAC length
2965 | | | | +-----> digest algorithm
2966 | | | +--------> cipher algorithm
2967 | | +--> 128 bits key
2968 | +--> name of the daemon that wants the key
2969 +----------> name of the daemon that uses this key
2972 +--> daemon that has changed it's packet key
2973 ------------------------------------------------------------------
2976 The keys used to encrypt VPN packets are not sent out directly. This is
2977 because it would generate a lot of traffic on VPNs with many daemons, and
2978 chances are that not every tinc daemon will ever send a packet to every
2979 other daemon. Instead, if a daemon needs a key it sends a request for it
2980 via the meta connection of the nearest hop in the direction of the
2987 ------------------------------------------------------------------
2990 ------------------------------------------------------------------
2993 There is also a mechanism to check if hosts are still alive. Since network
2994 failures or a crash can cause a daemon to be killed without properly
2995 shutting down the TCP connection, this is necessary to keep an up to date
2996 connection list. PINGs are sent at regular intervals, except when there
2997 is also some other traffic. A little bit of salt (random data) is added
2998 with each PING and PONG message, to make sure that long sequences of PING/PONG
2999 messages without any other traffic won't result in known plaintext.
3001 This basically covers what is sent over the meta connection by tinc.
3004 @c ==================================================================
3010 Tinc got its name from ``TINC,'' short for @emph{There Is No Cabal}; the
3011 alleged Cabal was/is an organisation that was said to keep an eye on the
3012 entire Internet. As this is exactly what you @emph{don't} want, we named
3013 the tinc project after TINC.
3016 But in order to be ``immune'' to eavesdropping, you'll have to encrypt
3017 your data. Because tinc is a @emph{Secure} VPN (SVPN) daemon, it does
3018 exactly that: encrypt.
3019 However, encryption in itself does not prevent an attacker from modifying the encrypted data.
3020 Therefore, tinc also authenticates the data.
3021 Finally, tinc uses sequence numbers (which themselves are also authenticated) to prevent an attacker from replaying valid packets.
3023 Since version 1.1pre3, tinc has two protocols used to protect your data; the legacy protocol, and the new Simple Peer-to-Peer Security (SPTPS) protocol.
3024 The SPTPS protocol is designed to address some weaknesses in the legacy protocol.
3025 The new authentication protocol is used when two nodes connect to each other that both have the ExperimentalProtocol option set to yes,
3026 otherwise the legacy protocol will be used.
3029 * Legacy authentication protocol::
3030 * Simple Peer-to-Peer Security::
3031 * Encryption of network packets::
3036 @c ==================================================================
3037 @node Legacy authentication protocol
3038 @subsection Legacy authentication protocol
3040 @cindex legacy authentication protocol
3049 --------------------------------------------------------------------------
3050 client <attempts connection>
3052 server <accepts connection>
3054 client ID client 17.2
3055 | | +-> minor protocol version
3056 | +----> major protocol version
3057 +--------> name of tinc daemon
3059 server ID server 17.2
3060 | | +-> minor protocol version
3061 | +----> major protocol version
3062 +--------> name of tinc daemon
3064 client META_KEY 94 64 0 0 5f0823a93e35b69e...7086ec7866ce582b
3065 | | | | \_________________________________/
3066 | | | | +-> RSAKEYLEN bits totally random string S1,
3067 | | | | encrypted with server's public RSA key
3068 | | | +-> compression level
3069 | | +---> MAC length
3070 | +------> digest algorithm NID
3071 +---------> cipher algorithm NID
3073 server META_KEY 94 64 0 0 6ab9c1640388f8f0...45d1a07f8a672630
3074 | | | | \_________________________________/
3075 | | | | +-> RSAKEYLEN bits totally random string S2,
3076 | | | | encrypted with client's public RSA key
3077 | | | +-> compression level
3078 | | +---> MAC length
3079 | +------> digest algorithm NID
3080 +---------> cipher algorithm NID
3081 --------------------------------------------------------------------------
3084 The protocol allows each side to specify encryption algorithms and parameters,
3085 but in practice they are always fixed, since older versions of tinc did not
3086 allow them to be different from the default values. The cipher is always
3087 Blowfish in OFB mode, the digest is SHA1, but the MAC length is zero and no
3088 compression is used.
3092 @item the client will symmetrically encrypt outgoing traffic using S1
3093 @item the server will symmetrically encrypt outgoing traffic using S2
3097 --------------------------------------------------------------------------
3098 client CHALLENGE da02add1817c1920989ba6ae2a49cecbda0
3099 \_________________________________/
3100 +-> CHALLEN bits totally random string H1
3102 server CHALLENGE 57fb4b2ccd70d6bb35a64c142f47e61d57f
3103 \_________________________________/
3104 +-> CHALLEN bits totally random string H2
3106 client CHAL_REPLY 816a86
3107 +-> 160 bits SHA1 of H2
3109 server CHAL_REPLY 928ffe
3110 +-> 160 bits SHA1 of H1
3112 After the correct challenge replies are received, both ends have proved
3113 their identity. Further information is exchanged.
3115 client ACK 655 123 0
3117 | +----> estimated weight
3118 +--------> listening port of client
3120 server ACK 655 321 0
3122 | +----> estimated weight
3123 +--------> listening port of server
3124 --------------------------------------------------------------------------
3127 This legacy authentication protocol has several weaknesses, pointed out by security export Peter Gutmann.
3128 First, data is encrypted with RSA without padding.
3129 Padding schemes are designed to prevent attacks when the size of the plaintext is not equal to the size of the RSA key.
3130 Tinc always encrypts random nonces that have the same size as the RSA key, so we do not believe this leads to a break of the security.
3131 There might be timing or other side-channel attacks against RSA encryption and decryption, tinc does not employ any protection against those.
3132 Furthermore, both sides send identical messages to each other, there is no distinction between server and client,
3133 which could make a MITM attack easier.
3134 However, no exploit is known in which a third party who is not already trusted by other nodes in the VPN could gain access.
3135 Finally, the RSA keys are used to directly encrypt the session keys, which means that if the RSA keys are compromised, it is possible to decrypt all previous VPN traffic.
3136 In other words, the legacy protocol does not provide perfect forward secrecy.
3138 @c ==================================================================
3139 @node Simple Peer-to-Peer Security
3140 @subsection Simple Peer-to-Peer Security
3143 The SPTPS protocol is designed to address the weaknesses in the legacy protocol.
3144 SPTPS is based on TLS 1.2, but has been simplified: there is no support for exchanging public keys, and there is no cipher suite negotiation.
3145 Instead, SPTPS always uses a very strong cipher suite:
3146 peers authenticate each other using 521 bits ECC keys,
3147 Diffie-Hellman using ephemeral 521 bits ECC keys is used to provide perfect forward secrecy (PFS),
3148 AES-256-CTR is used for encryption, and HMAC-SHA-256 for message authentication.
3150 Similar to TLS, messages are split up in records.
3151 A complete logical record contains the following information:
3154 @item uint32_t seqno (network byte order)
3155 @item uint16_t length (network byte order)
3157 @item opaque data[length]
3158 @item opaque hmac[HMAC_SIZE] (HMAC over all preceding fields)
3161 Depending on whether SPTPS records are sent via TCP or UDP, either the seqno or the length field is omitted on the wire
3162 (but they are still included in the calculation of the HMAC);
3163 for TCP packets are guaranteed to arrive in-order so we can infer the seqno, but packets can be split or merged, so we still need the length field to determine the boundaries between records;
3164 for UDP packets we know that there is exactly one record per packet, and we know the length of a packet, but packets can be dropped, duplicated and/or reordered, so we need to include the seqno.
3166 The type field is used to distinguish between application records or handshake records.
3167 Types 0 to 127 are application records, type 128 is a handshake record, and types 129 to 255 are reserved.
3169 Before the initial handshake, no fields are encrypted, and the HMAC field is not present.
3170 After the authentication handshake, the length (if present), type and data fields are encrypted, and the HMAC field is present.
3171 For UDP packets, the seqno field is not encrypted, as it is used to determine the value of the counter used for encryption.
3173 The authentication consists of an exchange of Key EXchange, SIGnature and ACKnowledge messages, transmitted using type 128 records.
3179 ---------------------
3185 ...encrypt and HMAC using session keys from now on...
3192 ...key renegotiation starts here...
3201 ...encrypt and HMAC using new session keys from now on...
3207 ---------------------
3210 Note that the responder does not need to wait before it receives the first KEX message,
3211 it can immediately send its own once it has accepted an incoming connection.
3213 Key EXchange message:
3216 @item uint8_t kex_version (always 0 in this version of SPTPS)
3217 @item opaque nonce[32] (random number)
3218 @item opaque ecdh_key[ECDH_SIZE]
3224 @item opaque ecdsa_signature[ECDSA_SIZE]
3227 ACKnowledge message:
3230 @item empty (only sent after key renegotiation)
3236 @item At the start, both peers generate a random nonce and an Elliptic Curve public key and send it to the other in the KEX message.
3237 @item After receiving the other's KEX message, both KEX messages are concatenated (see below),
3238 and the result is signed using ECDSA.
3239 The result is sent to the other.
3240 @item After receiving the other's SIG message, the signature is verified.
3241 If it is correct, the shared secret is calculated from the public keys exchanged in the KEX message using the Elliptic Curve Diffie-Helman algorithm.
3242 @item The shared secret key is expanded using a PRF.
3243 Both nonces and the application specific label are also used as input for the PRF.
3244 @item An ACK message is sent only when doing key renegotiation, and is sent using the old encryption keys.
3245 @item The expanded key is used to key the encryption and HMAC algorithms.
3248 The signature is calculated over this string:
3251 @item uint8_t initiator (0 = local peer, 1 = remote peer is initiator)
3252 @item opaque remote_kex_message[1 + 32 + ECDH_SIZE]
3253 @item opaque local_kex_message[1 + 32 + ECDH_SIZE]
3254 @item opaque label[label_length]
3257 The PRF is calculated as follows:
3260 @item A HMAC using SHA512 is used, the shared secret is used as the key.
3261 @item For each block of 64 bytes, a HMAC is calculated. For block n: hmac[n] =
3262 HMAC_SHA512(hmac[n - 1] + seed)
3263 @item For the first block (n = 1), hmac[0] is given by HMAC_SHA512(zeroes + seed),
3264 where zeroes is a block of 64 zero bytes.
3267 The seed is as follows:
3270 @item const char[13] "key expansion"
3271 @item opaque responder_nonce[32]
3272 @item opaque initiator_nonce[32]
3273 @item opaque label[label_length]
3276 The expanded key is used as follows:
3279 @item opaque responder_cipher_key[CIPHER_KEYSIZE]
3280 @item opaque responder_digest_key[DIGEST_KEYSIZE]
3281 @item opaque initiator_cipher_key[CIPHER_KEYSIZE]
3282 @item opaque initiator_digest_key[DIGEST_KEYSIZE]
3285 Where initiator_cipher_key is the key used by session initiator to encrypt
3286 messages sent to the responder.
3288 When using 256 bits Ed25519 keys, the AES-256-CTR cipher and HMAC-SHA-256 digest algorithm,
3289 the sizes are as follows:
3292 ECDH_SIZE: 32 (= 256/8)
3293 ECDSA_SIZE: 64 (= 2 * 256/8)
3294 CIPHER_KEYSIZE: 48 (= 256/8 + 128/8)
3295 DIGEST_KEYSIZE: 32 (= 256/8)
3298 Note that the cipher key also includes the initial value for the counter.
3300 @c ==================================================================
3301 @node Encryption of network packets
3302 @subsection Encryption of network packets
3305 A data packet can only be sent if the encryption key is known to both
3306 parties, and the connection is activated. If the encryption key is not
3307 known, a request is sent to the destination using the meta connection
3311 The UDP packets can be either encrypted with the legacy protocol or with SPTPS.
3312 In case of the legacy protocol, the UDP packet containing the network packet from the VPN has the following layout:
3315 ... | IP header | UDP header | seqno | VPN packet | MAC | UDP trailer
3316 \___________________/\_____/
3318 V +---> digest algorithm
3319 Encrypted with symmetric cipher
3325 So, the entire VPN packet is encrypted using a symmetric cipher, including a 32 bits
3326 sequence number that is added in front of the actual VPN packet, to act as a unique
3327 IV for each packet and to prevent replay attacks. A message authentication code
3328 is added to the UDP packet to prevent alteration of packets.
3329 Tinc by default encrypts network packets using Blowfish with 128 bit keys in CBC mode
3330 and uses 4 byte long message authentication codes to make sure
3331 eavesdroppers cannot get and cannot change any information at all from the
3332 packets they can intercept. The encryption algorithm and message authentication
3333 algorithm can be changed in the configuration. The length of the message
3334 authentication codes is also adjustable. The length of the key for the
3335 encryption algorithm is always the default length used by LibreSSL/OpenSSL.
3337 The SPTPS protocol is described in @ref{Simple Peer-to-Peer Security}.
3338 For comparison, this is how SPTPS UDP packets look:
3341 ... | IP header | UDP header | seqno | type | VPN packet | MAC | UDP trailer
3342 \__________________/\_____/
3344 V +---> digest algorithm
3345 Encrypted with symmetric cipher
3348 The difference is that the seqno is not encrypted, since the encryption cipher is used in CTR mode,
3349 and therefore the seqno must be known before the packet can be decrypted.
3350 Furthermore, the MAC is never truncated.
3351 The SPTPS protocol always uses the AES-256-CTR cipher and HMAC-SHA-256 digest,
3352 this cannot be changed.
3355 @c ==================================================================
3356 @node Security issues
3357 @subsection Security issues
3359 In August 2000, we discovered the existence of a security hole in all versions
3360 of tinc up to and including 1.0pre2. This had to do with the way we exchanged
3361 keys. Since then, we have been working on a new authentication scheme to make
3362 tinc as secure as possible. The current version uses the LibreSSL or OpenSSL library and
3363 uses strong authentication with RSA keys.
3365 On the 29th of December 2001, Jerome Etienne posted a security analysis of tinc
3366 1.0pre4. Due to a lack of sequence numbers and a message authentication code
3367 for each packet, an attacker could possibly disrupt certain network services or
3368 launch a denial of service attack by replaying intercepted packets. The current
3369 version adds sequence numbers and message authentication codes to prevent such
3372 On the 15th of September 2003, Peter Gutmann posted a security analysis of tinc
3373 1.0.1. He argues that the 32 bit sequence number used by tinc is not a good IV,
3374 that tinc's default length of 4 bytes for the MAC is too short, and he doesn't
3375 like tinc's use of RSA during authentication. We do not know of a security hole
3376 in the legacy protocol of tinc, but it is not as strong as TLS or IPsec.
3378 The Sweet32 attack affects versions of tinc prior to 1.0.30.
3380 On September 6th, 2018, Michael Yonly contacted us and provided
3381 proof-of-concept code that allowed a remote attacker to create an
3382 authenticated, one-way connection with a node, and also that there was a
3383 possibility for a man-in-the-middle to force UDP packets from a node to be sent
3384 in plaintext. The first issue was trivial to exploit on tinc versions prior to
3385 1.0.30, but the changes in 1.0.30 to mitigate the Sweet32 attack made this
3386 weakness much harder to exploit. These issues have been fixed in tinc 1.0.35.
3388 This version of tinc comes with an improved protocol, called Simple
3389 Peer-to-Peer Security (SPTPS), which aims to be as strong as TLS with one of
3390 the strongest cipher suites. None of the above security issues affected SPTPS.
3391 However, be aware that SPTPS is only used between nodes running tinc 1.1pre* or
3392 later, and in a VPN with nodes running different versions, the security might
3393 only be as good as that of the oldest version.
3395 Cryptography is a hard thing to get right. We cannot make any
3396 guarantees. Time, review and feedback are the only things that can
3397 prove the security of any cryptographic product. If you wish to review
3398 tinc or give us feedback, you are strongly encouraged to do so.
3401 @c ==================================================================
3402 @node Platform specific information
3403 @chapter Platform specific information
3406 * Interface configuration::
3408 * Automatically starting tinc::
3411 @c ==================================================================
3412 @node Interface configuration
3413 @section Interface configuration
3415 When configuring an interface, one normally assigns it an address and a
3416 netmask. The address uniquely identifies the host on the network attached to
3417 the interface. The netmask, combined with the address, forms a subnet. It is
3418 used to add a route to the routing table instructing the kernel to send all
3419 packets which fall into that subnet to that interface. Because all packets for
3420 the entire VPN should go to the virtual network interface used by tinc, the
3421 netmask should be such that it encompasses the entire VPN.
3425 @multitable {Darwin (MacOS/X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
3427 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3428 @item Linux iproute2
3429 @tab @code{ip addr add} @var{address}@code{/}@var{prefixlength} @code{dev} @var{interface}
3431 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3433 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3435 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3437 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3438 @item Darwin (MacOS/X)
3439 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
3441 @tab @code{netsh interface ip set address} @var{interface} @code{static} @var{address} @var{netmask}
3446 @multitable {Darwin (MacOS/X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
3448 @tab @code{ifconfig} @var{interface} @code{add} @var{address}@code{/}@var{prefixlength}
3450 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
3452 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
3454 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
3456 @tab @code{ifconfig} @var{interface} @code{inet6 plumb up}
3458 @tab @code{ifconfig} @var{interface} @code{inet6 addif} @var{address} @var{address}
3459 @item Darwin (MacOS/X)
3460 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
3462 @tab @code{netsh interface ipv6 add address} @var{interface} @code{static} @var{address}/@var{prefixlength}
3465 On Linux, it is possible to create a persistent tun/tap interface which will
3466 continue to exist even if tinc quit, although this is normally not required.
3467 It can be useful to set up a tun/tap interface owned by a non-root user, so
3468 tinc can be started without needing any root privileges at all.
3470 @multitable {Darwin (MacOS/X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
3472 @tab @code{ip tuntap add dev} @var{interface} @code{mode} @var{tun|tap} @code{user} @var{username}
3475 @c ==================================================================
3479 In some cases it might be necessary to add more routes to the virtual network
3480 interface. There are two ways to indicate which interface a packet should go
3481 to, one is to use the name of the interface itself, another way is to specify
3482 the (local) address that is assigned to that interface (@var{local_address}). The
3483 former way is unambiguous and therefore preferable, but not all platforms
3486 Adding routes to IPv4 subnets:
3488 @multitable {Darwin (MacOS/X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
3490 @tab @code{route add -net} @var{network_address} @code{netmask} @var{netmask} @var{interface}
3491 @item Linux iproute2
3492 @tab @code{ip route add} @var{network_address}@code{/}@var{prefixlength} @code{dev} @var{interface}
3494 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
3496 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
3498 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
3500 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address} @code{-interface}
3501 @item Darwin (MacOS/X)
3502 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
3504 @tab @code{netsh routing ip add persistentroute} @var{network_address} @var{netmask} @var{interface} @var{local_address}
3507 Adding routes to IPv6 subnets:
3509 @multitable {Darwin (MacOS/X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
3511 @tab @code{route add -A inet6} @var{network_address}@code{/}@var{prefixlength} @var{interface}
3512 @item Linux iproute2
3513 @tab @code{ip route add} @var{network_address}@code{/}@var{prefixlength} @code{dev} @var{interface}
3515 @tab @code{route add -inet6} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
3517 @tab @code{route add -inet6} @var{network_address} @var{local_address} @code{-prefixlen} @var{prefixlength}
3519 @tab @code{route add -inet6} @var{network_address} @var{local_address} @code{-prefixlen} @var{prefixlength}
3521 @tab @code{route add -inet6} @var{network_address}@code{/}@var{prefixlength} @var{local_address} @code{-interface}
3522 @item Darwin (MacOS/X)
3525 @tab @code{netsh interface ipv6 add route} @var{network address}/@var{prefixlength} @var{interface}
3528 @c ==================================================================
3529 @node Automatically starting tinc
3530 @section Automatically starting tinc
3538 @c ==================================================================
3543 There are many Linux distributions, and historically, many of them had their
3544 own way of starting programs at boot time. Today, a number of major Linux
3545 distributions have chosen to use systemd as their init system. Tinc ships with
3546 systemd service files that allow you to start and stop tinc using systemd.
3547 There are two service files: @code{tinc.service} is used to globally enable or
3548 disable all tinc daemons managed by systemd, and
3549 @code{tinc@@@var{netname}.service} is used to enable or disable specific tinc
3550 daemons. So if one has created a tinc network with netname @code{foo}, then
3551 you have to run the following two commands to ensure it is started at boot
3555 systemctl enable tinc
3556 systemctl enable tinc@@foo
3559 To start the tinc daemon immediately if it wasn't already running, use the
3563 systemctl start tinc@@foo
3566 You can also use @samp{systemctl start tinc}, this will start all tinc daemons
3567 that are enabled. You can stop and disable tinc networks in the same way.
3569 If your system is not using systemd, then you have to look up your
3570 distribution's way of starting tinc at boot time.
3572 @c ==================================================================
3576 On Windows, if tinc is started with the @code{tinc start} command without using
3577 the @code{-D} or @code{--no-detach} option, it will automatically register
3578 itself as a service that is started at boot time. When tinc is stopped using
3579 the @code{tinc stop} command, it will also automatically unregister itself.
3580 Once tinc is registered as a service, it is also possible to stop and start
3581 tinc using the Windows Services Manager.
3583 @c ==================================================================
3584 @node Other platforms
3585 @subsection Other platforms
3587 On platforms other than the ones mentioned in the earlier sections, you have to
3588 look up your platform's way of starting programs at boot time.
3590 @c ==================================================================
3596 * Contact information::
3601 @c ==================================================================
3602 @node Contact information
3603 @section Contact information
3606 Tinc's website is at @url{https://www.tinc-vpn.org/},
3607 this server is located in the Netherlands.
3610 We have an IRC channel on the FreeNode and OFTC IRC networks. Connect to
3611 @uref{https://freenode.net/, irc.freenode.net}
3613 @uref{https://www.oftc.net/, irc.oftc.net}
3614 and join channel #tinc.
3617 @c ==================================================================
3622 @item Ivo Timmermans (zarq)
3623 @item Guus Sliepen (guus) (@email{guus@@tinc-vpn.org})
3626 We have received a lot of valuable input from users. With their help,
3627 tinc has become the flexible and robust tool that it is today. We have
3628 composed a list of contributions, in the file called @file{THANKS} in
3629 the source distribution.
3632 @c ==================================================================
3634 @unnumbered Concept Index
3636 @c ==================================================================
3640 @c ==================================================================