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{} 1998-2014 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.
35 @subtitle Setting up a Virtual Private Network with tinc
36 @author Ivo Timmermans and Guus Sliepen
39 @vskip 0pt plus 1filll
40 This is the info manual for @value{PACKAGE} version @value{VERSION}, a Virtual Private Network daemon.
42 Copyright @copyright{} 1998-2014 Ivo Timmermans,
43 Guus Sliepen <guus@@tinc-vpn.org> and
44 Wessel Dankers <wsl@@tinc-vpn.org>.
46 Permission is granted to make and distribute verbatim copies of this
47 manual provided the copyright notice and this permission notice are
48 preserved on all copies.
50 Permission is granted to copy and distribute modified versions of this
51 manual under the conditions for verbatim copying, provided that the
52 entire resulting derived work is distributed under the terms of a
53 permission notice identical to this one.
58 @c ==================================================================
68 * Technical information::
69 * Platform specific information::
71 * Concept Index:: All used terms explained
75 @c ==================================================================
80 Tinc is a Virtual Private Network (VPN) daemon that uses tunneling and
81 encryption to create a secure private network between hosts on the
84 Because the tunnel appears to the IP level network code as a normal
85 network device, there is no need to adapt any existing software.
86 The encrypted tunnels allows VPN sites to share information with each other
87 over the Internet without exposing any information to others.
89 This document is the manual for tinc. Included are chapters on how to
90 configure your computer to use tinc, as well as the configuration
91 process of tinc itself.
94 * Virtual Private Networks::
96 * Supported platforms::
99 @c ==================================================================
100 @node Virtual Private Networks
101 @section Virtual Private Networks
104 A Virtual Private Network or VPN is a network that can only be accessed
105 by a few elected computers that participate. This goal is achievable in
106 more than just one way.
109 Private networks can consist of a single stand-alone Ethernet LAN. Or
110 even two computers hooked up using a null-modem cable. In these cases,
112 obvious that the network is @emph{private}, no one can access it from the
113 outside. But if your computers are linked to the Internet, the network
114 is not private anymore, unless one uses firewalls to block all private
115 traffic. But then, there is no way to send private data to trusted
116 computers on the other end of the Internet.
119 This problem can be solved by using @emph{virtual} networks. Virtual
120 networks can live on top of other networks, but they use encapsulation to
121 keep using their private address space so they do not interfere with
122 the Internet. Mostly, virtual networks appear like a single LAN, even though
123 they can span the entire world. But virtual networks can't be secured
124 by using firewalls, because the traffic that flows through it has to go
125 through the Internet, where other people can look at it.
127 As is the case with either type of VPN, anybody could eavesdrop. Or
128 worse, alter data. Hence it's probably advisable to encrypt the data
129 that flows over the network.
131 When one introduces encryption, we can form a true VPN. Other people may
132 see encrypted traffic, but if they don't know how to decipher it (they
133 need to know the key for that), they cannot read the information that flows
134 through the VPN. This is what tinc was made for.
137 @c ==================================================================
142 I really don't quite remember what got us started, but it must have been
143 Guus' idea. He wrote a simple implementation (about 50 lines of C) that
144 used the ethertap device that Linux knows of since somewhere
145 about kernel 2.1.60. It didn't work immediately and he improved it a
146 bit. At this stage, the project was still simply called "vpnd".
148 Since then, a lot has changed---to say the least.
151 Tinc now supports encryption, it consists of a single daemon (tincd) for
152 both the receiving and sending end, it has become largely
153 runtime-configurable---in short, it has become a full-fledged
154 professional package.
156 @cindex traditional VPNs
158 Tinc also allows more than two sites to connect to eachother and form a single VPN.
159 Traditionally VPNs are created by making tunnels, which only have two endpoints.
160 Larger VPNs with more sites are created by adding more tunnels.
161 Tinc takes another approach: only endpoints are specified,
162 the software itself will take care of creating the tunnels.
163 This allows for easier configuration and improved scalability.
165 A lot can---and will be---changed. We have a number of things that we would like to
166 see in the future releases of tinc. Not everything will be available in
167 the near future. Our first objective is to make tinc work perfectly as
168 it stands, and then add more advanced features.
170 Meanwhile, we're always open-minded towards new ideas. And we're
174 @c ==================================================================
175 @node Supported platforms
176 @section Supported platforms
179 Tinc has been verified to work under Linux, FreeBSD, OpenBSD, NetBSD, Mac OS X (Darwin), Solaris, and Windows (both natively and in a Cygwin environment),
180 with various hardware architectures. These are some of the platforms
181 that are supported by the universal tun/tap device driver or other virtual network device drivers.
182 Without such a driver, tinc will most
183 likely compile and run, but it will not be able to send or receive data
187 For an up to date list of supported platforms, please check the list on
189 @uref{http://www.tinc-vpn.org/platforms/}.
197 @c Preparing your system
204 @c ==================================================================
206 @chapter Preparations
208 This chapter contains information on how to prepare your system to
212 * Configuring the kernel::
217 @c ==================================================================
218 @node Configuring the kernel
219 @section Configuring the kernel
222 * Configuration of Linux kernels::
223 * Configuration of FreeBSD kernels::
224 * Configuration of OpenBSD kernels::
225 * Configuration of NetBSD kernels::
226 * Configuration of Solaris kernels::
227 * Configuration of Darwin (Mac OS X) kernels::
228 * Configuration of Windows::
232 @c ==================================================================
233 @node Configuration of Linux kernels
234 @subsection Configuration of Linux kernels
236 @cindex Universal tun/tap
237 For tinc to work, you need a kernel that supports the Universal tun/tap device.
238 Most distributions come with kernels that already support this.
239 Here are the options you have to turn on when configuring a new kernel:
242 Code maturity level options
243 [*] Prompt for development and/or incomplete code/drivers
244 Network device support
245 <M> Universal tun/tap device driver support
248 It's not necessary to compile this driver as a module, even if you are going to
249 run more than one instance of tinc.
251 If you decide to build the tun/tap driver as a kernel module, add these lines
252 to @file{/etc/modules.conf}:
255 alias char-major-10-200 tun
259 @c ==================================================================
260 @node Configuration of FreeBSD kernels
261 @subsection Configuration of FreeBSD kernels
263 For FreeBSD version 4.1 and higher, tun and tap drivers are included in the default kernel configuration.
264 The tap driver can be loaded with @code{kldload if_tap}, or by adding @code{if_tap_load="YES"} to @file{/boot/loader.conf}.
267 @c ==================================================================
268 @node Configuration of OpenBSD kernels
269 @subsection Configuration of OpenBSD kernels
271 For OpenBSD version 2.9 and higher,
272 the tun driver is included in the default kernel configuration.
273 There is also a kernel patch from @uref{http://diehard.n-r-g.com/stuff/openbsd/}
274 which adds a tap device to OpenBSD which should work with tinc,
275 but with recent versions of OpenBSD,
276 a tun device can act as a tap device by setting the link0 option with ifconfig.
279 @c ==================================================================
280 @node Configuration of NetBSD kernels
281 @subsection Configuration of NetBSD kernels
283 For NetBSD version 1.5.2 and higher,
284 the tun driver is included in the default kernel configuration.
286 Tunneling IPv6 may not work on NetBSD's tun device.
289 @c ==================================================================
290 @node Configuration of Solaris kernels
291 @subsection Configuration of Solaris kernels
293 For Solaris 8 (SunOS 5.8) and higher,
294 the tun driver may or may not be included in the default kernel configuration.
295 If it isn't, the source can be downloaded from @uref{http://vtun.sourceforge.net/tun/}.
296 For x86 and sparc64 architectures, precompiled versions can be found at @uref{http://www.monkey.org/~dugsong/fragroute/}.
297 If the @file{net/if_tun.h} header file is missing, install it from the source package.
300 @c ==================================================================
301 @node Configuration of Darwin (Mac OS X) kernels
302 @subsection Configuration of Darwin (Mac OS X) kernels
304 Tinc on Darwin relies on a tunnel driver for its data acquisition from the kernel.
305 Tinc supports either the driver from @uref{http://tuntaposx.sourceforge.net/},
306 which supports both tun and tap style devices.
309 @c ==================================================================
310 @node Configuration of Windows
311 @subsection Configuration of Windows
313 You will need to install the latest TAP-Win32 driver from OpenVPN.
314 You can download it from @uref{http://openvpn.sourceforge.net}.
315 Using the Network Connections control panel,
316 configure the TAP-Win32 network interface in the same way as you would do from the tinc-up script,
317 as explained in the rest of the documentation.
320 @c ==================================================================
326 Before you can configure or build tinc, you need to have the OpenSSL,
327 zlib and lzo libraries installed on your system. If you try to configure tinc without
328 having them installed, configure will give you an error message, and stop.
337 @c ==================================================================
342 For all cryptography-related functions, tinc uses the functions provided
343 by the OpenSSL library.
345 If this library is not installed, you will get an error when configuring
346 tinc for build. Support for running tinc with other cryptographic libraries
347 installed @emph{may} be added in the future.
349 You can use your operating system's package manager to install this if
350 available. Make sure you install the development AND runtime versions
353 If you have to install OpenSSL manually, you can get the source code
354 from @url{http://www.openssl.org/}. Instructions on how to configure,
355 build and install this package are included within the package. Please
356 make sure you build development and runtime libraries (which is the
359 If you installed the OpenSSL libraries from source, it may be necessary
360 to let configure know where they are, by passing configure one of the
361 --with-openssl-* parameters.
364 --with-openssl=DIR OpenSSL library and headers prefix
365 --with-openssl-include=DIR OpenSSL headers directory
366 (Default is OPENSSL_DIR/include)
367 --with-openssl-lib=DIR OpenSSL library directory
368 (Default is OPENSSL_DIR/lib)
372 @subsubheading License
375 The complete source code of tinc is covered by the GNU GPL version 2.
376 Since the license under which OpenSSL is distributed is not directly
377 compatible with the terms of the GNU GPL
378 @uref{http://www.openssl.org/support/faq.html#LEGAL2}, we
379 include an exemption to the GPL (see also the file COPYING.README) to allow
380 everyone to create a statically or dynamically linked executable:
383 This program is released under the GPL with the additional exemption
384 that compiling, linking, and/or using OpenSSL is allowed. You may
385 provide binary packages linked to the OpenSSL libraries, provided that
386 all other requirements of the GPL are met.
389 Since the LZO library used by tinc is also covered by the GPL,
390 we also present the following exemption:
393 Hereby I grant a special exception to the tinc VPN project
394 (http://www.tinc-vpn.org/) to link the LZO library with the OpenSSL library
395 (http://www.openssl.org).
397 Markus F.X.J. Oberhumer
401 @c ==================================================================
406 For the optional compression of UDP packets, tinc uses the functions provided
409 If this library is not installed, you will get an error when running the
410 configure script. You can either install the zlib library, or disable support
411 for zlib compression by using the "--disable-zlib" option when running the
412 configure script. Note that if you disable support for zlib, the resulting
413 binary will not work correctly on VPNs where zlib compression is used.
415 You can use your operating system's package manager to install this if
416 available. Make sure you install the development AND runtime versions
419 If you have to install zlib manually, you can get the source code
420 from @url{http://www.gzip.org/zlib/}. Instructions on how to configure,
421 build and install this package are included within the package. Please
422 make sure you build development and runtime libraries (which is the
426 @c ==================================================================
431 Another form of compression is offered using the LZO library.
433 If this library is not installed, you will get an error when running the
434 configure script. You can either install the LZO library, or disable support
435 for LZO compression by using the "--disable-lzo" option when running the
436 configure script. Note that if you disable support for LZO, the resulting
437 binary will not work correctly on VPNs where LZO compression is used.
439 You can use your operating system's package manager to install this if
440 available. Make sure you install the development AND runtime versions
443 If you have to install lzo manually, you can get the source code
444 from @url{http://www.oberhumer.com/opensource/lzo/}. Instructions on how to configure,
445 build and install this package are included within the package. Please
446 make sure you build development and runtime libraries (which is the
459 @c ==================================================================
461 @chapter Installation
463 If you use Debian, you may want to install one of the
464 precompiled packages for your system. These packages are equipped with
465 system startup scripts and sample configurations.
467 If you cannot use one of the precompiled packages, or you want to compile tinc
468 for yourself, you can use the source. The source is distributed under
469 the GNU General Public License (GPL). Download the source from the
470 @uref{http://www.tinc-vpn.org/download/, download page}, which has
471 the checksums of these files listed; you may wish to check these with
472 md5sum before continuing.
474 Tinc comes in a convenient autoconf/automake package, which you can just
475 treat the same as any other package. Which is just untar it, type
476 `./configure' and then `make'.
477 More detailed instructions are in the file @file{INSTALL}, which is
478 included in the source distribution.
481 * Building and installing tinc::
486 @c ==================================================================
487 @node Building and installing tinc
488 @section Building and installing tinc
490 Detailed instructions on configuring the source, building tinc and installing tinc
491 can be found in the file called @file{INSTALL}.
493 @cindex binary package
494 If you happen to have a binary package for tinc for your distribution,
495 you can use the package management tools of that distribution to install tinc.
496 The documentation that comes along with your distribution will tell you how to do that.
499 * Darwin (Mac OS X) build environment::
500 * Cygwin (Windows) build environment::
501 * MinGW (Windows) build environment::
505 @c ==================================================================
506 @node Darwin (Mac OS X) build environment
507 @subsection Darwin (Mac OS X) build environment
509 In order to build tinc on Darwin, you need to install the Mac OS X Developer Tools
510 from @uref{http://developer.apple.com/tools/macosxtools.html} and
511 preferably a recent version of Fink from @uref{http://www.finkproject.org/}.
513 After installation use fink to download and install the following packages:
514 autoconf25, automake, dlcompat, m4, openssl, zlib and lzo.
516 @c ==================================================================
517 @node Cygwin (Windows) build environment
518 @subsection Cygwin (Windows) build environment
520 If Cygwin hasn't already been installed, install it directly from
521 @uref{http://www.cygwin.com/}.
523 When tinc is compiled in a Cygwin environment, it can only be run in this environment,
524 but all programs, including those started outside the Cygwin environment, will be able to use the VPN.
525 It will also support all features.
527 @c ==================================================================
528 @node MinGW (Windows) build environment
529 @subsection MinGW (Windows) build environment
531 You will need to install the MinGW environment from @uref{http://www.mingw.org}.
533 When tinc is compiled using MinGW it runs natively under Windows,
534 it is not necessary to keep MinGW installed.
536 When detaching, tinc will install itself as a service,
537 which will be restarted automatically after reboots.
540 @c ==================================================================
542 @section System files
544 Before you can run tinc, you must make sure you have all the needed
545 files on your system.
553 @c ==================================================================
555 @subsection Device files
558 Most operating systems nowadays come with the necessary device files by default,
559 or they have a mechanism to create them on demand.
561 If you use Linux and do not have udev installed,
562 you may need to create the following device file if it does not exist:
565 mknod -m 600 /dev/net/tun c 10 200
569 @c ==================================================================
571 @subsection Other files
573 @subsubheading @file{/etc/networks}
575 You may add a line to @file{/etc/networks} so that your VPN will get a
576 symbolic name. For example:
582 @subsubheading @file{/etc/services}
585 You may add this line to @file{/etc/services}. The effect is that you
586 may supply a @samp{tinc} as a valid port number to some programs. The
587 number 655 is registered with the IANA.
592 # Ivo Timmermans <ivo@@tinc-vpn.org>
607 @c ==================================================================
609 @chapter Configuration
612 * Configuration introduction::
613 * Multiple networks::
614 * How connections work::
615 * Configuration files::
616 * Generating keypairs::
617 * Network interfaces::
618 * Example configuration::
621 @c ==================================================================
622 @node Configuration introduction
623 @section Configuration introduction
625 Before actually starting to configure tinc and editing files,
626 make sure you have read this entire section so you know what to expect.
627 Then, make it clear to yourself how you want to organize your VPN:
628 What are the nodes (computers running tinc)?
629 What IP addresses/subnets do they have?
630 What is the network mask of the entire VPN?
631 Do you need special firewall rules?
632 Do you have to set up masquerading or forwarding rules?
633 Do you want to run tinc in router mode or switch mode?
634 These questions can only be answered by yourself,
635 you will not find the answers in this documentation.
636 Make sure you have an adequate understanding of networks in general.
637 @cindex Network Administrators Guide
638 A good resource on networking is the
639 @uref{http://www.tldp.org/LDP/nag2/, Linux Network Administrators Guide}.
641 If you have everything clearly pictured in your mind,
642 proceed in the following order:
643 First, generate the configuration files (@file{tinc.conf}, your host configuration file, @file{tinc-up} and perhaps @file{tinc-down}).
644 Then generate the keypairs.
645 Finally, distribute the host configuration files.
646 These steps are described in the subsections below.
649 @c ==================================================================
650 @node Multiple networks
651 @section Multiple networks
653 @cindex multiple networks
655 In order to allow you to run more than one tinc daemon on one computer,
656 for instance if your computer is part of more than one VPN,
657 you can assign a @var{netname} to your VPN.
658 It is not required if you only run one tinc daemon,
659 it doesn't even have to be the same on all the sites of your VPN,
660 but it is recommended that you choose one anyway.
662 We will assume you use a netname throughout this document.
663 This means that you call tincd with the -n argument,
664 which will assign a netname to this daemon.
666 The effect of this is that the daemon will set its configuration
667 root to @file{@value{sysconfdir}/tinc/@var{netname}/}, where @var{netname} is your argument to the -n
668 option. You'll notice that it appears in syslog as @file{tinc.@var{netname}}.
670 However, it is not strictly necessary that you call tinc with the -n
671 option. In this case, the network name would just be empty, and it will
672 be used as such. tinc now looks for files in @file{@value{sysconfdir}/tinc/}, instead of
673 @file{@value{sysconfdir}/tinc/@var{netname}/}; the configuration file should be @file{@value{sysconfdir}/tinc/tinc.conf},
674 and the host configuration files are now expected to be in @file{@value{sysconfdir}/tinc/hosts/}.
676 But it is highly recommended that you use this feature of tinc, because
677 it will be so much clearer whom your daemon talks to. Hence, we will
678 assume that you use it.
681 @c ==================================================================
682 @node How connections work
683 @section How connections work
685 When tinc starts up, it parses the command-line options and then
686 reads in the configuration file tinc.conf.
687 If it sees one or more `ConnectTo' values pointing to other tinc daemons in that file,
688 it will try to connect to those other daemons.
689 Whether this succeeds or not and whether `ConnectTo' is specified or not,
690 tinc will listen for incoming connection from other daemons.
691 If you did specify a `ConnectTo' value and the other side is not responding,
692 tinc will keep retrying.
693 This means that once started, tinc will stay running until you tell it to stop,
694 and failures to connect to other tinc daemons will not stop your tinc daemon
695 for trying again later.
696 This means you don't have to intervene if there are temporary network problems.
700 There is no real distinction between a server and a client in tinc.
701 If you wish, you can view a tinc daemon without a `ConnectTo' value as a server,
702 and one which does specify such a value as a client.
703 It does not matter if two tinc daemons have a `ConnectTo' value pointing to each other however.
706 @c ==================================================================
707 @node Configuration files
708 @section Configuration files
710 The actual configuration of the daemon is done in the file
711 @file{@value{sysconfdir}/tinc/@var{netname}/tinc.conf} and at least one other file in the directory
712 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/}.
714 An optional directory @file{@value{sysconfdir}/tinc/@var{netname}/conf.d} can be added from which
715 any .conf file will be read.
717 These file consists of comments (lines started with a #) or assignments
724 The variable names are case insensitive, and any spaces, tabs, newlines
725 and carriage returns are ignored. Note: it is not required that you put
726 in the `=' sign, but doing so improves readability. If you leave it
727 out, remember to replace it with at least one space character.
729 The server configuration is complemented with host specific configuration (see
730 the next section). Although all host configuration options for the local node
731 listed in this document can also be put in
732 @file{@value{sysconfdir}/tinc/@var{netname}/tinc.conf}, it is recommended to
733 put host specific configuration options in the host configuration file, as this
734 makes it easy to exchange with other nodes.
736 In this section all valid variables are listed in alphabetical order.
737 The default value is given between parentheses,
738 other comments are between square brackets.
741 * Main configuration variables::
742 * Host configuration variables::
748 @c ==================================================================
749 @node Main configuration variables
750 @subsection Main configuration variables
753 @cindex AddressFamily
754 @item AddressFamily = <ipv4|ipv6|any> (any)
755 This option affects the address family of listening and outgoing sockets.
756 If any is selected, then depending on the operating system
757 both IPv4 and IPv6 or just IPv6 listening sockets will be created.
759 @cindex BindToAddress
760 @item BindToAddress = <@var{address}> [<@var{port}>] [experimental]
761 If your computer has more than one IPv4 or IPv6 address, tinc
762 will by default listen on all of them for incoming connections.
763 Multiple BindToAddress variables may be specified,
764 in which case listening sockets for each specified address are made.
766 If no @var{port} is specified, the socket will be bound to the port specified by the Port option,
767 or to port 655 if neither is given.
768 To only bind to a specific port but not to a specific address, use "*" for the @var{address}.
770 This option may not work on all platforms.
772 @cindex BindToInterface
773 @item BindToInterface = <@var{interface}> [experimental]
774 If you have more than one network interface in your computer, tinc will
775 by default listen on all of them for incoming connections. It is
776 possible to bind tinc to a single interface like eth0 or ppp0 with this
779 This option may not work on all platforms.
782 @item Broadcast = <no | mst | direct> (mst) [experimental]
783 This option selects the way broadcast packets are sent to other daemons.
784 @emph{NOTE: all nodes in a VPN must use the same Broadcast mode, otherwise routing loops can form.}
788 Broadcast packets are never sent to other nodes.
791 Broadcast packets are sent and forwarded via the VPN's Minimum Spanning Tree.
792 This ensures broadcast packets reach all nodes.
795 Broadcast packets are sent directly to all nodes that can be reached directly.
796 Broadcast packets received from other nodes are never forwarded.
797 If the IndirectData option is also set, broadcast packets will only be sent to nodes which we have a meta connection to.
801 @item ConnectTo = <@var{name}>
802 Specifies which other tinc daemon to connect to on startup.
803 Multiple ConnectTo variables may be specified,
804 in which case outgoing connections to each specified tinc daemon are made.
805 The names should be known to this tinc daemon
806 (i.e., there should be a host configuration file for the name on the ConnectTo line).
808 If you don't specify a host with ConnectTo,
809 tinc won't try to connect to other daemons at all,
810 and will instead just listen for incoming connections.
813 @item DecrementTTL = <yes | no> (no) [experimental]
814 When enabled, tinc will decrement the Time To Live field in IPv4 packets, or the Hop Limit field in IPv6 packets,
815 before forwarding a received packet to the virtual network device or to another node,
816 and will drop packets that have a TTL value of zero,
817 in which case it will send an ICMP Time Exceeded packet back.
819 Do not use this option if you use switch mode and want to use IPv6.
822 @item Device = <@var{device}> (@file{/dev/tap0}, @file{/dev/net/tun} or other depending on platform)
823 The virtual network device to use.
824 Tinc will automatically detect what kind of device it is.
825 Under Windows, use @var{Interface} instead of @var{Device}.
826 Note that you can only use one device per daemon.
827 See also @ref{Device files}.
830 @item DeviceType = <@var{type}> (platform dependent)
831 The type of the virtual network device.
832 Tinc will normally automatically select the right type of tun/tap interface, and this option should not be used.
833 However, this option can be used to select one of the special interface types, if support for them is compiled in.
838 Use a dummy interface.
839 No packets are ever read or written to a virtual network device.
840 Useful for testing, or when setting up a node that only forwards packets for other nodes.
844 Open a raw socket, and bind it to a pre-existing
845 @var{Interface} (eth0 by default).
846 All packets are read from this interface.
847 Packets received for the local node are written to the raw socket.
848 However, at least on Linux, the operating system does not process IP packets destined for the local host.
852 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}.
853 Packets are read from and written to this multicast socket.
854 This can be used to connect to UML, QEMU or KVM instances listening on the same multicast address.
855 Do NOT connect multiple tinc daemons to the same multicast address, this will very likely cause routing loops.
856 Also note that this can cause decrypted VPN packets to be sent out on a real network if misconfigured.
859 @item uml (not compiled in by default)
860 Create a UNIX socket with the filename specified by
861 @var{Device}, or @file{@value{localstatedir}/run/@var{netname}.umlsocket}
863 Tinc will wait for a User Mode Linux instance to connect to this socket.
866 @item vde (not compiled in by default)
867 Uses the libvdeplug library to connect to a Virtual Distributed Ethernet switch,
868 using the UNIX socket specified by
869 @var{Device}, or @file{@value{localstatedir}/run/vde.ctl}
873 Also, in case tinc does not seem to correctly interpret packets received from the virtual network device,
874 it can be used to change the way packets are interpreted:
877 @item tun (BSD and Linux)
879 Depending on the platform, this can either be with or without an address family header (see below).
882 @item tunnohead (BSD)
883 Set type to tun without an address family header.
884 Tinc will expect packets read from the virtual network device to start with an IP header.
885 On some platforms IPv6 packets cannot be read from or written to the device in this mode.
888 @item tunifhead (BSD)
889 Set type to tun with an address family header.
890 Tinc will expect packets read from the virtual network device
891 to start with a four byte header containing the address family,
892 followed by an IP header.
893 This mode should support both IPv4 and IPv6 packets.
895 @item tap (BSD and Linux)
897 Tinc will expect packets read from the virtual network device
898 to start with an Ethernet header.
902 @item DirectOnly = <yes|no> (no) [experimental]
903 When this option is enabled, packets that cannot be sent directly to the destination node,
904 but which would have to be forwarded by an intermediate node, are dropped instead.
905 When combined with the IndirectData option,
906 packets for nodes for which we do not have a meta connection with are also dropped.
909 @item Forwarding = <off|internal|kernel> (internal) [experimental]
910 This option selects the way indirect packets are forwarded.
914 Incoming packets that are not meant for the local node,
915 but which should be forwarded to another node, are dropped.
918 Incoming packets that are meant for another node are forwarded by tinc internally.
920 This is the default mode, and unless you really know you need another forwarding mode, don't change it.
923 Incoming packets are always sent to the TUN/TAP device, even if the packets are not for the local node.
924 This is less efficient, but allows the kernel to apply its routing and firewall rules on them,
925 and can also help debugging.
928 @cindex GraphDumpFile
929 @item GraphDumpFile = <@var{filename}> [experimental]
930 If this option is present,
931 tinc will dump the current network graph to the file @var{filename}
932 every minute, unless there were no changes to the graph.
933 The file is in a format that can be read by graphviz tools.
934 If @var{filename} starts with a pipe symbol |,
935 then the rest of the filename is interpreted as a shell command
936 that is executed, the graph is then sent to stdin.
939 @item Hostnames = <yes|no> (no)
940 This option selects whether IP addresses (both real and on the VPN)
941 should be resolved. Since DNS lookups are blocking, it might affect
942 tinc's efficiency, even stopping the daemon for a few seconds every time
943 it does a lookup if your DNS server is not responding.
945 This does not affect resolving hostnames to IP addresses from the
946 configuration file, but whether hostnames should be resolved while logging.
949 @item IffOneQueue = <yes|no> (no) [experimental]
950 (Linux only) Set IFF_ONE_QUEUE flag on TUN/TAP devices.
953 @item Interface = <@var{interface}>
954 Defines the name of the interface corresponding to the virtual network device.
955 Depending on the operating system and the type of device this may or may not actually set the name of the interface.
956 Under Windows, this variable is used to select which network interface will be used.
957 If you specified a Device, this variable is almost always already correctly set.
960 @item KeyExpire = <@var{seconds}> (3600)
961 This option controls the time the encryption keys used to encrypt the data
962 are valid. It is common practice to change keys at regular intervals to
963 make it even harder for crackers, even though it is thought to be nearly
964 impossible to crack a single key.
966 @cindex LocalDiscovery
967 @item LocalDiscovery = <yes | no> (no) [experimental]
968 When enabled, tinc will try to detect peers that are on the same local network.
969 This will allow direct communication using LAN addresses, even if both peers are behind a NAT
970 and they only ConnectTo a third node outside the NAT,
971 which normally would prevent the peers from learning each other's LAN address.
973 Currently, local discovery is implemented by sending broadcast packets to the LAN during path MTU discovery.
974 This feature may not work in all possible situations.
977 @item MACExpire = <@var{seconds}> (600)
978 This option controls the amount of time MAC addresses are kept before they are removed.
979 This only has effect when Mode is set to "switch".
982 @item MaxTimeout = <@var{seconds}> (900)
983 This is the maximum delay before trying to reconnect to other tinc daemons.
986 @item Mode = <router|switch|hub> (router)
987 This option selects the way packets are routed to other daemons.
993 variables in the host configuration files will be used to form a routing table.
994 Only unicast packets of routable protocols (IPv4 and IPv6) are supported in this mode.
996 This is the default mode, and unless you really know you need another mode, don't change it.
1000 In this mode the MAC addresses of the packets on the VPN will be used to
1001 dynamically create a routing table just like an Ethernet switch does.
1002 Unicast, multicast and broadcast packets of every protocol that runs over Ethernet are supported in this mode
1003 at the cost of frequent broadcast ARP requests and routing table updates.
1005 This mode is primarily useful if you want to bridge Ethernet segments.
1009 This mode is almost the same as the switch mode, but instead
1010 every packet will be broadcast to the other daemons
1011 while no routing table is managed.
1015 @item Name = <@var{name}> [required]
1016 This is a symbolic name for this connection.
1017 The name should consist only of alphanumeric and underscore characters (a-z, A-Z, 0-9 and _).
1019 If Name starts with a $, then the contents of the environment variable that follows will be used.
1020 In that case, invalid characters will be converted to underscores.
1021 If Name is $HOST, but no such environment variable exist,
1022 the hostname will be read using the gethostnname() system call.
1024 @cindex PingInterval
1025 @item PingInterval = <@var{seconds}> (60)
1026 The number of seconds of inactivity that tinc will wait before sending a
1027 probe to the other end.
1030 @item PingTimeout = <@var{seconds}> (5)
1031 The number of seconds to wait for a response to pings or to allow meta
1032 connections to block. If the other end doesn't respond within this time,
1033 the connection is terminated, and the others will be notified of this.
1035 @cindex PriorityInheritance
1036 @item PriorityInheritance = <yes|no> (no) [experimental]
1037 When this option is enabled the value of the TOS field of tunneled IPv4 packets
1038 will be inherited by the UDP packets that are sent out.
1041 @item PrivateKey = <@var{key}> [obsolete]
1042 This is the RSA private key for tinc. However, for safety reasons it is
1043 advised to store private keys of any kind in separate files. This prevents
1044 accidental eavesdropping if you are editing the configuration file.
1046 @cindex PrivateKeyFile
1047 @item PrivateKeyFile = <@var{path}> (@file{@value{sysconfdir}/tinc/@var{netname}/rsa_key.priv})
1048 This is the full path name of the RSA private key file that was
1049 generated by @samp{tincd --generate-keys}. It must be a full path, not a
1052 @cindex ProcessPriority
1053 @item ProcessPriority = <low|normal|high>
1054 When this option is used the priority of the tincd process will be adjusted.
1055 Increasing the priority may help to reduce latency and packet loss on the VPN.
1058 @item Proxy = socks4 | socks5 | http | exec @var{...} [experimental]
1059 Use a proxy when making outgoing connections.
1060 The following proxy types are currently supported:
1064 @item socks4 <@var{address}> <@var{port}> [<@var{username}>]
1065 Connects to the proxy using the SOCKS version 4 protocol.
1066 Optionally, a @var{username} can be supplied which will be passed on to the proxy server.
1069 @item socks5 <@var{address}> <@var{port}> [<@var{username}> <@var{password}>]
1070 Connect to the proxy using the SOCKS version 5 protocol.
1071 If a @var{username} and @var{password} are given, basic username/password authentication will be used,
1072 otherwise no authentication will be used.
1075 @item http <@var{address}> <@var{port}>
1076 Connects to the proxy and sends a HTTP CONNECT request.
1079 @item exec <@var{command}>
1080 Executes the given command which should set up the outgoing connection.
1081 The environment variables @env{NAME}, @env{NODE}, @env{REMOTEADDRES} and @env{REMOTEPORT} are available.
1084 @cindex ReplayWindow
1085 @item ReplayWindow = <bytes> (16)
1086 This is the size of the replay tracking window for each remote node, in bytes.
1087 The window is a bitfield which tracks 1 packet per bit, so for example
1088 the default setting of 16 will track up to 128 packets in the window. In high
1089 bandwidth scenarios, setting this to a higher value can reduce packet loss from
1090 the interaction of replay tracking with underlying real packet loss and/or
1091 reordering. Setting this to zero will disable replay tracking completely and
1092 pass all traffic, but leaves tinc vulnerable to replay-based attacks on your
1095 @cindex StrictSubnets
1096 @item StrictSubnets = <yes|no> (no) [experimental]
1097 When this option is enabled tinc will only use Subnet statements which are
1098 present in the host config files in the local
1099 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/} directory.
1100 Subnets learned via connections to other nodes and which are not
1101 present in the local host config files are ignored.
1103 @cindex TunnelServer
1104 @item TunnelServer = <yes|no> (no) [experimental]
1105 When this option is enabled tinc will no longer forward information between other tinc daemons,
1106 and will only allow connections with nodes for which host config files are present in the local
1107 @file{@value{sysconfdir}/tinc/@var{netname}/hosts/} directory.
1108 Setting this options also implicitly sets StrictSubnets.
1111 @item UDPRcvBuf = <bytes> (OS default)
1112 Sets the socket receive buffer size for the UDP socket, in bytes.
1113 If unset, the default buffer size will be used by the operating system.
1116 @item UDPSndBuf = <bytes> Pq OS default
1117 Sets the socket send buffer size for the UDP socket, in bytes.
1118 If unset, the default buffer size will be used by the operating system.
1123 @c ==================================================================
1124 @node Host configuration variables
1125 @subsection Host configuration variables
1129 @item Address = <@var{IP address}|@var{hostname}> [<port>] [recommended]
1130 This variable is only required if you want to connect to this host. It
1131 must resolve to the external IP address where the host can be reached,
1132 not the one that is internal to the VPN.
1133 If no port is specified, the default Port is used.
1134 Multiple Address variables can be specified, in which case each address will be
1135 tried until a working connection has been established.
1138 @item Cipher = <@var{cipher}> (blowfish)
1139 The symmetric cipher algorithm used to encrypt UDP packets.
1140 Any cipher supported by OpenSSL is recognized.
1141 Furthermore, specifying "none" will turn off packet encryption.
1142 It is best to use only those ciphers which support CBC mode.
1145 @item ClampMSS = <yes|no> (yes)
1146 This option specifies whether tinc should clamp the maximum segment size (MSS)
1147 of TCP packets to the path MTU. This helps in situations where ICMP
1148 Fragmentation Needed or Packet too Big messages are dropped by firewalls.
1151 @item Compression = <@var{level}> (0)
1152 This option sets the level of compression used for UDP packets.
1153 Possible values are 0 (off), 1 (fast zlib) and any integer up to 9 (best zlib),
1154 10 (fast lzo) and 11 (best lzo).
1157 @item Digest = <@var{digest}> (sha1)
1158 The digest algorithm used to authenticate UDP packets.
1159 Any digest supported by OpenSSL is recognized.
1160 Furthermore, specifying "none" will turn off packet authentication.
1162 @cindex IndirectData
1163 @item IndirectData = <yes|no> (no)
1164 This option specifies whether other tinc daemons besides the one you
1165 specified with ConnectTo can make a direct connection to you. This is
1166 especially useful if you are behind a firewall and it is impossible to
1167 make a connection from the outside to your tinc daemon. Otherwise, it
1168 is best to leave this option out or set it to no.
1171 @item MACLength = <@var{bytes}> (4)
1172 The length of the message authentication code used to authenticate UDP packets.
1173 Can be anything from 0
1174 up to the length of the digest produced by the digest algorithm.
1177 @item PMTU = <@var{mtu}> (1514)
1178 This option controls the initial path MTU to this node.
1180 @cindex PMTUDiscovery
1181 @item PMTUDiscovery = <yes|no> (yes)
1182 When this option is enabled, tinc will try to discover the path MTU to this node.
1183 After the path MTU has been discovered, it will be enforced on the VPN.
1186 @item Port = <@var{port}> (655)
1187 This is the port this tinc daemon listens on.
1188 You can use decimal portnumbers or symbolic names (as listed in @file{/etc/services}).
1191 @item PublicKey = <@var{key}> [obsolete]
1192 This is the RSA public key for this host.
1194 @cindex PublicKeyFile
1195 @item PublicKeyFile = <@var{path}> [obsolete]
1196 This is the full path name of the RSA public key file that was generated
1197 by @samp{tincd --generate-keys}. It must be a full path, not a relative
1201 From version 1.0pre4 on tinc will store the public key directly into the
1202 host configuration file in PEM format, the above two options then are not
1203 necessary. Either the PEM format is used, or exactly
1204 @strong{one of the above two options} must be specified
1205 in each host configuration file, if you want to be able to establish a
1206 connection with that host.
1209 @item Subnet = <@var{address}[/@var{prefixlength}[#@var{weight}]]>
1210 The subnet which this tinc daemon will serve.
1211 Tinc tries to look up which other daemon it should send a packet to by searching the appropriate subnet.
1212 If the packet matches a subnet,
1213 it will be sent to the daemon who has this subnet in his host configuration file.
1214 Multiple subnet lines can be specified for each daemon.
1216 Subnets can either be single MAC, IPv4 or IPv6 addresses,
1217 in which case a subnet consisting of only that single address is assumed,
1218 or they can be a IPv4 or IPv6 network address with a prefixlength.
1219 For example, IPv4 subnets must be in a form like 192.168.1.0/24,
1220 where 192.168.1.0 is the network address and 24 is the number of bits set in the netmask.
1221 Note that subnets like 192.168.1.1/24 are invalid!
1222 Read a networking HOWTO/FAQ/guide if you don't understand this.
1223 IPv6 subnets are notated like fec0:0:0:1::/64.
1224 MAC addresses are notated like 0:1a:2b:3c:4d:5e.
1226 @cindex CIDR notation
1227 Prefixlength is the number of bits set to 1 in the netmask part; for
1228 example: netmask 255.255.255.0 would become /24, 255.255.252.0 becomes
1229 /22. This conforms to standard CIDR notation as described in
1230 @uref{http://www.ietf.org/rfc/rfc1519.txt, RFC1519}
1232 @cindex Subnet weight
1233 A Subnet can be given a weight to indicate its priority over identical Subnets
1234 owned by different nodes. The default weight is 10. Lower values indicate
1235 higher priority. Packets will be sent to the node with the highest priority,
1236 unless that node is not reachable, in which case the node with the next highest
1237 priority will be tried, and so on.
1240 @item TCPonly = <yes|no> (no) [deprecated]
1241 If this variable is set to yes, then the packets are tunnelled over a
1242 TCP connection instead of a UDP connection. This is especially useful
1243 for those who want to run a tinc daemon from behind a masquerading
1244 firewall, or if UDP packet routing is disabled somehow.
1245 Setting this options also implicitly sets IndirectData.
1247 Since version 1.0.10, tinc will automatically detect whether communication via
1248 UDP is possible or not.
1252 @c ==================================================================
1257 Apart from reading the server and host configuration files,
1258 tinc can also run scripts at certain moments.
1259 Under Windows (not Cygwin), the scripts should have the extension .bat.
1263 @item @value{sysconfdir}/tinc/@var{netname}/tinc-up
1264 This is the most important script.
1265 If it is present it will be executed right after the tinc daemon has been
1266 started and has connected to the virtual network device.
1267 It should be used to set up the corresponding network interface,
1268 but can also be used to start other things.
1269 Under Windows you can use the Network Connections control panel instead of creating this script.
1272 @item @value{sysconfdir}/tinc/@var{netname}/tinc-down
1273 This script is started right before the tinc daemon quits.
1275 @item @value{sysconfdir}/tinc/@var{netname}/hosts/@var{host}-up
1276 This script is started when the tinc daemon with name @var{host} becomes reachable.
1278 @item @value{sysconfdir}/tinc/@var{netname}/hosts/@var{host}-down
1279 This script is started when the tinc daemon with name @var{host} becomes unreachable.
1281 @item @value{sysconfdir}/tinc/@var{netname}/host-up
1282 This script is started when any host becomes reachable.
1284 @item @value{sysconfdir}/tinc/@var{netname}/host-down
1285 This script is started when any host becomes unreachable.
1287 @item @value{sysconfdir}/tinc/@var{netname}/subnet-up
1288 This script is started when a subnet becomes reachable.
1289 The Subnet and the node it belongs to are passed in environment variables.
1291 @item @value{sysconfdir}/tinc/@var{netname}/subnet-down
1292 This script is started when a subnet becomes unreachable.
1295 @cindex environment variables
1296 The scripts are started without command line arguments,
1297 but can make use of certain environment variables.
1298 Under UNIX like operating systems the names of environment variables must be preceded by a $ in scripts.
1299 Under Windows, in @file{.bat} files, they have to be put between % signs.
1304 If a netname was specified, this environment variable contains it.
1308 Contains the name of this tinc daemon.
1312 Contains the name of the virtual network device that tinc uses.
1316 Contains the name of the virtual network interface that tinc uses.
1317 This should be used for commands like ifconfig.
1321 When a host becomes (un)reachable, this is set to its name.
1322 If a subnet becomes (un)reachable, this is set to the owner of that subnet.
1324 @cindex REMOTEADDRESS
1326 When a host becomes (un)reachable, this is set to its real address.
1330 When a host becomes (un)reachable,
1331 this is set to the port number it uses for communication with other tinc daemons.
1335 When a subnet becomes (un)reachable, this is set to the subnet.
1339 When a subnet becomes (un)reachable, this is set to the subnet weight.
1344 @c ==================================================================
1345 @node How to configure
1346 @subsection How to configure
1348 @subsubheading Step 1. Creating the main configuration file
1350 The main configuration file will be called @file{@value{sysconfdir}/tinc/@var{netname}/tinc.conf}.
1351 Adapt the following example to create a basic configuration file:
1354 Name = @var{yourname}
1355 Device = @file{/dev/tap0}
1358 Then, if you know to which other tinc daemon(s) yours is going to connect,
1359 add `ConnectTo' values.
1361 @subsubheading Step 2. Creating your host configuration file
1363 If you added a line containing `Name = yourname' in the main configuration file,
1364 you will need to create a host configuration file @file{@value{sysconfdir}/tinc/@var{netname}/hosts/yourname}.
1365 Adapt the following example to create a host configuration file:
1368 Address = your.real.hostname.org
1369 Subnet = 192.168.1.0/24
1372 You can also use an IP address instead of a hostname.
1373 The `Subnet' specifies the address range that is local for @emph{your part of the VPN only}.
1374 If you have multiple address ranges you can specify more than one `Subnet'.
1375 You might also need to add a `Port' if you want your tinc daemon to run on a different port number than the default (655).
1378 @c ==================================================================
1379 @node Generating keypairs
1380 @section Generating keypairs
1382 @cindex key generation
1383 Now that you have already created the main configuration file and your host configuration file,
1384 you can easily create a public/private keypair by entering the following command:
1387 tincd -n @var{netname} -K
1390 Tinc will generate a public and a private key and ask you where to put them.
1391 Just press enter to accept the defaults.
1394 @c ==================================================================
1395 @node Network interfaces
1396 @section Network interfaces
1398 Before tinc can start transmitting data over the tunnel, it must
1399 set up the virtual network interface.
1401 First, decide which IP addresses you want to have associated with these
1402 devices, and what network mask they must have.
1404 Tinc will open a virtual network device (@file{/dev/tun}, @file{/dev/tap0} or similar),
1405 which will also create a network interface called something like @samp{tun0}, @samp{tap0}.
1406 If you are using the Linux tun/tap driver, the network interface will by default have the same name as the @var{netname}.
1407 Under Windows you can change the name of the network interface from the Network Connections control panel.
1410 You can configure the network interface by putting ordinary ifconfig, route, and other commands
1411 to a script named @file{@value{sysconfdir}/tinc/@var{netname}/tinc-up}.
1412 When tinc starts, this script will be executed. When tinc exits, it will execute the script named
1413 @file{@value{sysconfdir}/tinc/@var{netname}/tinc-down}, but normally you don't need to create that script.
1415 An example @file{tinc-up} script:
1419 ifconfig $INTERFACE 192.168.1.1 netmask 255.255.0.0
1422 This script gives the interface an IP address and a netmask.
1423 The kernel will also automatically add a route to this interface, so normally you don't need
1424 to add route commands to the @file{tinc-up} script.
1425 The kernel will also bring the interface up after this command.
1427 The netmask is the mask of the @emph{entire} VPN network, not just your
1430 The exact syntax of the ifconfig and route commands differs from platform to platform.
1431 You can look up the commands for setting addresses and adding routes in @ref{Platform specific information},
1432 but it is best to consult the manpages of those utilities on your platform.
1435 @c ==================================================================
1436 @node Example configuration
1437 @section Example configuration
1441 Imagine the following situation. Branch A of our example `company' wants to connect
1442 three branch offices in B, C and D using the Internet. All four offices
1443 have a 24/7 connection to the Internet.
1445 A is going to serve as the center of the network. B and C will connect
1446 to A, and D will connect to C. Each office will be assigned their own IP
1450 A: net 10.1.0.0 mask 255.255.0.0 gateway 10.1.54.1 internet IP 1.2.3.4
1451 B: net 10.2.0.0 mask 255.255.0.0 gateway 10.2.1.12 internet IP 2.3.4.5
1452 C: net 10.3.0.0 mask 255.255.0.0 gateway 10.3.69.254 internet IP 3.4.5.6
1453 D: net 10.4.0.0 mask 255.255.0.0 gateway 10.4.3.32 internet IP 4.5.6.7
1456 Here, ``gateway'' is the VPN IP address of the machine that is running the
1457 tincd, and ``internet IP'' is the IP address of the firewall, which does not
1458 need to run tincd, but it must do a port forwarding of TCP and UDP on port
1459 655 (unless otherwise configured).
1461 In this example, it is assumed that eth0 is the interface that points to
1462 the inner (physical) LAN of the office, although this could also be the
1463 same as the interface that leads to the Internet. The configuration of
1464 the real interface is also shown as a comment, to give you an idea of
1465 how these example host is set up. All branches use the netname `company'
1466 for this particular VPN.
1468 @subsubheading For Branch A
1470 @emph{BranchA} would be configured like this:
1472 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1475 # Real interface of internal network:
1476 # ifconfig eth0 10.1.54.1 netmask 255.255.0.0
1478 ifconfig $INTERFACE 10.1.54.1 netmask 255.0.0.0
1481 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1488 On all hosts, @file{@value{sysconfdir}/tinc/company/hosts/BranchA} contains:
1491 Subnet = 10.1.0.0/16
1494 -----BEGIN RSA PUBLIC KEY-----
1496 -----END RSA PUBLIC KEY-----
1499 Note that the IP addresses of eth0 and tap0 are the same.
1500 This is quite possible, if you make sure that the netmasks of the interfaces are different.
1501 It is in fact recommended to give both real internal network interfaces and tap interfaces the same IP address,
1502 since that will make things a lot easier to remember and set up.
1505 @subsubheading For Branch B
1507 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1510 # Real interface of internal network:
1511 # ifconfig eth0 10.2.43.8 netmask 255.255.0.0
1513 ifconfig $INTERFACE 10.2.1.12 netmask 255.0.0.0
1516 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1523 Note here that the internal address (on eth0) doesn't have to be the
1524 same as on the tap0 device. Also, ConnectTo is given so that this node will
1525 always try to connect to BranchA.
1527 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchB}:
1530 Subnet = 10.2.0.0/16
1533 -----BEGIN RSA PUBLIC KEY-----
1535 -----END RSA PUBLIC KEY-----
1539 @subsubheading For Branch C
1541 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1544 # Real interface of internal network:
1545 # ifconfig eth0 10.3.69.254 netmask 255.255.0.0
1547 ifconfig $INTERFACE 10.3.69.254 netmask 255.0.0.0
1550 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1558 C already has another daemon that runs on port 655, so they have to
1559 reserve another port for tinc. It knows the portnumber it has to listen on
1560 from it's own host configuration file.
1562 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchC}:
1566 Subnet = 10.3.0.0/16
1569 -----BEGIN RSA PUBLIC KEY-----
1571 -----END RSA PUBLIC KEY-----
1575 @subsubheading For Branch D
1577 In @file{@value{sysconfdir}/tinc/company/tinc-up}:
1580 # Real interface of internal network:
1581 # ifconfig eth0 10.4.3.32 netmask 255.255.0.0
1583 ifconfig $INTERFACE 10.4.3.32 netmask 255.0.0.0
1586 and in @file{@value{sysconfdir}/tinc/company/tinc.conf}:
1591 Device = /dev/net/tun
1594 D will be connecting to C, which has a tincd running for this network on
1595 port 2000. It knows the port number from the host configuration file.
1596 Also note that since D uses the tun/tap driver, the network interface
1597 will not be called `tun' or `tap0' or something like that, but will
1598 have the same name as netname.
1600 On all hosts, in @file{@value{sysconfdir}/tinc/company/hosts/BranchD}:
1603 Subnet = 10.4.0.0/16
1606 -----BEGIN RSA PUBLIC KEY-----
1608 -----END RSA PUBLIC KEY-----
1611 @subsubheading Key files
1613 A, B, C and D all have generated a public/private keypair with the following command:
1619 The private key is stored in @file{@value{sysconfdir}/tinc/company/rsa_key.priv},
1620 the public key is put into the host configuration file in the @file{@value{sysconfdir}/tinc/company/hosts/} directory.
1621 During key generation, tinc automatically guesses the right filenames based on the -n option and
1622 the Name directive in the @file{tinc.conf} file (if it is available).
1624 @subsubheading Starting
1626 After each branch has finished configuration and they have distributed
1627 the host configuration files amongst them, they can start their tinc daemons.
1628 They don't necessarily have to wait for the other branches to have started
1629 their daemons, tinc will try connecting until they are available.
1632 @c ==================================================================
1634 @chapter Running tinc
1636 If everything else is done, you can start tinc by typing the following command:
1639 tincd -n @var{netname}
1643 Tinc will detach from the terminal and continue to run in the background like a good daemon.
1644 If there are any problems however you can try to increase the debug level
1645 and look in the syslog to find out what the problems are.
1651 * Solving problems::
1653 * Sending bug reports::
1657 @c ==================================================================
1658 @node Runtime options
1659 @section Runtime options
1661 Besides the settings in the configuration file, tinc also accepts some
1662 command line options.
1664 @cindex command line
1665 @cindex runtime options
1669 @item -c, --config=@var{path}
1670 Read configuration options from the directory @var{path}. The default is
1671 @file{@value{sysconfdir}/tinc/@var{netname}/}.
1673 @item -D, --no-detach
1674 Don't fork and detach.
1675 This will also disable the automatic restart mechanism for fatal errors.
1678 @item -d, --debug=@var{level}
1679 Set debug level to @var{level}. The higher the debug level, the more gets
1680 logged. Everything goes via syslog.
1682 @item -k, --kill[=@var{signal}]
1683 Attempt to kill a running tincd (optionally with the specified @var{signal} instead of SIGTERM) and exit.
1684 Use it in conjunction with the -n option to make sure you kill the right tinc daemon.
1685 Under native Windows the optional argument is ignored,
1686 the service will always be stopped and removed.
1688 @item -n, --net=@var{netname}
1689 Use configuration for net @var{netname}.
1690 This will let tinc read all configuration files from
1691 @file{@value{sysconfdir}/tinc/@var{netname}/}.
1692 Specifying . for @var{netname} is the same as not specifying any @var{netname}.
1693 @xref{Multiple networks}.
1695 @item -K, --generate-keys[=@var{bits}]
1696 Generate public/private keypair of @var{bits} length. If @var{bits} is not specified,
1697 2048 is the default. tinc will ask where you want to store the files,
1698 but will default to the configuration directory (you can use the -c or -n option
1699 in combination with -K). After that, tinc will quit.
1701 @item -o, --option=[@var{HOST}.]@var{KEY}=@var{VALUE}
1702 Without specifying a @var{HOST}, this will set server configuration variable @var{KEY} to @var{VALUE}.
1703 If specified as @var{HOST}.@var{KEY}=@var{VALUE},
1704 this will set the host configuration variable @var{KEY} of the host named @var{HOST} to @var{VALUE}.
1705 This option can be used more than once to specify multiple configuration variables.
1708 Lock tinc into main memory.
1709 This will prevent sensitive data like shared private keys to be written to the system swap files/partitions.
1711 @item --logfile[=@var{file}]
1712 Write log entries to a file instead of to the system logging facility.
1713 If @var{file} is omitted, the default is @file{@value{localstatedir}/log/tinc.@var{netname}.log}.
1715 @item --pidfile=@var{file}
1716 Write PID to @var{file} instead of @file{@value{localstatedir}/run/tinc.@var{netname}.pid}.
1718 @item --bypass-security
1719 Disables encryption and authentication.
1720 Only useful for debugging.
1723 Change process root directory to the directory where the config file is
1724 located (@file{@value{sysconfdir}/tinc/@var{netname}/} as determined by
1725 -n/--net option or as given by -c/--config option), for added security.
1726 The chroot is performed after all the initialization is done, after
1727 writing pid files and opening network sockets.
1729 Note that this option alone does not do any good without -U/--user, below.
1731 Note also that tinc can't run scripts anymore (such as tinc-down or host-up),
1732 unless it's setup to be runnable inside chroot environment.
1734 @item -U, --user=@var{user}
1735 Switch to the given @var{user} after initialization, at the same time as
1736 chroot is performed (see --chroot above). With this option tinc drops
1737 privileges, for added security.
1740 Display a short reminder of these runtime options and terminate.
1743 Output version information and exit.
1747 @c ==================================================================
1752 You can also send the following signals to a running tincd process:
1758 Forces tinc to try to connect to all uplinks immediately.
1759 Usually tinc attempts to do this itself,
1760 but increases the time it waits between the attempts each time it failed,
1761 and if tinc didn't succeed to connect to an uplink the first time after it started,
1762 it defaults to the maximum time of 15 minutes.
1765 Partially rereads configuration files.
1766 Connections to hosts whose host config file are removed are closed.
1767 New outgoing connections specified in @file{tinc.conf} will be made.
1768 If the --logfile option is used, this will also close and reopen the log file,
1769 useful when log rotation is used.
1772 Temporarily increases debug level to 5.
1773 Send this signal again to revert to the original level.
1776 Dumps the connection list to syslog.
1779 Dumps virtual network device statistics, all known nodes, edges and subnets to syslog.
1782 Purges all information remembered about unreachable nodes.
1786 @c ==================================================================
1788 @section Debug levels
1790 @cindex debug levels
1791 The tinc daemon can send a lot of messages to the syslog.
1792 The higher the debug level, the more messages it will log.
1793 Each level inherits all messages of the previous level:
1799 This will log a message indicating tinc has started along with a version number.
1800 It will also log any serious error.
1803 This will log all connections that are made with other tinc daemons.
1806 This will log status and error messages from scripts and other tinc daemons.
1809 This will log all requests that are exchanged with other tinc daemons. These include
1810 authentication, key exchange and connection list updates.
1813 This will log a copy of everything received on the meta socket.
1816 This will log all network traffic over the virtual private network.
1820 @c ==================================================================
1821 @node Solving problems
1822 @section Solving problems
1824 If tinc starts without problems, but if the VPN doesn't work, you will have to find the cause of the problem.
1825 The first thing to do is to start tinc with a high debug level in the foreground,
1826 so you can directly see everything tinc logs:
1829 tincd -n @var{netname} -d5 -D
1832 If tinc does not log any error messages, then you might want to check the following things:
1835 @item @file{tinc-up} script
1836 Does this script contain the right commands?
1837 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.
1840 Does the Subnet (or Subnets) in the host configuration file of this host match the portion of the VPN that belongs to this host?
1842 @item Firewalls and NATs
1843 Do you have a firewall or a NAT device (a masquerading firewall or perhaps an ADSL router that performs masquerading)?
1844 If so, check that it allows TCP and UDP traffic on port 655.
1845 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.
1846 You can add @samp{TCPOnly = yes} to your host config file to force tinc to only use a single TCP connection,
1847 this works through most firewalls and NATs. Since version 1.0.10, tinc will automatically fall back to TCP if direct communication via UDP is not possible.
1852 @c ==================================================================
1853 @node Error messages
1854 @section Error messages
1856 What follows is a list of the most common error messages you might find in the logs.
1857 Some of them will only be visible if the debug level is high enough.
1860 @item Could not open /dev/tap0: No such device
1863 @item You forgot to `modprobe netlink_dev' or `modprobe ethertap'.
1864 @item You forgot to compile `Netlink device emulation' in the kernel.
1867 @item Can't write to /dev/net/tun: No such device
1870 @item You forgot to `modprobe tun'.
1871 @item You forgot to compile `Universal TUN/TAP driver' in the kernel.
1872 @item The tun device is located somewhere else in @file{/dev/}.
1875 @item Network address and prefix length do not match!
1878 @item The Subnet field must contain a @emph{network} address, trailing bits should be 0.
1879 @item If you only want to use one IP address, set the netmask to /32.
1882 @item Error reading RSA key file `rsa_key.priv': No such file or directory
1885 @item You forgot to create a public/private keypair.
1886 @item Specify the complete pathname to the private key file with the @samp{PrivateKeyFile} option.
1889 @item Warning: insecure file permissions for RSA private key file `rsa_key.priv'!
1892 @item The private key file is readable by users other than root.
1893 Use chmod to correct the file permissions.
1896 @item Creating metasocket failed: Address family not supported
1899 @item By default tinc tries to create both IPv4 and IPv6 sockets.
1900 On some platforms this might not be implemented.
1901 If the logs show @samp{Ready} later on, then at least one metasocket was created,
1902 and you can ignore this message.
1903 You can add @samp{AddressFamily = ipv4} to @file{tinc.conf} to prevent this from happening.
1906 @item Cannot route packet: unknown IPv4 destination 1.2.3.4
1909 @item You try to send traffic to a host on the VPN for which no Subnet is known.
1910 @item If it is a broadcast address (ending in .255), it probably is a samba server or a Windows host sending broadcast packets.
1914 @item Cannot route packet: ARP request for unknown address 1.2.3.4
1917 @item You try to send traffic to a host on the VPN for which no Subnet is known.
1920 @item Packet with destination 1.2.3.4 is looping back to us!
1923 @item Something is not configured right. Packets are being sent out to the
1924 virtual network device, but according to the Subnet directives in your host configuration
1925 file, those packets should go to your own host. Most common mistake is that
1926 you have a Subnet line in your host configuration file with a prefix length which is
1927 just as large as the prefix of the virtual network interface. The latter should in almost all
1928 cases be larger. Rethink your configuration.
1929 Note that you will only see this message if you specified a debug
1930 level of 5 or higher!
1931 @item Chances are that a @samp{Subnet = ...} line in the host configuration file of this tinc daemon is wrong.
1932 Change it to a subnet that is accepted locally by another interface,
1933 or if that is not the case, try changing the prefix length into /32.
1936 @item Node foo (1.2.3.4) is not reachable
1939 @item Node foo does not have a connection anymore, its tinc daemon is not running or its connection to the Internet is broken.
1942 @item Received UDP packet from unknown source 1.2.3.4 (port 12345)
1945 @item If you see this only sporadically, it is harmless and caused by a node sending packets using an old key.
1948 @item Got bad/bogus/unauthorized REQUEST from foo (1.2.3.4 port 12345)
1951 @item Node foo does not have the right public/private keypair.
1952 Generate new keypairs and distribute them again.
1953 @item An attacker tries to gain access to your VPN.
1954 @item A network error caused corruption of metadata sent from foo.
1959 @c ==================================================================
1960 @node Sending bug reports
1961 @section Sending bug reports
1963 If you really can't find the cause of a problem, or if you suspect tinc is not working right,
1964 you can send us a bugreport, see @ref{Contact information}.
1965 Be sure to include the following information in your bugreport:
1968 @item A clear description of what you are trying to achieve and what the problem is.
1969 @item What platform (operating system, version, hardware architecture) and which version of tinc you use.
1970 @item If compiling tinc fails, a copy of @file{config.log} and the error messages you get.
1971 @item Otherwise, a copy of @file{tinc.conf}, @file{tinc-up} and all files in the @file{hosts/} directory.
1972 @item The output of the commands @samp{ifconfig -a} and @samp{route -n} (or @samp{netstat -rn} if that doesn't work).
1973 @item The output of any command that fails to work as it should (like ping or traceroute).
1976 @c ==================================================================
1977 @node Technical information
1978 @chapter Technical information
1983 * The meta-protocol::
1988 @c ==================================================================
1989 @node The connection
1990 @section The connection
1993 Tinc is a daemon that takes VPN data and transmit that to another host
1994 computer over the existing Internet infrastructure.
1998 * The meta-connection::
2002 @c ==================================================================
2003 @node The UDP tunnel
2004 @subsection The UDP tunnel
2006 @cindex virtual network device
2008 The data itself is read from a character device file, the so-called
2009 @emph{virtual network device}. This device is associated with a network
2010 interface. Any data sent to this interface can be read from the device,
2011 and any data written to the device gets sent from the interface.
2012 There are two possible types of virtual network devices:
2013 `tun' style, which are point-to-point devices which can only handle IPv4 and/or IPv6 packets,
2014 and `tap' style, which are Ethernet devices and handle complete Ethernet frames.
2016 So when tinc reads an Ethernet frame from the device, it determines its
2017 type. When tinc is in its default routing mode, it can handle IPv4 and IPv6
2018 packets. Depending on the Subnet lines, it will send the packets off to their destination IP address.
2019 In the `switch' and `hub' mode, tinc will use broadcasts and MAC address discovery
2020 to deduce the destination of the packets.
2021 Since the latter modes only depend on the link layer information,
2022 any protocol that runs over Ethernet is supported (for instance IPX and Appletalk).
2023 However, only `tap' style devices provide this information.
2025 After the destination has been determined,
2026 the packet will be compressed (optionally),
2027 a sequence number will be added to the packet,
2028 the packet will then be encrypted
2029 and a message authentication code will be appended.
2031 @cindex encapsulating
2033 When that is done, time has come to actually transport the
2034 packet to the destination computer. We do this by sending the packet
2035 over an UDP connection to the destination host. This is called
2036 @emph{encapsulating}, the VPN packet (though now encrypted) is
2037 encapsulated in another IP datagram.
2039 When the destination receives this packet, the same thing happens, only
2040 in reverse. So it checks the message authentication code, decrypts the contents of the UDP datagram,
2041 checks the sequence number
2042 and writes the decrypted information to its own virtual network device.
2044 If the virtual network device is a `tun' device (a point-to-point tunnel),
2045 there is no problem for the kernel to accept a packet.
2046 However, if it is a `tap' device (this is the only available type on FreeBSD),
2047 the destination MAC address must match that of the virtual network interface.
2048 If tinc is in its default routing mode, ARP does not work, so the correct destination MAC
2049 can not be known by the sending host.
2050 Tinc solves this by letting the receiving end detect the MAC address of its own virtual network interface
2051 and overwriting the destination MAC address of the received packet.
2053 In switch or hub modes ARP does work so the sender already knows the correct destination MAC address.
2054 In those modes every interface should have a unique MAC address, so make sure they are not the same.
2055 Because switch and hub modes rely on MAC addresses to function correctly,
2056 these modes cannot be used on the following operating systems which don't have a `tap' style virtual network device:
2057 OpenBSD, NetBSD, Darwin and Solaris.
2060 @c ==================================================================
2061 @node The meta-connection
2062 @subsection The meta-connection
2064 Having only a UDP connection available is not enough. Though suitable
2065 for transmitting data, we want to be able to reliably send other
2066 information, such as routing and session key information to somebody.
2069 TCP is a better alternative, because it already contains protection
2070 against information being lost, unlike UDP.
2072 So we establish two connections. One for the encrypted VPN data, and one
2073 for other information, the meta-data. Hence, we call the second
2074 connection the meta-connection. We can now be sure that the
2075 meta-information doesn't get lost on the way to another computer.
2077 @cindex data-protocol
2078 @cindex meta-protocol
2079 Like with any communication, we must have a protocol, so that everybody
2080 knows what everything stands for, and how she should react. Because we
2081 have two connections, we also have two protocols. The protocol used for
2082 the UDP data is the ``data-protocol,'' the other one is the
2085 The reason we don't use TCP for both protocols is that UDP is much
2086 better for encapsulation, even while it is less reliable. The real
2087 problem is that when TCP would be used to encapsulate a TCP stream
2088 that's on the private network, for every packet sent there would be
2089 three ACKs sent instead of just one. Furthermore, if there would be
2090 a timeout, both TCP streams would sense the timeout, and both would
2091 start re-sending packets.
2094 @c ==================================================================
2095 @node The meta-protocol
2096 @section The meta-protocol
2098 The meta protocol is used to tie all tinc daemons together, and
2099 exchange information about which tinc daemon serves which virtual
2102 The meta protocol consists of requests that can be sent to the other
2103 side. Each request has a unique number and several parameters. All
2104 requests are represented in the standard ASCII character set. It is
2105 possible to use tools such as telnet or netcat to connect to a tinc
2106 daemon started with the --bypass-security option
2107 and to read and write requests by hand, provided that one
2108 understands the numeric codes sent.
2110 The authentication scheme is described in @ref{Authentication protocol}. After a
2111 successful authentication, the server and the client will exchange all the
2112 information about other tinc daemons and subnets they know of, so that both
2113 sides (and all the other tinc daemons behind them) have their information
2120 ------------------------------------------------------------------
2121 ADD_EDGE node1 node2 21.32.43.54 655 222 0
2122 | | | | | +-> options
2123 | | | | +----> weight
2124 | | | +--------> UDP port of node2
2125 | | +----------------> real address of node2
2126 | +-------------------------> name of destination node
2127 +-------------------------------> name of source node
2129 ADD_SUBNET node 192.168.1.0/24
2130 | | +--> prefixlength
2131 | +--------> network address
2132 +------------------> owner of this subnet
2133 ------------------------------------------------------------------
2136 The ADD_EDGE messages are to inform other tinc daemons that a connection between
2137 two nodes exist. The address of the destination node is available so that
2138 VPN packets can be sent directly to that node.
2140 The ADD_SUBNET messages inform other tinc daemons that certain subnets belong
2141 to certain nodes. tinc will use it to determine to which node a VPN packet has
2148 ------------------------------------------------------------------
2149 DEL_EDGE node1 node2
2150 | +----> name of destination node
2151 +----------> name of source node
2153 DEL_SUBNET node 192.168.1.0/24
2154 | | +--> prefixlength
2155 | +--------> network address
2156 +------------------> owner of this subnet
2157 ------------------------------------------------------------------
2160 In case a connection between two daemons is closed or broken, DEL_EDGE messages
2161 are sent to inform the other daemons of that fact. Each daemon will calculate a
2162 new route to the the daemons, or mark them unreachable if there isn't any.
2169 ------------------------------------------------------------------
2170 REQ_KEY origin destination
2171 | +--> name of the tinc daemon it wants the key from
2172 +----------> name of the daemon that wants the key
2174 ANS_KEY origin destination 4ae0b0a82d6e0078 91 64 4
2175 | | \______________/ | | +--> MAC length
2176 | | | | +-----> digest algorithm
2177 | | | +--------> cipher algorithm
2178 | | +--> 128 bits key
2179 | +--> name of the daemon that wants the key
2180 +----------> name of the daemon that uses this key
2183 +--> daemon that has changed it's packet key
2184 ------------------------------------------------------------------
2187 The keys used to encrypt VPN packets are not sent out directly. This is
2188 because it would generate a lot of traffic on VPNs with many daemons, and
2189 chances are that not every tinc daemon will ever send a packet to every
2190 other daemon. Instead, if a daemon needs a key it sends a request for it
2191 via the meta connection of the nearest hop in the direction of the
2198 ------------------------------------------------------------------
2201 ------------------------------------------------------------------
2204 There is also a mechanism to check if hosts are still alive. Since network
2205 failures or a crash can cause a daemon to be killed without properly
2206 shutting down the TCP connection, this is necessary to keep an up to date
2207 connection list. PINGs are sent at regular intervals, except when there
2208 is also some other traffic. A little bit of salt (random data) is added
2209 with each PING and PONG message, to make sure that long sequences of PING/PONG
2210 messages without any other traffic won't result in known plaintext.
2212 This basically covers what is sent over the meta connection by tinc.
2215 @c ==================================================================
2221 Tinc got its name from ``TINC,'' short for @emph{There Is No Cabal}; the
2222 alleged Cabal was/is an organisation that was said to keep an eye on the
2223 entire Internet. As this is exactly what you @emph{don't} want, we named
2224 the tinc project after TINC.
2227 But in order to be ``immune'' to eavesdropping, you'll have to encrypt
2228 your data. Because tinc is a @emph{Secure} VPN (SVPN) daemon, it does
2229 exactly that: encrypt.
2230 Tinc by default uses blowfish encryption with 128 bit keys in CBC mode, 32 bit
2231 sequence numbers and 4 byte long message authentication codes to make sure
2232 eavesdroppers cannot get and cannot change any information at all from the
2233 packets they can intercept. The encryption algorithm and message authentication
2234 algorithm can be changed in the configuration. The length of the message
2235 authentication codes is also adjustable. The length of the key for the
2236 encryption algorithm is always the default length used by OpenSSL.
2239 * Authentication protocol::
2240 * Encryption of network packets::
2245 @c ==================================================================
2246 @node Authentication protocol
2247 @subsection Authentication protocol
2249 @cindex authentication
2250 A new scheme for authentication in tinc has been devised, which offers some
2251 improvements over the protocol used in 1.0pre2 and 1.0pre3. Explanation is
2261 --------------------------------------------------------------------------
2262 client <attempts connection>
2264 server <accepts connection>
2268 +-------> name of tinc daemon
2272 +-------> name of tinc daemon
2274 client META_KEY 5f0823a93e35b69e...7086ec7866ce582b
2275 \_________________________________/
2276 +-> RSAKEYLEN bits totally random string S1,
2277 encrypted with server's public RSA key
2279 server META_KEY 6ab9c1640388f8f0...45d1a07f8a672630
2280 \_________________________________/
2281 +-> RSAKEYLEN bits totally random string S2,
2282 encrypted with client's public RSA key
2285 - the client will symmetrically encrypt outgoing traffic using S1
2286 - the server will symmetrically encrypt outgoing traffic using S2
2288 client CHALLENGE da02add1817c1920989ba6ae2a49cecbda0
2289 \_________________________________/
2290 +-> CHALLEN bits totally random string H1
2292 server CHALLENGE 57fb4b2ccd70d6bb35a64c142f47e61d57f
2293 \_________________________________/
2294 +-> CHALLEN bits totally random string H2
2296 client CHAL_REPLY 816a86
2297 +-> 160 bits SHA1 of H2
2299 server CHAL_REPLY 928ffe
2300 +-> 160 bits SHA1 of H1
2302 After the correct challenge replies are received, both ends have proved
2303 their identity. Further information is exchanged.
2305 client ACK 655 123 0
2307 | +----> estimated weight
2308 +--------> listening port of client
2310 server ACK 655 321 0
2312 | +----> estimated weight
2313 +--------> listening port of server
2314 --------------------------------------------------------------------------
2317 This new scheme has several improvements, both in efficiency and security.
2319 First of all, the server sends exactly the same kind of messages over the wire
2320 as the client. The previous versions of tinc first authenticated the client,
2321 and then the server. This scheme even allows both sides to send their messages
2322 simultaneously, there is no need to wait for the other to send something first.
2323 This means that any calculations that need to be done upon sending or receiving
2324 a message can also be done in parallel. This is especially important when doing
2325 RSA encryption/decryption. Given that these calculations are the main part of
2326 the CPU time spent for the authentication, speed is improved by a factor 2.
2328 Second, only one RSA encrypted message is sent instead of two. This reduces the
2329 amount of information attackers can see (and thus use for a cryptographic
2330 attack). It also improves speed by a factor two, making the total speedup a
2333 Third, and most important:
2334 The symmetric cipher keys are exchanged first, the challenge is done
2335 afterwards. In the previous authentication scheme, because a man-in-the-middle
2336 could pass the challenge/chal_reply phase (by just copying the messages between
2337 the two real tinc daemons), but no information was exchanged that was really
2338 needed to read the rest of the messages, the challenge/chal_reply phase was of
2339 no real use. The man-in-the-middle was only stopped by the fact that only after
2340 the ACK messages were encrypted with the symmetric cipher. Potentially, it
2341 could even send it's own symmetric key to the server (if it knew the server's
2342 public key) and read some of the metadata the server would send it (it was
2343 impossible for the mitm to read actual network packets though). The new scheme
2344 however prevents this.
2346 This new scheme makes sure that first of all, symmetric keys are exchanged. The
2347 rest of the messages are then encrypted with the symmetric cipher. Then, each
2348 side can only read received messages if they have their private key. The
2349 challenge is there to let the other side know that the private key is really
2350 known, because a challenge reply can only be sent back if the challenge is
2351 decrypted correctly, and that can only be done with knowledge of the private
2354 Fourth: the first thing that is sent via the symmetric cipher encrypted
2355 connection is a totally random string, so that there is no known plaintext (for
2356 an attacker) in the beginning of the encrypted stream.
2359 @c ==================================================================
2360 @node Encryption of network packets
2361 @subsection Encryption of network packets
2364 A data packet can only be sent if the encryption key is known to both
2365 parties, and the connection is activated. If the encryption key is not
2366 known, a request is sent to the destination using the meta connection
2367 to retrieve it. The packet is stored in a queue while waiting for the
2371 The UDP packet containing the network packet from the VPN has the following layout:
2374 ... | IP header | UDP header | seqno | VPN packet | MAC | UDP trailer
2375 \___________________/\_____/
2377 V +---> digest algorithm
2378 Encrypted with symmetric cipher
2381 So, the entire VPN packet is encrypted using a symmetric cipher, including a 32 bits
2382 sequence number that is added in front of the actual VPN packet, to act as a unique
2383 IV for each packet and to prevent replay attacks. A message authentication code
2384 is added to the UDP packet to prevent alteration of packets. By default the
2385 first 4 bytes of the digest are used for this, but this can be changed using
2386 the MACLength configuration variable.
2388 @c ==================================================================
2389 @node Security issues
2390 @subsection Security issues
2392 In August 2000, we discovered the existence of a security hole in all versions
2393 of tinc up to and including 1.0pre2. This had to do with the way we exchanged
2394 keys. Since then, we have been working on a new authentication scheme to make
2395 tinc as secure as possible. The current version uses the OpenSSL library and
2396 uses strong authentication with RSA keys.
2398 On the 29th of December 2001, Jerome Etienne posted a security analysis of tinc
2399 1.0pre4. Due to a lack of sequence numbers and a message authentication code
2400 for each packet, an attacker could possibly disrupt certain network services or
2401 launch a denial of service attack by replaying intercepted packets. The current
2402 version adds sequence numbers and message authentication codes to prevent such
2405 On the 15th of September 2003, Peter Gutmann posted a security analysis of tinc
2406 1.0.1. He argues that the 32 bit sequence number used by tinc is not a good IV,
2407 that tinc's default length of 4 bytes for the MAC is too short, and he doesn't
2408 like tinc's use of RSA during authentication. We do not know of a security hole
2409 in this version of tinc, but tinc's security is not as strong as TLS or IPsec.
2410 We will address these issues in tinc 2.0.
2412 Cryptography is a hard thing to get right. We cannot make any
2413 guarantees. Time, review and feedback are the only things that can
2414 prove the security of any cryptographic product. If you wish to review
2415 tinc or give us feedback, you are stronly encouraged to do so.
2418 @c ==================================================================
2419 @node Platform specific information
2420 @chapter Platform specific information
2423 * Interface configuration::
2427 @c ==================================================================
2428 @node Interface configuration
2429 @section Interface configuration
2431 When configuring an interface, one normally assigns it an address and a
2432 netmask. The address uniquely identifies the host on the network attached to
2433 the interface. The netmask, combined with the address, forms a subnet. It is
2434 used to add a route to the routing table instructing the kernel to send all
2435 packets which fall into that subnet to that interface. Because all packets for
2436 the entire VPN should go to the virtual network interface used by tinc, the
2437 netmask should be such that it encompasses the entire VPN.
2441 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2443 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2444 @item Linux iproute2
2445 @tab @code{ip addr add} @var{address}@code{/}@var{prefixlength} @code{dev} @var{interface}
2447 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2449 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2451 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2453 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2454 @item Darwin (Mac OS X)
2455 @tab @code{ifconfig} @var{interface} @var{address} @code{netmask} @var{netmask}
2457 @tab @code{netsh interface ip set address} @var{interface} @code{static} @var{address} @var{netmask}
2462 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2464 @tab @code{ifconfig} @var{interface} @code{add} @var{address}@code{/}@var{prefixlength}
2466 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
2468 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
2470 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
2472 @tab @code{ifconfig} @var{interface} @code{inet6 plumb up}
2474 @tab @code{ifconfig} @var{interface} @code{inet6 addif} @var{address} @var{address}
2475 @item Darwin (Mac OS X)
2476 @tab @code{ifconfig} @var{interface} @code{inet6} @var{address} @code{prefixlen} @var{prefixlength}
2478 @tab @code{netsh interface ipv6 add address} @var{interface} @code{static} @var{address}/@var{prefixlength}
2481 On some platforms, when running tinc in switch mode, the VPN interface must be set to tap mode with an ifconfig command:
2483 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2485 @tab @code{ifconfig} @var{interface} @code{link0}
2488 On Linux, it is possible to create a persistent tun/tap interface which will
2489 continue to exist even if tinc quit, although this is normally not required.
2490 It can be useful to set up a tun/tap interface owned by a non-root user, so
2491 tinc can be started without needing any root privileges at all.
2493 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2495 @tab @code{ip tuntap add dev} @var{interface} @code{mode} @var{tun|tap} @code{user} @var{username}
2498 @c ==================================================================
2502 In some cases it might be necessary to add more routes to the virtual network
2503 interface. There are two ways to indicate which interface a packet should go
2504 to, one is to use the name of the interface itself, another way is to specify
2505 the (local) address that is assigned to that interface (@var{local_address}). The
2506 former way is unambiguous and therefore preferable, but not all platforms
2509 Adding routes to IPv4 subnets:
2511 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2513 @tab @code{route add -net} @var{network_address} @code{netmask} @var{netmask} @var{interface}
2514 @item Linux iproute2
2515 @tab @code{ip route add} @var{network_address}@code{/}@var{prefixlength} @code{dev} @var{interface}
2517 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
2519 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
2521 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
2523 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @var{local_address} @code{-interface}
2524 @item Darwin (Mac OS X)
2525 @tab @code{route add} @var{network_address}@code{/}@var{prefixlength} @code{-interface} @var{interface}
2527 @tab @code{netsh routing ip add persistentroute} @var{network_address} @var{netmask} @var{interface} @var{local_address}
2530 Adding routes to IPv6 subnets:
2532 @multitable {Darwin (Mac OS X)} {ifconfig route add -bla network address netmask netmask prefixlength interface}
2534 @tab @code{route add -A inet6} @var{network_address}@code{/}@var{prefixlength} @var{interface}
2535 @item Linux iproute2
2536 @tab @code{ip route add} @var{network_address}@code{/}@var{prefixlength} @code{dev} @var{interface}
2538 @tab @code{route add -inet6} @var{network_address}@code{/}@var{prefixlength} @var{local_address}
2540 @tab @code{route add -inet6} @var{network_address} @var{local_address} @code{-prefixlen} @var{prefixlength}
2542 @tab @code{route add -inet6} @var{network_address} @var{local_address} @code{-prefixlen} @var{prefixlength}
2544 @tab @code{route add -inet6} @var{network_address}@code{/}@var{prefixlength} @var{local_address} @code{-interface}
2545 @item Darwin (Mac OS X)
2546 @tab @code{route add -inet6} @var{network_address}@code{/}@var{prefixlength} @code{-interface} @var{interface}
2548 @tab @code{netsh interface ipv6 add route} @var{network address}/@var{prefixlength} @var{interface}
2552 @c ==================================================================
2558 * Contact information::
2563 @c ==================================================================
2564 @node Contact information
2565 @section Contact information
2568 Tinc's website is at @url{http://www.tinc-vpn.org/},
2569 this server is located in the Netherlands.
2572 We have an IRC channel on the FreeNode and OFTC IRC networks. Connect to
2573 @uref{http://www.freenode.net/, irc.freenode.net}
2575 @uref{http://www.oftc.net/, irc.oftc.net}
2576 and join channel #tinc.
2579 @c ==================================================================
2584 @item Ivo Timmermans (zarq)
2585 @item Guus Sliepen (guus) (@email{guus@@tinc-vpn.org})
2588 We have received a lot of valuable input from users. With their help,
2589 tinc has become the flexible and robust tool that it is today. We have
2590 composed a list of contributions, in the file called @file{THANKS} in
2591 the source distribution.
2594 @c ==================================================================
2596 @unnumbered Concept Index
2598 @c ==================================================================
2602 @c ==================================================================