en/handbook/hb-net-modules.xml

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE sections SYSTEM "/dtd/book.dtd">

<!-- The content of this document is licensed under the CC-BY-SA license -->
<!-- See http://creativecommons.org/licenses/by-sa/2.5 -->

<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/handbook/hb-net-modules.xml,v 1.5 2005/06/14 18:34:32 neysx Exp $ -->

<sections>

<version>1.2</version>
<date>2005-07-20</date>

<section>
<title>Network Modules</title>
<body>

<p>
We now support modular networking scripts, which means we can easily
add support for new interface types and configuration modules while keeping
compatibility with existing ones.
</p>

<p>
Modules load by default if the package they need is installed.  If
you specify a module here that doesn't have its package installed
then you get an error stating which package you need to install.
Ideally, you only use the modules setting when you have two or more
packages installed that supply the same service and you need to prefer
one over the other.
</p>

<pre caption="Module preference">
<comment># Prefer iproute2 over ifconfig</comment>
modules=( "iproute2" )

<comment># You can also specify other modules for an interface
# In this case we prefer udhcpc over dhcpcd</comment>
modules_eth0=( "udhcpc" )

<comment># You can also specify which modules not to use - for example you may be
# using a supplicant or linux-wlan-ng to control wireless configuration but
# you still want to configure network settings per ESSID associated with.</comment>
modules=( "!iwconfig" )
</pre>

</body>
</section>
<section>
<title>Interface Handlers</title>
<body>

<p>
We provide two interface handlers presently: ifconfig and iproute2.
You need one of these to do any kind of network configuration.
</p>

<p>
ifconfig is the current Gentoo default and it's included in the system profile.
iproute2 is a more powerful and flexible package, but it's not included by 
default.
</p>

<pre caption="To install iproute2">
# <i>emerge sys-apps/iproute2</i>

<comment># To prefer iproute2 over ifconfig if both are installed</comment>
modules=( "iproute2" )
</pre>

<p>
As both ifconfig and iproute2 do very similar things we allow their basic 
configuration to work with each other. For example both the below code 
snippets work regardless of which module you are using.
</p>

<pre caption="ifconfig and iproute2 examples">
config_eth0=( "192.168.0.2/24" )
config_eth0=( "192.168.0.2 netmask 255.255.255.0" )

<comment># We can also specify broadcast</comment>
config_eth0=( "192.168.0.2/24 brd 192.168.0.255" )
config_eth0=( "192.168.0.2 netmask 255.255.255.0 broadcast 192.168.0.255" )
</pre>

</body>
</section>
<section id="dhcp">
<title>DHCP</title>
<body>

<p>
DHCP is a means of obtaining network information (IP address, DNS servers,
Gateway, etc) from a DHCP server. This means that if there is a DHCP server
running on the network, you just have to tell each client to use DHCP and it
sets up the network all by itself. Of course, you will have to configure for
other things like wireless, ppp or other things if required before you can use
DHCP.
</p>

<p>
DHCP can be provided by dhclient, dhcpcd, pump or udhcpc. Each DHCP module has
its pros and cons - here's a quick run down.
</p>

<table>
<tr>
  <th>DHCP Module</th>
  <th>Package</th>
  <th>Pros</th>
  <th>Cons</th>
</tr>
<tr>
  <ti>dhclient</ti>
  <ti>net-misc/dhcp</ti>
  <ti>
    Made by ISC, the same people who make the BIND DNS software. Very 
    configurable
  </ti>
  <ti>
    Configuration is overly complex, software is quite bloated, cannot get 
    NTP servers from DHCP, does not send hostname by default
  </ti>
</tr>
<tr>
  <ti>dhcpcd</ti>
  <ti>net-misc/dhcpcd</ti>
  <ti>
    Long time Gentoo default, no reliance on outside tools
  </ti>
  <ti>
    No longer maintained upstream, can be slow at times, does not daemonize 
    when lease is infinite
  </ti>
</tr>
<tr>
  <ti>pump</ti>
  <ti>net-misc/pump</ti>
  <ti>
    Lightweight, no reliance on outside tools
  </ti>
  <ti>
    No longer maintained upstream, unreliable, especially over modems, cannot
    get NIS servers from DHCP
  </ti>
</tr>
<tr>
  <ti>udhcpc</ti>
  <ti>net-misc/udhcp</ti>
  <ti>
    Lightweight - smallest dhcp client around, made for embedded systems
  </ti>
  <ti>
    Unproven - no distro uses it by default, cannot define a timeout beyond 3 
    seconds
  </ti>
</tr>
</table>
     
<p>
If you have more than one DHCP client installed, you need to specify which
one to use - otherwise we default to dhcpcd if available.
</p>

<p>
To send specific options to the dhcp module, use module_eth0="..."
<e>(change module to the DHCP module you're using - ie dhcpcd_eth0)</e>
</p>

<p>
We try and make DHCP relatively agnostic - as such we support the following
commands using the dhcp_eth0 variable. The default is not to set any of them
</p>

<ul>
  <li>release - releases the IP address for re-use</li>
  <li>nodns - don't overwrite /etc/resolv.conf</li>
  <li>nontp - don't overwrite /etc/ntp.conf</li>
  <li>nonis - don't overwrite /etc/yp.conf</li>
</ul>

<pre caption="Sample DHCP configuration in /etc/conf.d/net">
<comment># Only needed if you have more than one DHCP module installed</comment>
modules=( "dhcpcd" ) 

config_eth0=( "dhcp" )
dhcpcd_eth0="-t 10" <comment># Timeout after 10 seconds</comment>
dhcp_eth0="release nodns nontp nonis" <comment># Only get an address</comment>
</pre>

<note>
dhcpcd, udhcpc and pump send the current hostname to the DHCP server by
default so you don't need to specify this anymore.
</note>

</body>
</section>
<section>
<title>ADSL Modem</title>
<body>

<p>
First we need to install the ADSL software.
</p>

<pre caption="Install the rp-pppoe package">
# <i>emerge net-dialup/rp-pppoe</i>
</pre>

<warn>
baselayout-1.11.x supports PPPOE only. Hopefully future versions will support 
PPPOA.
</warn>

<p>
Now we need to instruct configure eth0 to be an ADSL interface and enter our
username.
</p>

<pre caption="Configure eth0 for ADSL">
config_eth0=( "adsl" )
user_eth0="username"
</pre>

<p>
Finally you need to define your username and password in
<path>/etc/ppp/pap-secrets</path>
</p>

<pre caption="sample /etc/ppp/pap-secrets">
<comment># The * is important</comment>
"username"  *  "password"
</pre>

</body>
</section>
<section id="apipa">
<title>APIPA (Automatic Private IP Addressing)</title>
<body>

<p>
APIPA tries to find a free address in the range 169.254.0.0-169.254.255.255
by arping a random address in that range on the interface. If no reply is
found then we assign that address to the interface.
</p>

<p>
This is only useful for LANs where there is no DHCP server and you don't
connect directly to the internet and all other computers use APIPA.
</p>

<p>
For APIPA support, emerge net-misc/iputils or net-analyzer/arping
</p>

<pre caption="APIPA configuration in /etc/conf.d/net">
<comment># Try DHCP first - if that fails then fallback to APIPA</comment>
config_eth0=( "dhcp" )
fallback_eth0=( "apipa" )

<comment># Just use APIPA</comment>
config_eth0=( "apipa" )
</pre>

</body>
</section>
<section>
<title>Bonding</title>
<body>

<p>
For link bonding/trunking emerge net-misc/ifenslave
</p>

<p>
Bonding is used to increase network bandwidth. If you have two network
cards going to the same network, you can bond them together so your
applications see just one interface but they really use both network cards.
</p>

<pre caption="bonding configuration in /etc/conf.d/net">
<comment>To bond interfaces together</comment>
slaves_bond0="eth0 eth1 eth2"

<comment># You may not want to assign an IP to the bonded interface</comment>
config_bond0=( "null" )

<comment># Depend on eth0, eth1 and eth2 as they may require extra configuration</comment>
depend_bond0() {
  need net.eth0 net.eth1 net.eth2
}
</pre>

</body>
</section>
<section>
<title>Bridging (802.1d support)</title>
<body>

<p>
For bridging support emerge net-misc/bridge-utils
</p>

<p>
Bridging is used to join networks together. For example, you may have a
server that connects to the internet via an ADSL modem and a wireless
access card to enable other computers to connect to the internet via the
ADSL modem. You could create a bridge to join the two interfaces together.
</p>

<pre caption="Bridge configuration in /etc/conf.d/net">
<comment># Configure the bridge - "man btctl" for more details</comment>
brctl_br0=( "setfd 0" "sethello 0" "stp off" )

<comment># To add ports to bridge br0</comment>
bridge_br0="eth0 eth1"

<comment># You need to configure the ports to null values so dhcp does not get started</comment>
config_eth0=( "null" )
config_eth1=( "null" )

<comment># Finally give the bridge an address - you could use DHCP as well</comment>
config_br0=( "192.168.0.1/24" )

<comment># Depend on eth0 and eth1 as they may require extra configuration</comment>
depend_br0() {
  need net.eth0 net.eth1
}
</pre>

<impo>
For using some bridge setups, you may need to consult the
<uri link="?part=3&amp;chap=2#variable_name">variable name</uri>
documentation.
</impo>

</body>
</section>
<section>
<title>MAC Address</title>
<body>

<p>
You don't need to emerge anything for changing the MAC address of your
interface if you change to a specific address. However, if you need to
change to a random address or a random address of a given type then you
need to emerge net-analyzer/macchanger.
</p>

<pre caption="MAC Address change example">
<comment># To set the MAC address of the interface</comment>
mac_eth0="00:11:22:33:44:55"

<comment># To randomize the last 3 bytes only</comment>
mac_eth0="random-ending"

<comment># To randomize between the same physical type of connection (eg fibre,
# copper, wireless) , all vendors</comment>
mac_eth0="random-samekind"

<comment># To randomize between any physical type of connection (eg fibre, copper,
# wireless) , all vendors</comment>
mac_eth0="random-anykind"

<comment># Full randomization - WARNING: some MAC addresses generated by this may
# NOT act as expected</comment>
mac_eth0="random-full"
</pre>

</body>
</section>
<section>
<title>Tunnelling</title>
<body>

<p>
You don't need to emerge anything for tunnelling as the interface handler
can do it for you.
</p>

<pre caption="Tunnelling configuration in /etc/conf.d/net">
<comment># For GRE tunnels</comment>
iptunnel_vpn0="mode gre remote 207.170.82.1 key 0xffffffff ttl 255"

<comment># For IPIP tunnels</comment>
iptunnel_vpn0="mode ipip remote 207.170.82.2 ttl 255"

<comment># To configure the interface</comment>
config_vpn0=( "192.168.0.2 peer 192.168.1.1" ) 
</pre>

</body>
</section>
<section>
<title>VLAN (802.1q support)</title>
<body>

<p>
For VLAN support, emerge net-misc/vconfig
</p>

<p>
Virtual LAN is a group of network devices that behave as if they were
connected to a single network segment - even though they may not be.
VLAN members can only see members of the same VLAN even though they may
share the same physical network.
</p>

<pre caption="VLAN configuration in /etc/conf.d/net">
<comment># Specify the VLAN numbers for the interface like so</comment>
<comment># Please ensure your VLAN IDs are NOT zero-padded</comment>
vlans_eth0="1 2"

<comment># You can also configure the VLAN</comment>
<comment># see for vconfig man page for more details</comment>
vconfig_eth0=( "set_name_type VLAN_PLUS_VID_NO_PAD" )
vconfig_vlan1=( "set_flag 1" "set_egress_map 2 6" )

<comment># Configure the interface as usual</comment>
config_vlan1=( "172.16.3.1 netmask 255.255.254.0" )
config_vlan2=( "172.16.2.1 netmask 255.255.254.0" )
</pre>

<impo>
For using some VLAN setups, you may need to consult the
<uri link="?part=3&amp;chap=2#variable_name">variable name</uri>
documentation.
</impo>

</body>
</section>

</sections>
1	swift	1.1	<?xml version="1.0" encoding="UTF-8"?>
2			<!DOCTYPE sections SYSTEM "/dtd/book.dtd">
3
4			<!-- The content of this document is licensed under the CC-BY-SA license -->
5	swift	1.2	<!-- See http://creativecommons.org/licenses/by-sa/2.5 -->
6	swift	1.1
7	fox2mike	1.6	<!-- $Header: /var/cvsroot/gentoo/xml/htdocs/doc/en/handbook/hb-net-modules.xml,v 1.5 2005/06/14 18:34:32 neysx Exp $ -->
8	neysx	1.5
9	swift	1.1	<sections>
10
11	fox2mike	1.6	<version>1.2</version>
12			<date>2005-07-20</date>
13	swift	1.1
14			<section>
15			<title>Network Modules</title>
16			<body>
17
18			<p>
19			We now support modular networking scripts, which means we can easily
20			add support for new interface types and configuration modules while keeping
21			compatibility with existing ones.
22			</p>
23
24			<p>
25			Modules load by default if the package they need is installed. If
26			you specify a module here that doesn't have its package installed
27			then you get an error stating which package you need to install.
28			Ideally, you only use the modules setting when you have two or more
29			packages installed that supply the same service and you need to prefer
30			one over the other.
31			</p>
32
33			<pre caption="Module preference">
34			<comment># Prefer iproute2 over ifconfig</comment>
35			modules=( "iproute2" )
36
37			<comment># You can also specify other modules for an interface
38			# In this case we prefer udhcpc over dhcpcd</comment>
39			modules_eth0=( "udhcpc" )
40
41			<comment># You can also specify which modules not to use - for example you may be
42			# using a supplicant or linux-wlan-ng to control wireless configuration but
43			# you still want to configure network settings per ESSID associated with.</comment>
44			modules=( "!iwconfig" )
45			</pre>
46
47			</body>
48			</section>
49			<section>
50			<title>Interface Handlers</title>
51			<body>
52
53			<p>
54			We provide two interface handlers presently: ifconfig and iproute2.
55			You need one of these to do any kind of network configuration.
56			</p>
57
58			<p>
59	swift	1.3	ifconfig is the current Gentoo default and it's included in the system profile.
60	swift	1.2	iproute2 is a more powerful and flexible package, but it's not included by
61			default.
62	swift	1.1	</p>
63
64			<pre caption="To install iproute2">
65			# <i>emerge sys-apps/iproute2</i>
66
67			<comment># To prefer iproute2 over ifconfig if both are installed</comment>
68			modules=( "iproute2" )
69			</pre>
70
71			<p>
72	swift	1.2	As both ifconfig and iproute2 do very similar things we allow their basic
73			configuration to work with each other. For example both the below code
74			snippets work regardless of which module you are using.
75	swift	1.1	</p>
76
77			<pre caption="ifconfig and iproute2 examples">
78			config_eth0=( "192.168.0.2/24" )
79			config_eth0=( "192.168.0.2 netmask 255.255.255.0" )
80
81			<comment># We can also specify broadcast</comment>
82			config_eth0=( "192.168.0.2/24 brd 192.168.0.255" )
83			config_eth0=( "192.168.0.2 netmask 255.255.255.0 broadcast 192.168.0.255" )
84			</pre>
85
86			</body>
87			</section>
88			<section id="dhcp">
89			<title>DHCP</title>
90			<body>
91
92			<p>
93			DHCP is a means of obtaining network information (IP address, DNS servers,
94			Gateway, etc) from a DHCP server. This means that if there is a DHCP server
95			running on the network, you just have to tell each client to use DHCP and it
96			sets up the network all by itself. Of course, you will have to configure for
97			other things like wireless, ppp or other things if required before you can use
98			DHCP.
99			</p>
100
101			<p>
102	fox2mike	1.6	DHCP can be provided by dhclient, dhcpcd, pump or udhcpc. Each DHCP module has
103			its pros and cons - here's a quick run down.
104	swift	1.1	</p>
105
106			<table>
107	swift	1.2	<tr>
108			<th>DHCP Module</th>
109			<th>Package</th>
110			<th>Pros</th>
111			<th>Cons</th>
112			</tr>
113			<tr>
114			<ti>dhclient</ti>
115			<ti>net-misc/dhcp</ti>
116			<ti>
117	swift	1.3	Made by ISC, the same people who make the BIND DNS software. Very
118			configurable
119	swift	1.2	</ti>
120			<ti>
121	swift	1.3	Configuration is overly complex, software is quite bloated, cannot get
122			NTP servers from DHCP, does not send hostname by default
123	swift	1.2	</ti>
124			</tr>
125			<tr>
126			<ti>dhcpcd</ti>
127			<ti>net-misc/dhcpcd</ti>
128			<ti>
129	swift	1.3	Long time Gentoo default, no reliance on outside tools
130	swift	1.2	</ti>
131			<ti>
132	swift	1.3	No longer maintained upstream, can be slow at times, does not daemonize
133			when lease is infinite
134	swift	1.2	</ti>
135			</tr>
136			<tr>
137			<ti>pump</ti>
138			<ti>net-misc/pump</ti>
139			<ti>
140	swift	1.3	Lightweight, no reliance on outside tools
141	swift	1.2	</ti>
142			<ti>
143	swift	1.3	No longer maintained upstream, unreliable, especially over modems, cannot
144			get NIS servers from DHCP
145	swift	1.2	</ti>
146			</tr>
147			<tr>
148			<ti>udhcpc</ti>
149			<ti>net-misc/udhcp</ti>
150			<ti>
151	swift	1.3	Lightweight - smallest dhcp client around, made for embedded systems
152	swift	1.2	</ti>
153			<ti>
154	swift	1.3	Unproven - no distro uses it by default, cannot define a timeout beyond 3
155			seconds
156	swift	1.2	</ti>
157			</tr>
158	swift	1.1	</table>
159
160			<p>
161			If you have more than one DHCP client installed, you need to specify which
162			one to use - otherwise we default to dhcpcd if available.
163			</p>
164
165			<p>
166			To send specific options to the dhcp module, use module_eth0="..."
167			<e>(change module to the DHCP module you're using - ie dhcpcd_eth0)</e>
168			</p>
169
170			<p>
171			We try and make DHCP relatively agnostic - as such we support the following
172			commands using the dhcp_eth0 variable. The default is not to set any of them
173			</p>
174
175			<ul>
176			<li>release - releases the IP address for re-use</li>
177			<li>nodns - don't overwrite /etc/resolv.conf</li>
178			<li>nontp - don't overwrite /etc/ntp.conf</li>
179			<li>nonis - don't overwrite /etc/yp.conf</li>
180			</ul>
181
182			<pre caption="Sample DHCP configuration in /etc/conf.d/net">
183			<comment># Only needed if you have more than one DHCP module installed</comment>
184			modules=( "dhcpcd" )
185
186			config_eth0=( "dhcp" )
187			dhcpcd_eth0="-t 10" <comment># Timeout after 10 seconds</comment>
188			dhcp_eth0="release nodns nontp nonis" <comment># Only get an address</comment>
189			</pre>
190
191			<note>
192			dhcpcd, udhcpc and pump send the current hostname to the DHCP server by
193			default so you don't need to specify this anymore.
194			</note>
195
196			</body>
197			</section>
198			<section>
199			<title>ADSL Modem</title>
200			<body>
201
202			<p>
203			First we need to install the ADSL software.
204			</p>
205
206			<pre caption="Install the rp-pppoe package">
207			# <i>emerge net-dialup/rp-pppoe</i>
208			</pre>
209
210			<warn>
211	swift	1.3	baselayout-1.11.x supports PPPOE only. Hopefully future versions will support
212			PPPOA.
213	swift	1.1	</warn>
214
215			<p>
216			Now we need to instruct configure eth0 to be an ADSL interface and enter our
217			username.
218			</p>
219
220			<pre caption="Configure eth0 for ADSL">
221			config_eth0=( "adsl" )
222			user_eth0="username"
223			</pre>
224
225			<p>
226			Finally you need to define your username and password in
227			<path>/etc/ppp/pap-secrets</path>
228			</p>
229
230			<pre caption="sample /etc/ppp/pap-secrets">
231			<comment># The * is important</comment>
232	swift	1.4	"username" * "password"
233	swift	1.1	</pre>
234
235			</body>
236			</section>
237			<section id="apipa">
238			<title>APIPA (Automatic Private IP Addressing)</title>
239			<body>
240
241			<p>
242			APIPA tries to find a free address in the range 169.254.0.0-169.254.255.255
243			by arping a random address in that range on the interface. If no reply is
244			found then we assign that address to the interface.
245			</p>
246
247			<p>
248			This is only useful for LANs where there is no DHCP server and you don't
249			connect directly to the internet and all other computers use APIPA.
250			</p>
251
252			<p>
253			For APIPA support, emerge net-misc/iputils or net-analyzer/arping
254			</p>
255
256			<pre caption="APIPA configuration in /etc/conf.d/net">
257			<comment># Try DHCP first - if that fails then fallback to APIPA</comment>
258			config_eth0=( "dhcp" )
259			fallback_eth0=( "apipa" )
260
261			<comment># Just use APIPA</comment>
262			config_eth0=( "apipa" )
263			</pre>
264
265			</body>
266			</section>
267			<section>
268			<title>Bonding</title>
269			<body>
270
271			<p>
272			For link bonding/trunking emerge net-misc/ifenslave
273			</p>
274
275			<p>
276			Bonding is used to increase network bandwidth. If you have two network
277			cards going to the same network, you can bond them together so your
278			applications see just one interface but they really use both network cards.
279			</p>
280
281			<pre caption="bonding configuration in /etc/conf.d/net">
282			<comment>To bond interfaces together</comment>
283			slaves_bond0="eth0 eth1 eth2"
284
285			<comment># You may not want to assign an IP to the bonded interface</comment>
286			config_bond0=( "null" )
287
288			<comment># Depend on eth0, eth1 and eth2 as they may require extra configuration</comment>
289			depend_bond0() {
290	swift	1.4	need net.eth0 net.eth1 net.eth2
291	swift	1.1	}
292			</pre>
293
294			</body>
295			</section>
296			<section>
297			<title>Bridging (802.1d support)</title>
298			<body>
299
300			<p>
301			For bridging support emerge net-misc/bridge-utils
302			</p>
303
304			<p>
305			Bridging is used to join networks together. For example, you may have a
306			server that connects to the internet via an ADSL modem and a wireless
307			access card to enable other computers to connect to the internet via the
308			ADSL modem. You could create a bridge to join the two interfaces together.
309			</p>
310
311			<pre caption="Bridge configuration in /etc/conf.d/net">
312			<comment># Configure the bridge - "man btctl" for more details</comment>
313			brctl_br0=( "setfd 0" "sethello 0" "stp off" )
314
315			<comment># To add ports to bridge br0</comment>
316			bridge_br0="eth0 eth1"
317
318			<comment># You need to configure the ports to null values so dhcp does not get started</comment>
319			config_eth0=( "null" )
320			config_eth1=( "null" )
321
322			<comment># Finally give the bridge an address - you could use DHCP as well</comment>
323			config_br0=( "192.168.0.1/24" )
324
325			<comment># Depend on eth0 and eth1 as they may require extra configuration</comment>
326			depend_br0() {
327	swift	1.4	need net.eth0 net.eth1
328	swift	1.1	}
329			</pre>
330
331			<impo>
332			For using some bridge setups, you may need to consult the
333			<uri link="?part=3&chap=2#variable_name">variable name</uri>
334			documentation.
335			</impo>
336
337			</body>
338			</section>
339			<section>
340			<title>MAC Address</title>
341			<body>
342
343			<p>
344			You don't need to emerge anything for changing the MAC address of your
345			interface if you change to a specific address. However, if you need to
346			change to a random address or a random address of a given type then you
347			need to emerge net-analyzer/macchanger.
348			</p>
349
350			<pre caption="MAC Address change example">
351			<comment># To set the MAC address of the interface</comment>
352			mac_eth0="00:11:22:33:44:55"
353
354			<comment># To randomize the last 3 bytes only</comment>
355			mac_eth0="random-ending"
356
357			<comment># To randomize between the same physical type of connection (eg fibre,
358			# copper, wireless) , all vendors</comment>
359			mac_eth0="random-samekind"
360
361			<comment># To randomize between any physical type of connection (eg fibre, copper,
362			# wireless) , all vendors</comment>
363			mac_eth0="random-anykind"
364
365			<comment># Full randomization - WARNING: some MAC addresses generated by this may
366			# NOT act as expected</comment>
367			mac_eth0="random-full"
368			</pre>
369
370			</body>
371			</section>
372			<section>
373			<title>Tunnelling</title>
374			<body>
375
376			<p>
377			You don't need to emerge anything for tunnelling as the interface handler
378			can do it for you.
379			</p>
380
381			<pre caption="Tunnelling configuration in /etc/conf.d/net">
382			<comment># For GRE tunnels</comment>
383			iptunnel_vpn0="mode gre remote 207.170.82.1 key 0xffffffff ttl 255"
384
385			<comment># For IPIP tunnels</comment>
386			iptunnel_vpn0="mode ipip remote 207.170.82.2 ttl 255"
387
388			<comment># To configure the interface</comment>
389			config_vpn0=( "192.168.0.2 peer 192.168.1.1" )
390			</pre>
391
392			</body>
393			</section>
394			<section>
395			<title>VLAN (802.1q support)</title>
396			<body>
397
398			<p>
399			For VLAN support, emerge net-misc/vconfig
400			</p>
401
402	swift	1.2	<p>
403			Virtual LAN is a group of network devices that behave as if they were
404	swift	1.1	connected to a single network segment - even though they may not be.
405			VLAN members can only see members of the same VLAN even though they may
406			share the same physical network.
407			</p>
408
409			<pre caption="VLAN configuration in /etc/conf.d/net">
410			<comment># Specify the VLAN numbers for the interface like so</comment>
411			<comment># Please ensure your VLAN IDs are NOT zero-padded</comment>
412			vlans_eth0="1 2"
413
414			<comment># You can also configure the VLAN</comment>
415			<comment># see for vconfig man page for more details</comment>
416			vconfig_eth0=( "set_name_type VLAN_PLUS_VID_NO_PAD" )
417			vconfig_vlan1=( "set_flag 1" "set_egress_map 2 6" )
418
419			<comment># Configure the interface as usual</comment>
420			config_vlan1=( "172.16.3.1 netmask 255.255.254.0" )
421			config_vlan2=( "172.16.2.1 netmask 255.255.254.0" )
422			</pre>
423
424			<impo>
425			For using some VLAN setups, you may need to consult the
426			<uri link="?part=3&chap=2#variable_name">variable name</uri>
427			documentation.
428			</impo>
429
430			</body>
431			</section>
432
433			</sections>