ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/gvpe/doc/gvpe.conf.5.pod
Revision: 1.24
Committed: Sat Jul 18 05:59:16 2009 UTC (14 years, 10 months ago) by pcg
Branch: MAIN
Changes since 1.23: +15 -5 lines
Log Message:
riddify us of meta.yml garbage in manifest

File Contents

# User Rev Content
1 pcg 1.1 =head1 NAME
2    
3     gvpe.conf - configuration file for the GNU VPE daemon
4    
5     =head1 SYNOPSIS
6    
7 pcg 1.21 # global options for all nodes
8 pcg 1.1 udp-port = 407
9     mtu = 1492
10     ifname = vpn0
11    
12 pcg 1.21 # first node is named branch1 and is at 1.2.3.4
13 pcg 1.1 node = branch1
14     hostname = 1.2.3.4
15    
16 pcg 1.21 # second node uses dns to resolve the address
17 pcg 1.1 node = branch2
18     hostname = www.example.net
19     udp-port = 500 # this host uses a different udp-port
20    
21 pcg 1.21 # third node has no fixed ip address
22 pcg 1.1 node = branch3
23     connect = ondemand
24    
25     =head1 DESCRIPTION
26    
27     The gvpe config file consists of a series of lines that contain C<variable
28     = value> pairs. Empty lines are ignored. Comments start with a C<#> and
29     extend to the end of the line. They can be used on their own lines, or
30 pcg 1.13 after any directives. Whitespace is allowed around the C<=> sign or after
31     values, but not within the variable names or values themselves.
32 pcg 1.1
33     The only exception to the above is the "on" directive that can prefix any
34     C<name = value> setting and will only "execute" it on the named node, or
35     (if the nodename starts with "!") on all nodes except the named one.
36    
37 pcg 1.21 For example, set the MTU to C<1450> everywhere, loglevel to C<noise> on
38     branch1, and connect to C<ondemand> everywhere but on branch2:
39    
40     mtu = 1450
41 pcg 1.1 on branch1 loglevel = noise
42     on !branch2 connect = ondemand
43    
44 pcg 1.21 All settings are applied "in order", that is, later settings of the same
45 pcg 1.1 variable overwrite earlier ones.
46    
47     =head1 ANATOMY OF A CONFIG FILE
48    
49 pcg 1.21 Usually, a config file starts with a few global settings (like the UDP
50     port to listen on), followed by node-specific sections that begin with a
51     C<node = nickname> line.
52 pcg 1.1
53     Every node that is part of the network must have a section that starts
54     with C<node = nickname>. The number and order of the nodes is important
55 pcg 1.20 and must be the same on all nodes. It is not uncommon for node sections to
56 pcg 1.1 be completely empty - if the default values are right.
57    
58     Node-specific settings can be used at any time. If used before the first
59     node section they will set the default values for all following nodes.
60    
61     =head1 CONFIG VARIABLES
62    
63     =head2 GLOBAL SETTINGS
64    
65     Global settings will affect the behaviour of the running gvpe daemon, that
66     is, they are in some sense node-specific (config files can set different
67     values on different nodes using C<on>), but will affect the behaviour of
68     the gvpe daemon and all connections it creates.
69    
70     =over 4
71    
72 pcg 1.6 =item dns-forw-host = hostname/ip
73 pcg 1.1
74 pcg 1.21 The DNS server to forward DNS requests to for the DNS tunnel protocol
75 pcg 1.6 (default: C<127.0.0.1>, changing it is highly recommended).
76 pcg 1.1
77 pcg 1.6 =item dns-forw-port = port-number
78 pcg 1.1
79 pcg 1.6 The port where the C<dns-forw-host> is to be contacted (default: C<53>,
80     which is fine in most cases).
81 pcg 1.1
82 pcg 1.12 =item dns-max-outstanding = integer-number-of-requests
83    
84     The maximum number of outstanding DNS transport requests
85     (default: C<100>). GVPE will never issue more requests then the given
86     limit without receiving replies. In heavily overloaded situations it might
87     help to set this to a low number (e.g. C<3> or even C<1>) to limit the
88     number of parallel requests.
89    
90 pcg 1.21 The default should be working OK for most links.
91 pcg 1.12
92     =item dns-overlap-factor = float
93    
94     The DNS transport uses the minimum request latency (B<min_latency>) seen
95     during a connection as it's timing base. This factor (default: C<0.5>,
96     must be > 0) is multiplied by B<min_latency> to get the maximum sending
97     rate (= minimum send interval), i.e. a factor of C<1> means that a new
98     request might be generated every B<min_latency> seconds, which means on
99     average there should only ever be one outstanding request. A factor of
100     C<0.5> means that GVPE will send requests twice as often as the minimum
101     latency measured.
102    
103 pcg 1.21 For congested or picky DNS forwarders you could use a value nearer to or
104 pcg 1.12 exceeding C<1>.
105    
106 pcg 1.21 The default should be working OK for most links.
107 pcg 1.12
108     =item dns-send-interval = send-interval-in-seconds
109    
110     The minimum send interval (= maximum rate) that the DNS transport will
111     use to send new DNS requests. GVPE will not exceed this rate even when
112     the latency is very low. The default is C<0.01>, which means GVPE will
113     not send more than 100 DNS requests per connection per second. For
114     high-bandwidth links you could go lower, e.g. to C<0.001> or so. For
115     congested or rate-limited links, you might want to go higher, say C<0.1>,
116     C<0.2> or even higher.
117    
118 pcg 1.21 The default should be working OK for most links.
119 pcg 1.12
120     =item dns-timeout-factor = float
121    
122     Factor to multiply the C<min_latency> (see C<dns-overlap-factor>) by to
123     get request timeouts. The default of C<8> means that the DNS transport
124     will resend the request when no reply has been received for longer than
125     eight times the minimum (= expected) latency, assuming the request or
126     reply has been lost.
127    
128 pcg 1.17 For congested links a higher value might be necessary (e.g. C<30>). If
129     the link is very stable lower values (e.g. C<2>) might work
130     nicely. Values near or below C<1> makes no sense whatsoever.
131 pcg 1.12
132 pcg 1.21 The default should be working OK for most links but will result in low
133 pcg 1.17 throughput if packet loss is high.
134 pcg 1.12
135 pcg 1.1 =item if-up = relative-or-absolute-path
136    
137     Sets the path of a script that should be called immediately after the
138 pcg 1.21 network interface is initialized (but not necessarily up). The following
139 pcg 1.13 environment variables are passed to it (the values are just examples).
140    
141     Variables that have the same value on all nodes:
142 pcg 1.1
143     =over 4
144    
145     =item CONFBASE=/etc/gvpe
146    
147     The configuration base directory.
148    
149     =item IFNAME=vpn0
150    
151 pcg 1.13 The network interface to initialize.
152    
153     =item IFTYPE=native # or tincd
154    
155     =item IFSUBTYPE=linux # or freebsd, darwin etc..
156    
157     The interface type (C<native> or C<tincd>) and the subtype (usually the
158     OS name in lowercase) that this GVPE was configured for. Can be used to
159     select the correct syntax to use for network-related commands.
160 pcg 1.1
161     =item MTU=1436
162    
163     The MTU to set the interface to. You can use lower values (if done
164 pcg 1.20 consistently on all nodes), but this is usually either inefficient or
165     simply ineffective.
166 pcg 1.1
167 pcg 1.13 =item NODES=5
168    
169     The number of nodes in this GVPE network.
170    
171     =back
172    
173     Variables that are node-specific and with values pertaining to the node
174     running this GVPE:
175 pcg 1.1
176 pcg 1.13 =over 4
177 pcg 1.1
178 pcg 1.13 =item IFUPDATA=string
179 pcg 1.1
180 pcg 1.13 The value of the configuration directive C<if-up-data>.
181 pcg 1.1
182 pcg 1.13 =item MAC=fe:fd:80:00:00:01
183 pcg 1.1
184 pcg 1.13 The MAC address the network interface has to use.
185 pcg 1.1
186 pcg 1.13 Might be used to initialize interfaces on platforms where GVPE does not
187 pcg 1.21 do this automatically. Please see the C<gvpe.osdep(5)> man page for
188 pcg 1.13 platform-specific information.
189 pcg 1.1
190     =item NODENAME=branch1
191    
192 pcg 1.13 The nickname of the node.
193 pcg 1.1
194     =item NODEID=1
195    
196 pcg 1.13 The numerical node ID of the node running this instance of GVPE. The first
197     node mentioned in the config file gets ID 1, the second ID 2 and so on.
198 pcg 1.1
199     =back
200    
201 pcg 1.13 In addition, all node-specific variables (except C<NODEID>) will be
202     available with a postfix of C<_nodeid>, which contains the value for that
203     node, e.g. the C<MAC_1> variable contains the MAC address of node #1, while
204     the C<NODENAME_22> variable contains the name of node #22.
205    
206 pcg 1.1 Here is a simple if-up script:
207    
208     #!/bin/sh
209 pcg 1.13 ip link set $IFNAME up
210 pcg 1.1 [ $NODENAME = branch1 ] && ip addr add 10.0.0.1 dev $IFNAME
211     [ $NODENAME = branch2 ] && ip addr add 10.1.0.1 dev $IFNAME
212     ip route add 10.0.0.0/8 dev $IFNAME
213    
214 pcg 1.21 More complicated examples (using routing to reduce ARP traffic) can be
215     found in the F<etc/> subdirectory of the distribution.
216 pcg 1.1
217 pcg 1.6 =item ifname = devname
218    
219     Sets the tun interface name to the given name. The default is OS-specific
220     and most probably something like C<tun0>.
221    
222     =item ifpersist = yes|true|on | no|false|off
223    
224     Should the tun/tap device be made persistent, that is, should the device
225     stay up even when gvpe exits? Some versions of the tunnel device have
226     problems sending packets when gvpe is restarted in persistent mode, so
227     if the connections can be established but you cannot send packets from
228     the local node, try to set this to C<off> and do an ifconfig down on the
229     device.
230    
231     =item ip-proto = numerical-ip-protocol
232    
233     Sets the protocol number to be used for the rawip protocol. This is a
234 pcg 1.20 global option because all nodes must use the same protocol, and since
235 pcg 1.6 there are no port numbers, you cannot easily run more than one gvpe
236     instance using the same protocol, nor can you share the protocol with
237     other programs.
238    
239 pcg 1.21 The default is 47 (GRE), which has a good chance of tunneling
240     through firewalls (but note that gvpe's rawip protocol is not GRE
241     compatible). Other common choices are 50 (IPSEC, ESP), 51 (IPSEC, AH), 4
242     (IPIP tunnels) or 98 (ENCAP, rfc1241)
243 pcg 1.6
244     =item http-proxy-host = hostname/ip
245    
246     The C<http-proxy-*> family of options are only available if gvpe was
247     compiled with the C<--enable-http-proxy> option and enable tunneling of
248     tcp connections through a http proxy server.
249    
250     C<http-proxy-host> and C<http-proxy-port> should specify the hostname and
251     port number of the proxy server. See C<http-proxy-loginpw> if your proxy
252     requires authentication.
253    
254     Please note that gvpe will still try to resolve all hostnames in the
255 pcg 1.21 configuration file, so if you are behind a proxy without access to a DNS
256 pcg 1.6 server better use numerical IP addresses.
257    
258 pcg 1.21 To make best use of this option disable all protocols except TCP in your
259 pcg 1.20 config file and make sure your routers (or all other nodes) are listening
260 pcg 1.6 on a port that the proxy allows (443, https, is a common choice).
261    
262 pcg 1.21 If you have a router, connecting to it will suffice. Otherwise TCP must be
263 pcg 1.20 enabled on all nodes.
264 pcg 1.6
265     Example:
266    
267     http-proxy-host = proxy.example.com
268     http-proxy-port = 3128 # 8080 is another common choice
269     http-proxy-auth = schmorp:grumbeere
270    
271     =item http-proxy-port = proxy-tcp-port
272    
273     The port where your proxy server listens.
274    
275     =item http-proxy-auth = login:password
276    
277     The optional login and password used to authenticate to the proxy server,
278 pcg 1.21 separated by a literal colon (C<:>). Only basic authentication is
279 pcg 1.6 currently supported.
280    
281     =item keepalive = seconds
282    
283     Sets the keepalive probe interval in seconds (default: C<60>). After this
284     many seconds of inactivity the daemon will start to send keepalive probe
285 pcg 1.21 every 3 seconds until it receives a reply from the other end. If no reply
286     is received within 15 seconds, the peer is considered unreachable and the
287 pcg 1.6 connection is closed.
288    
289     =item loglevel = noise|trace|debug|info|notice|warn|error|critical
290    
291     Set the logging level. Connection established messages are logged at level
292     C<info>, notable errors are logged with C<error>. Default is C<info>.
293    
294     =item mtu = bytes
295    
296     Sets the maximum MTU that should be used on outgoing packets (basically
297     the MTU of the outgoing interface) The daemon will automatically calculate
298 pcg 1.21 maximum overhead (e.g. UDP header size, encryption blocksize...) and pass
299 pcg 1.6 this information to the C<if-up> script.
300    
301     Recommended values are 1500 (ethernet), 1492 (pppoe), 1472 (pptp).
302    
303 pcg 1.21 This value must be the minimum of the MTU values of all nodes.
304 pcg 1.6
305     =item node = nickname
306    
307     Not really a config setting but introduces a node section. The nickname is
308     used to select the right configuration section and must be passed as an
309     argument to the gvpe daemon.
310    
311 pcg 1.1 =item node-up = relative-or-absolute-path
312    
313 pcg 1.19 Sets a command (default: none) that should be called whenever a connection
314     is established (even on rekeying operations). Note that node-up/down
315     scripts will be run asynchronously, but execution is serialised, so there
316     will only ever be one such script running.
317    
318     In addition to all the variables passed to C<if-up> scripts, the following
319 pcg 1.24 environment variables will be set (values are just examples):
320 pcg 1.1
321     =over 4
322    
323     =item DESTNODE=branch2
324    
325     The name of the remote node.
326    
327     =item DESTID=2
328    
329     The node id of the remote node.
330    
331 pcg 1.24 =item DESTSI=rawip/88.99.77.55:0
332    
333     The "socket info" of the target node, protocol dependent but usually in
334     the format protocol/ip:port.
335    
336 pcg 1.1 =item DESTIP=188.13.66.8
337    
338 pcg 1.20 The numerical IP address of the remote node (gvpe accepts connections from
339     everywhere, as long as the other node can authenticate itself).
340 pcg 1.1
341     =item DESTPORT=655 # deprecated
342    
343 pcg 1.24 The protocol port used by the other side, if applicable.
344 pcg 1.1
345 pcg 1.24 =item STATE=up
346 pcg 1.1
347 pcg 1.24 Node-up scripts get called with STATE=up, node-change scripts get called
348     with STATE=change and node-down scripts get called with STATE=down.
349 pcg 1.1
350     =back
351    
352     Here is a nontrivial example that uses nsupdate to update the name => ip
353 pcg 1.21 mapping in some DNS zone:
354 pcg 1.1
355     #!/bin/sh
356     {
357     echo update delete $DESTNODE.lowttl.example.net. a
358     echo update add $DESTNODE.lowttl.example.net. 1 in a $DESTIP
359     echo
360     } | nsupdate -d -k $CONFBASE:key.example.net.
361    
362 pcg 1.24 =item node-change = relative-or-absolute-path
363    
364     Same as C<node-change>, but gets called whenever something about a
365     connection changes (such as the source IP address).
366    
367 pcg 1.1 =item node-down = relative-or-absolute-path
368    
369     Same as C<node-up>, but gets called whenever a connection is lost.
370    
371 pcg 1.6 =item pid-file = path
372    
373     The path to the pid file to check and create
374     (default: C<LOCALSTATEDIR/run/gvpe.pid>).
375    
376     =item private-key = relative-path-to-key
377    
378     Sets the path (relative to the config directory) to the private key
379     (default: C<hostkey>). This is a printf format string so every C<%> must
380     be doubled. A single C<%s> is replaced by the hostname, so you could
381     use paths like C<hostkeys/%s> to fetch the files at the location where
382     C<gvpectrl> puts them.
383 pcg 1.1
384 pcg 1.6 Since only the private key file of the current node is used and the
385 pcg 1.21 private key file should be kept secret per-node to avoid spoofing, it is
386 pcg 1.6 not recommended to use this feature.
387 pcg 1.1
388 pcg 1.6 =item rekey = seconds
389 pcg 1.1
390 pcg 1.6 Sets the rekeying interval in seconds (default: C<3600>). Connections are
391 pcg 1.21 reestablished every C<rekey> seconds, making them use a new encryption
392     key.
393 pcg 1.1
394 pcg 1.23 =item nfmark = integer
395    
396     This advanced option, when set to a nonzero value (default: C<0>), tries
397     to set the netfilter mark (or fwmark) value on all sockets gvpe uses to
398     send packets.
399    
400     This can be used to make gvpe use a different set of routing rules. For
401     example, on GNU/Linux, the C<if-up> could set C<nfmark> to 1000 and then
402     put all routing rules into table C<99> and then use an ip rule to make
403     gvpe traffic avoid that routing table, in effect routing normal traffic
404     via gvpe and gvpe traffic via the normal system routing tables:
405    
406     ip rule add not fwmark 1000 lookup 99
407    
408 pcg 1.6 =back
409 pcg 1.1
410 pcg 1.6 =head2 NODE SPECIFIC SETTINGS
411 pcg 1.1
412 pcg 1.6 The following settings are node-specific, that is, every node can have
413     different settings, even within the same gvpe instance. Settings that are
414 pcg 1.15 set before the first node section set the defaults, settings that are
415     set within a node section only apply to the given node.
416 pcg 1.1
417 pcg 1.6 =over 4
418 pcg 1.1
419 pcg 1.15 =item allow-direct = nodename
420    
421     Allow direct connections to this node. See C<deny-direct> for more info.
422    
423 pcg 1.6 =item compress = yes|true|on | no|false|off
424 pcg 1.1
425 pcg 1.20 Wether to compress data packets sent to this node (default: C<yes>).
426 pcg 1.6 Compression is really cheap even on slow computers and has no size
427 pcg 1.21 overhead at all, so enabling this is often a good idea.
428 pcg 1.1
429 pcg 1.6 =item connect = ondemand | never | always | disabled
430 pcg 1.1
431 pcg 1.6 Sets the connect mode (default: C<always>). It can be C<always> (always
432 pcg 1.20 try to establish and keep a connection to the given node), C<never>
433 pcg 1.6 (never initiate a connection to the given host, but accept connections),
434 pcg 1.18 C<ondemand> (try to establish a connection when there are outstanding
435     packets in the queue and take it down after the keepalive interval) or
436     C<disabled> (node is bad, don't talk to it).
437 pcg 1.1
438 pcg 1.20 Routers will automatically be forced to C<always> unless they are
439     C<disabled>, to ensure all nodes can talk to each other.
440    
441 pcg 1.15 =item deny-direct = nodename | *
442    
443     Deny direct connections to the specified node (or all nodes when C<*>
444     is given). Only one node can be specified, but you can use multiple
445     C<allow-direct> and C<deny-direct> statements. This only makes sense in
446     networks with routers, as routers are required for indirect connections.
447    
448     Sometimes, a node cannot reach some other nodes for reasons of network
449     connectivity. For example, a node behind a firewall that only allows
450 pcg 1.21 connections to/from a single other node in the network. In this case one
451 pcg 1.15 should specify C<deny-direct = *> and C<allow-direct = othernodename> (the other
452     node I<must> be a router for this to work).
453    
454 pcg 1.21 The algorithm to check whether a connection may be direct is as follows:
455 pcg 1.15
456 pcg 1.21 1. Other node mentioned in an C<allow-direct>? If yes, allow the connection.
457 pcg 1.15
458     2. Other node mentioned in a C<deny-direct>? If yes, deny direct connections.
459    
460     3. Allow the connection.
461    
462 pcg 1.16 That is, C<allow-direct> takes precedence over C<deny-direct>.
463 pcg 1.15
464     The check is done in both directions, i.e. both nodes must allow a direct
465     connection before one is attempted, so you only need to specify connect
466     limitations on one node.
467    
468 pcg 1.6 =item dns-domain = domain-suffix
469 pcg 1.1
470 pcg 1.7 The DNS domain suffix that points to the DNS tunnel server for this node.
471 pcg 1.1
472 pcg 1.6 The domain must point to a NS record that points to the I<dns-hostname>,
473     i.e.
474 pcg 1.1
475 pcg 1.6 dns-domainname = tunnel.example.net
476     dns-hostname = tunnel-server.example.net
477 pcg 1.1
478 pcg 1.6 Corresponds to the following DNS entries in the C<example.net> domain:
479 pcg 1.1
480 pcg 1.6 tunnel.example.net. NS tunnel-server.example.net.
481     tunnel-server.example.net. A 13.13.13.13
482 pcg 1.1
483 pcg 1.6 =item dns-hostname = hostname/ip
484 pcg 1.1
485 pcg 1.6 The address to bind the DNS tunnel socket to, similar to the C<hostname>,
486     but for the DNS tunnel protocol only. Default: C<0.0.0.0>, but that might
487     change.
488 pcg 1.1
489 pcg 1.6 =item dns-port = port-number
490 pcg 1.1
491 pcg 1.8 The port to bind the DNS tunnel socket to. Must be C<53> on DNS tunnel servers.
492 pcg 1.1
493 pcg 1.7 =item enable-dns = yes|true|on | no|false|off
494    
495 pcg 1.10 See gvpe.protocol(7) for a description of the DNS transport
496     protocol. Avoid this protocol if you can.
497    
498 pcg 1.8 Enable the DNS tunneling protocol on this node, either as server or as
499 pcg 1.10 client. Support for this transport protocol is only available when gvpe
500     was compiled using the C<--enable-dns> option.
501    
502     =item enable-icmp = yes|true|on | no|false|off
503    
504     See gvpe.protocol(7) for a description of the ICMP transport protocol.
505 pcg 1.8
506 pcg 1.21 Enable the ICMP transport using ICMP packets of type C<icmp-type> on this
507 pcg 1.10 node.
508 pcg 1.7
509 pcg 1.1 =item enable-rawip = yes|true|on | no|false|off
510    
511 pcg 1.10 See gvpe.protocol(7) for a description of the RAW IP transport protocol.
512    
513 pcg 1.1 Enable the RAW IPv4 transport using the C<ip-proto> protocol
514 pcg 1.10 (default: C<no>).
515 pcg 1.1
516 pcg 1.6 =item enable-tcp = yes|true|on | no|false|off
517    
518 pcg 1.10 See gvpe.protocol(7) for a description of the TCP transport protocol.
519    
520 pcg 1.6 Enable the TCPv4 transport using the C<tcp-port> port
521 pcg 1.10 (default: C<no>). Support for this transport protocol is only available
522     when gvpe was compiled using the C<--enable-tcp> option.
523 pcg 1.6
524 pcg 1.1 =item enable-udp = yes|true|on | no|false|off
525    
526 pcg 1.10 See gvpe.protocol(7) for a description of the UDP transport protocol.
527    
528 pcg 1.5 Enable the UDPv4 transport using the C<udp-port> port (default: C<no>,
529     unless no other protocol is enabled for a node, in which case this
530 pcg 1.10 protocol is enabled automatically).
531 pcg 1.5
532 pcg 1.21 NOTE: Please specify C<enable-udp = yes> if you want to use it even though
533 pcg 1.5 it might get switched on automatically, as some future version might
534     default to another default protocol.
535 pcg 1.15
536     =item hostname = hostname | ip [can not be defaulted]
537    
538 pcg 1.21 Forces the address of this node to be set to the given DNS hostname or IP
539 pcg 1.15 address. It will be resolved before each connect request, so dyndns should
540     work fine. If this setting is not specified and a router is available,
541     then the router will be queried for the address of this node. Otherwise,
542     the connection attempt will fail.
543 pcg 1.1
544 pcg 1.21 Note that DNS resolving is done synchronously, pausing the daemon. If that
545     is an issue you need to specify IP addresses.
546    
547 pcg 1.11 =item icmp-type = integer
548    
549     Sets the type value to be used for outgoing (and incoming) packets sent
550     via the ICMP transport.
551    
552     The default is C<0> (which is C<echo-reply>, also known as
553 pcg 1.21 "ping-reply"). Other useful values include C<8> (C<echo-request>, a.k.a.
554 pcg 1.11 "ping") and C<11> (C<time-exceeded>), but any 8-bit value can be used.
555    
556 pcg 1.13 =item if-up-data = value
557    
558     The value specified using this directive will be passed to the C<if-up>
559     script in the environment variable C<IFUPDATA>.
560    
561 pcg 1.6 =item inherit-tos = yes|true|on | no|false|off
562    
563     Wether to inherit the TOS settings of packets sent to the tunnel when
564     sending packets to this node (default: C<yes>). If set to C<yes> then
565     outgoing tunnel packets will have the same TOS setting as the packets sent
566     to the tunnel device, which is usually what you want.
567    
568     =item max-retry = positive-number
569 pcg 1.1
570 pcg 1.8 The maximum interval in seconds (default: C<3600>, one hour) between
571 pcg 1.6 retries to establish a connection to this node. When a connection cannot
572 pcg 1.21 be established, gvpe uses exponential back-off capped at this value. It's
573 pcg 1.6 sometimes useful to set this to a much lower value (e.g. C<120>) on
574     connections to routers that usually are stable but sometimes are down, to
575 pcg 1.8 assure quick reconnections even after longer downtimes.
576 pcg 1.1
577 pcg 1.18 =item max-ttl = seconds
578    
579     Expire packets that couldn't be sent after this many seconds
580     (default: C<60>). Gvpe will normally queue packets for a node without an
581     active connection, in the hope of establishing a connection soon. This
582     value specifies the maximum lifetime a packet will stay in the queue, if a
583     packet gets older, it will be thrown away.
584    
585 pcg 1.20 =item max-queue = positive-number>=1
586 pcg 1.18
587     The maximum number of packets that will be queued (default: C<512>)
588     for this node. If more packets are sent then earlier packets will be
589     expired. See C<max-ttl>, above.
590    
591 pcg 1.8 =item router-priority = 0 | 1 | positive-number>=2
592 pcg 1.1
593 pcg 1.20 Sets the router priority of the given node (default: C<0>, disabled).
594    
595     If some node tries to connect to another node but it doesn't have a
596     hostname, it asks a router node for it's IP address. The router node
597     chosen is the one with the highest priority larger than C<1> that is
598     currently reachable. This is called a I<mediated> connection, as the
599     connection itself will still be direct, but it uses another node to
600     mediate between the two nodes.
601 pcg 1.1
602 pcg 1.20 The value C<0> disables routing, that means if the node receives a packet
603     not for itself it will not forward it but instead drop it.
604 pcg 1.2
605     The special value C<1> allows other hosts to route through the router
606 pcg 1.20 host, but they will never route through it by default (i.e. the config
607     file of another node needs to specify a router priority higher than one
608     to choose such a node for routing).
609    
610     The idea behind this is that some hosts can, if required, bump the
611     C<router-priority> setting to higher than C<1> in their local config to
612     route through specific hosts. If C<router-priority> is C<0>, then routing
613     will be refused, so C<1> serves as a "enable, but do not use by default"
614     switch.
615    
616     Nodes with C<router-priority> set to C<2> or higher will always be forced
617     to C<connect> = C<always> (unless they are C<disabled>).
618 pcg 1.2
619 pcg 1.6 =item tcp-port = port-number
620 pcg 1.1
621 pcg 1.6 Similar to C<udp-port> (default: C<655>), but sets the TCP port number.
622 pcg 1.1
623 pcg 1.6 =item udp-port = port-number
624 pcg 1.1
625 pcg 1.6 Sets the port number used by the UDP protocol (default: C<655>, not
626     officially assigned by IANA!).
627 pcg 1.1
628     =back
629    
630     =head1 CONFIG DIRECTORY LAYOUT
631    
632     The default (or recommended) directory layout for the config directory is:
633    
634     =over 4
635    
636 pcg 1.22 =item gvpe.conf
637 pcg 1.1
638     The config file.
639    
640 pcg 1.22 =item if-up
641 pcg 1.1
642     The if-up script
643    
644 pcg 1.22 =item node-up, node-down
645 pcg 1.1
646     If used the node up or node-down scripts.
647    
648 pcg 1.22 =item hostkey
649 pcg 1.1
650     The private key (taken from C<hostkeys/nodename>) of the current host.
651    
652 pcg 1.22 =item pubkey/nodename
653 pcg 1.1
654     The public keys of the other nodes, one file per node.
655    
656     =back
657    
658     =head1 SEE ALSO
659    
660     gvpe(5), gvpe(8), gvpectrl(8).
661    
662     =head1 AUTHOR
663    
664 pcg 1.14 Marc Lehmann <gvpe@schmorp.de>
665 pcg 1.1