2 <!DOCTYPE refentry PUBLIC
"-//OASIS//DTD DocBook XML V4.5//EN"
3 "http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd">
4 <!-- SPDX-License-Identifier: LGPL-2.1-or-later -->
6 <refentry id=
"systemd.network" conditional='ENABLE_NETWORKD'
7 xmlns:
xi=
"http://www.w3.org/2001/XInclude">
10 <title>systemd.network
</title>
11 <productname>systemd
</productname>
15 <refentrytitle>systemd.network
</refentrytitle>
16 <manvolnum>5</manvolnum>
20 <refname>systemd.network
</refname>
21 <refpurpose>Network configuration
</refpurpose>
25 <para><filename><replaceable>network
</replaceable>.network
</filename></para>
29 <title>Description
</title>
31 <para>A plain ini-style text file that encodes network configuration for matching network
33 <citerefentry><refentrytitle>systemd-networkd
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
34 See
<citerefentry><refentrytitle>systemd.syntax
</refentrytitle><manvolnum>7</manvolnum></citerefentry>
35 for a general description of the syntax.
</para>
37 <para>The main network file must have the extension
<filename>.network
</filename>; other
38 extensions are ignored. Networks are applied to links whenever the links appear.
</para>
40 <para>The
<filename>.network
</filename> files are read from the files located in the system network
41 directories
<filename>/usr/lib/systemd/network
</filename> and
42 <filename>/usr/local/lib/systemd/network
</filename>, the volatile runtime network directory
43 <filename>/run/systemd/network
</filename> and the local administration network directory
44 <filename>/etc/systemd/network
</filename>. All configuration files are collectively sorted and
45 processed in alphanumeric order, regardless of the directories in which they live. However, files
46 with identical filenames replace each other. It is recommended that each filename is prefixed with
47 a number (e.g.
<filename>10-eth0.network
</filename>). Otherwise, the default
48 <filename>.network
</filename> files or those generated by
49 <citerefentry><refentrytitle>systemd-network-generator.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>
50 may take precedence over user configured files. Files in
<filename>/etc/
</filename> have the highest
51 priority, files in
<filename>/run/
</filename> take precedence over files with the same name under
52 <filename>/usr/
</filename>. This can be used to override a system-supplied configuration file with
53 a local file if needed. As a special case, an empty file (file size
0) or symlink with the same
54 name pointing to
<filename>/dev/null
</filename> disables the configuration file entirely (it is
57 <para>Along with the network file
<filename>foo.network
</filename>, a
"drop-in" directory
58 <filename>foo.network.d/
</filename> may exist. All files with the suffix
59 <literal>.conf
</literal> from this directory will be merged in the alphanumeric order and parsed
60 after the main file itself has been parsed. This is useful to alter or add configuration settings,
61 without having to modify the main configuration file. Each drop-in file must have appropriate
62 section headers.
</para>
64 <para>In addition to
<filename>/etc/systemd/network
</filename>, drop-in
<literal>.d
</literal>
65 directories can be placed in
<filename>/usr/lib/systemd/network
</filename> or
66 <filename>/run/systemd/network
</filename> directories. Drop-in files in
67 <filename>/etc/
</filename> take precedence over those in
<filename>/run/
</filename> which in turn
68 take precedence over those in
<filename>/usr/lib/
</filename>. Drop-in files under any of these
69 directories take precedence over the main network file wherever located.
</para>
73 <title>[Match] Section Options
</title>
75 <para>The network file contains a [Match] section, which determines if a given network file may
76 be applied to a given interface; and a [Network] section specifying how the interface should be
77 configured. The first (in alphanumeric order) of the network files that matches a given interface
78 is applied, all later files are ignored, even if they match as well.
</para>
80 <para>A network file is said to match a network interface if all matches specified by the [Match]
81 section are satisfied. When a network file does not contain valid settings in [Match] section, then
82 the file will match all interfaces and
<command>systemd-networkd
</command> warns about that. Hint:
83 to avoid the warning and to make it clear that all interfaces shall be matched, add the following:
84 <programlisting>Name=*
</programlisting> The following keys are accepted:
</para>
86 <variablelist class='network-directives'
>
87 <xi:include href=
"systemd.link.xml" xpointer=
"mac-address" />
88 <xi:include href=
"systemd.link.xml" xpointer=
"permanent-mac-address" />
89 <xi:include href=
"systemd.link.xml" xpointer=
"path" />
90 <xi:include href=
"systemd.link.xml" xpointer=
"driver" />
91 <xi:include href=
"systemd.link.xml" xpointer=
"type" />
92 <xi:include href=
"systemd.link.xml" xpointer=
"kind" />
93 <xi:include href=
"systemd.link.xml" xpointer=
"property" />
96 <term><varname>Name=
</varname></term>
98 <para>A whitespace-separated list of shell-style globs matching the device name, as exposed
99 by the udev property
<literal>INTERFACE
</literal>, or device's alternative names. If the
100 list is prefixed with a
"!", the test is inverted.
</para>
105 <term><varname>WLANInterfaceType=
</varname></term>
107 <para>A whitespace-separated list of wireless network type. Supported values are
108 <literal>ad-hoc
</literal>,
<literal>station
</literal>,
<literal>ap
</literal>,
109 <literal>ap-vlan
</literal>,
<literal>wds
</literal>,
<literal>monitor
</literal>,
110 <literal>mesh-point
</literal>,
<literal>p2p-client
</literal>,
<literal>p2p-go
</literal>,
111 <literal>p2p-device
</literal>,
<literal>ocb
</literal>, and
<literal>nan
</literal>. If the
112 list is prefixed with a
"!", the test is inverted.
</para>
117 <term><varname>SSID=
</varname></term>
119 <para>A whitespace-separated list of shell-style globs matching the SSID of the currently
120 connected wireless LAN. If the list is prefixed with a
"!", the test is inverted.
</para>
125 <term><varname>BSSID=
</varname></term>
127 <para>A whitespace-separated list of hardware address of the currently connected wireless
128 LAN. Use full colon-, hyphen- or dot-delimited hexadecimal. See the example in
129 <varname>MACAddress=
</varname>. This option may appear more than once, in which case the
130 lists are merged. If the empty string is assigned to this option, the list is reset.
</para>
134 <xi:include href=
"systemd.link.xml" xpointer=
"host" />
135 <xi:include href=
"systemd.link.xml" xpointer=
"virtualization" />
136 <xi:include href=
"systemd.link.xml" xpointer=
"kernel-command-line" />
137 <xi:include href=
"systemd.link.xml" xpointer=
"kernel-version" />
138 <xi:include href=
"systemd.link.xml" xpointer=
"credential" />
139 <xi:include href=
"systemd.link.xml" xpointer=
"architecture" />
140 <xi:include href=
"systemd.link.xml" xpointer=
"firmware" />
145 <title>[Link] Section Options
</title>
147 <para>The [Link] section accepts the following keys:
</para>
149 <variablelist class='network-directives'
>
151 <term><varname>MACAddress=
</varname></term>
153 <para>The hardware address to set for the device.
</para>
158 <term><varname>MTUBytes=
</varname></term>
160 <para>The maximum transmission unit in bytes to set for the device. The usual suffixes K, M,
161 G, are supported and are understood to the base of
1024.
</para>
162 <para>Note that if IPv6 is enabled on the interface, and the MTU is chosen below
1280 (the
163 minimum MTU for IPv6) it will automatically be increased to this value.
</para>
168 <term><varname>ARP=
</varname></term>
170 <para>Takes a boolean. If set to true, the ARP (low-level Address Resolution Protocol)
171 for this interface is enabled. When unset, the kernel's default will be used.
</para>
172 <para> For example, disabling ARP is useful when creating multiple MACVLAN or VLAN virtual
173 interfaces atop a single lower-level physical interface, which will then only serve as a
174 link/
"bridge" device aggregating traffic to the same physical link and not participate in
175 the network otherwise. Defaults to unset.
</para>
180 <term><varname>Multicast=
</varname></term>
182 <para>Takes a boolean. If set to true, the multicast flag on the device is enabled. Defaults
188 <term><varname>AllMulticast=
</varname></term>
190 <para>Takes a boolean. If set to true, the driver retrieves all multicast packets from the
191 network. This happens when multicast routing is enabled. Defaults to unset.
</para>
196 <term><varname>Promiscuous=
</varname></term>
198 <para>Takes a boolean. If set to true, promiscuous mode of the interface is enabled. Defaults
200 <para>If this is set to false for the underlying link of a
<literal>passthru
</literal> mode
201 MACVLAN/MACVTAP, the virtual interface will be created with the
<literal>nopromisc
</literal>
207 <term><varname>Unmanaged=
</varname></term>
209 <para>Takes a boolean. When
<literal>yes
</literal>, no attempts are made to bring up or
210 configure matching links, equivalent to when there are no matching network files. Defaults to
211 <literal>no
</literal>.
</para>
212 <para>This is useful for preventing later matching network files from interfering with
213 certain interfaces that are fully controlled by other applications.
</para>
218 <term><varname>Group=
</varname></term>
220 <para>Link groups are similar to port ranges found in managed switches. When network
221 interfaces are added to a numbered group, operations on all the interfaces from that group
222 can be performed at once. Takes an unsigned integer in the range
0…
2147483647. Defaults to
228 <term><varname>RequiredForOnline=
</varname></term>
230 <para>Takes a boolean or a minimum operational state and an optional maximum operational
232 <citerefentry><refentrytitle>networkctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
233 for possible operational states. When
<literal>yes
</literal>, the network is deemed required
234 when determining whether the system is online (including when running
235 <command>systemd-networkd-wait-online
</command>). When
<literal>no
</literal>, the network is
236 ignored when determining the online state. When a minimum operational state and an optional
237 maximum operational state are set,
<literal>yes
</literal> is implied, and this controls the
238 minimum and maximum operational state required for the network interface to be considered
241 <para>Defaults to
<literal>yes
</literal> when
<varname>ActivationPolicy=
</varname> is not
242 set, or set to
<literal>up
</literal>,
<literal>always-up
</literal>, or
243 <literal>bound
</literal>. Defaults to
<literal>no
</literal> when
244 <varname>ActivationPolicy=
</varname> is set to
<literal>manual
</literal> or
245 <literal>down
</literal>. This is forced to
<literal>no
</literal> when
246 <varname>ActivationPolicy=
</varname> is set to
<literal>always-down
</literal>.
</para>
248 <para>The network will be brought up normally (as configured by
249 <varname>ActivationPolicy=
</varname>), but in the event that there is no address being
250 assigned by DHCP or the cable is not plugged in, the link will simply remain offline and be
251 skipped automatically by
<command>systemd-networkd-wait-online
</command> if
252 <literal>RequiredForOnline=no
</literal>.
</para>
257 <term><varname>RequiredFamilyForOnline=
</varname></term>
259 <para>Takes an address family. When specified, an IP address in the given family is deemed
260 required when determining whether the link is online (including when running
261 <command>systemd-networkd-wait-online
</command>). Takes one of
<literal>ipv4
</literal>,
262 <literal>ipv6
</literal>,
<literal>both
</literal>, or
<literal>any
</literal>. Defaults to
263 <literal>any
</literal>. Note that this option has no effect if
264 <literal>RequiredForOnline=no
</literal>, or if
<literal>RequiredForOnline=
</literal>
265 specifies a minimum operational state below
<literal>degraded
</literal>.
</para>
270 <term><varname>ActivationPolicy=
</varname></term>
272 <para>Specifies the policy for
<command>systemd-networkd
</command> managing the link
273 administrative state. Specifically, this controls how
<command>systemd-networkd
</command>
274 changes the network device's
<literal>IFF_UP
</literal> flag, which is sometimes
275 controlled by system administrators by running e.g.,
276 <command>ip link set dev eth0 up
</command> or
<command>ip link set dev eth0 down
</command>,
277 and can also be changed with
<command>networkctl up eth0
</command> or
278 <command>networkctl down eth0
</command>.
</para>
280 <para>Takes one of
<literal>up
</literal>,
<literal>always-up
</literal>,
281 <literal>manual
</literal>,
<literal>always-down
</literal>,
<literal>down
</literal>,
282 or
<literal>bound
</literal>. When
<literal>manual
</literal>,
283 <command>systemd-networkd
</command> will not change the link's admin state automatically;
284 the system administrator must bring the interface up or down manually, as desired. When
285 <literal>up
</literal> (the default) or
<literal>always-up
</literal>, or
286 <literal>down
</literal> or
<literal>always-down
</literal>,
287 <command>systemd-networkd
</command> will set the link up or down, respectively, when the
288 interface is (re)configured. When
<literal>always-up
</literal> or
289 <literal>always-down
</literal>,
<command>systemd-networkd
</command> will set the link up or
290 down, respectively, any time
<command>systemd-networkd
</command> detects a change in the
291 administrative state. When
<varname>BindCarrier=
</varname> is also set, this is automatically
292 set to
<literal>bound
</literal> and any other value is ignored.
</para>
294 <para>When the policy is set to
<literal>down
</literal> or
<literal>manual
</literal>, the
295 default value of
<varname>RequiredForOnline=
</varname> is
<literal>no
</literal>. When the
296 policy is set to
<literal>always-down
</literal>, the value of
297 <varname>RequiredForOnline=
</varname> forced to
<literal>no
</literal>.
</para>
299 <para>The administrative state is not the same as the carrier state, so using
300 <literal>always-up
</literal> does not mean the link will never lose carrier. The link carrier
301 depends on both the administrative state as well as the network device's physical connection.
302 However, to avoid reconfiguration failures, when using
<literal>always-up
</literal>,
303 <varname>IgnoreCarrierLoss=
</varname> is forced to true.
</para>
309 <xi:include href=
"systemd.link.xml" xpointer=
"sr-iov" />
312 <title>[Network] Section Options
</title>
314 <para>The [Network] section accepts the following keys:
</para>
316 <variablelist class='network-directives'
>
318 <term><varname>Description=
</varname></term>
320 <para>A description of the device. This is only used for presentation purposes.
</para>
325 <term><varname>DHCP=
</varname></term>
327 <para>Enables DHCPv4 and/or DHCPv6 client support. Accepts
<literal>yes
</literal>,
328 <literal>no
</literal>,
<literal>ipv4
</literal>, or
<literal>ipv6
</literal>. Defaults to
329 <literal>no
</literal>.
</para>
331 <para>Note that DHCPv6 will by default be triggered by Router Advertisements, if reception is
332 enabled, regardless of this parameter. By explicitly enabling DHCPv6 support here, the DHCPv6
333 client will be started in the mode specified by the
<varname>WithoutRA=
</varname> setting in the
334 [DHCPv6] section, regardless of the presence of routers on the link, or what flags the routers
335 pass. See
<varname>IPv6AcceptRA=
</varname>.
</para>
337 <para>Furthermore, note that by default the domain name specified through DHCP is not used
338 for name resolution. See option
<option>UseDomains=
</option> below.
</para>
340 <para>See the [DHCPv4] or [DHCPv6] sections below for further configuration options for the
341 DHCP client support.
</para>
346 <term><varname>DHCPServer=
</varname></term>
348 <para>Takes a boolean. If set to
<literal>yes
</literal>, DHCPv4 server will be started.
349 Defaults to
<literal>no
</literal>. Further settings for the DHCP server may be set in the
350 [DHCPServer] section described below.
</para>
355 <term><varname>LinkLocalAddressing=
</varname></term>
357 <para>Enables link-local address autoconfiguration. Accepts
<option>yes
</option>,
358 <option>no
</option>,
<option>ipv4
</option>, and
<option>ipv6
</option>. An IPv6 link-local
359 address is configured when
<option>yes
</option> or
<option>ipv6
</option>. An IPv4 link-local
360 address is configured when
<option>yes
</option> or
<option>ipv4
</option> and when DHCPv4
361 autoconfiguration has been unsuccessful for some time. (IPv4 link-local address
362 autoconfiguration will usually happen in parallel with repeated attempts to acquire a DHCPv4
365 <para>Defaults to
<option>no
</option> when
<varname>KeepMaster=
</varname> or
366 <varname>Bridge=
</varname> is set or when the specified
367 <varname>MACVLAN=
</varname>/
<varname>MACVTAP=
</varname> has
<varname>Mode=passthru
</varname>,
368 or
<option>ipv6
</option> otherwise.
</para>
373 <term><varname>IPv6LinkLocalAddressGenerationMode=
</varname></term>
375 <para>Specifies how IPv6 link-local address is generated. Takes one of
376 <literal>eui64
</literal>,
<literal>none
</literal>,
<literal>stable-privacy
</literal> and
377 <literal>random
</literal>. When unset,
<literal>stable-privacy
</literal> is used if
378 <varname>IPv6StableSecretAddress=
</varname> is specified, and if not,
379 <literal>eui64
</literal> is used. Note that if
<varname>LinkLocalAddressing=
</varname> is
380 <literal>no
</literal> or
<literal>ipv4
</literal>, then
381 <varname>IPv6LinkLocalAddressGenerationMode=
</varname> will be ignored. Also, even if
382 <varname>LinkLocalAddressing=
</varname> is
<literal>yes
</literal> or
<literal>ipv6
</literal>,
383 setting
<varname>IPv6LinkLocalAddressGenerationMode=none
</varname>
384 disables to configure an IPv6 link-local address.
</para>
389 <term><varname>IPv6StableSecretAddress=
</varname></term>
391 <para>Takes an IPv6 address. The specified address will be used as a stable secret for
392 generating IPv6 link-local address. If this setting is specified, and
393 <varname>IPv6LinkLocalAddressGenerationMode=
</varname> is unset, then
394 <varname>IPv6LinkLocalAddressGenerationMode=stable-privacy
</varname> is implied.
395 If this setting is not specified, and
<literal>stable-privacy
</literal> is set to
396 <varname>IPv6LinkLocalAddressGenerationMode=
</varname>,
397 then a stable secret address will be generated from the local machine ID and the interface
403 <term><varname>IPv4LLStartAddress=
</varname></term>
405 <para>Specifies the first IPv4 link-local address to try. Takes an IPv4 address for example
406 169.254.1.2, from the link-local address range:
169.254.0.0/
16 except for
169.254.0.0/
24 and
407 169.254.255.0/
24. This setting may be useful if the device should always have the same address
408 as long as there is no address conflict. When unset, a random address will be automatically
409 selected. Defaults to unset.
</para>
414 <term><varname>IPv4LLRoute=
</varname></term>
416 <para>Takes a boolean. If set to true, sets up the route needed for non-IPv4LL hosts to
417 communicate with IPv4LL-only hosts. Defaults to false.
</para>
422 <term><varname>DefaultRouteOnDevice=
</varname></term>
424 <para>Takes a boolean. If set to true, sets up the IPv4 default route bound to the interface.
425 Defaults to false. This is useful when creating routes on point-to-point interfaces. This is
426 equivalent to e.g. the following,
427 <programlisting>ip route add default dev veth99
</programlisting>
429 <programlisting>[Route]
430 Gateway=
0.0.0.0</programlisting></para>
431 <para>Currently, there are no way to specify e.g., the table for the route configured by this
432 setting. To configure the default route with such an additional property, please use the
434 <programlisting>[Route]
436 Table=
1234</programlisting></para>
437 <para>If you'd like to create an IPv6 default route bound to the interface, please use the
439 <programlisting>[Route]
441 Table=
1234</programlisting></para>
446 <term><varname>LLMNR=
</varname></term>
448 <para>Takes a boolean or
<literal>resolve
</literal>. When true, enables
449 <ulink url=
"https://tools.ietf.org/html/rfc4795">Link-Local Multicast Name Resolution
</ulink>
450 on the link. When set to
<literal>resolve
</literal>, only resolution is enabled, but not host
451 registration and announcement. Defaults to true. This setting is read by
452 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
458 <term><varname>MulticastDNS=
</varname></term>
460 <para>Takes a boolean or
<literal>resolve
</literal>. When true, enables
461 <ulink url=
"https://tools.ietf.org/html/rfc6762">Multicast DNS
</ulink> support on the link.
462 When set to
<literal>resolve
</literal>, only resolution is enabled, but not host or service
463 registration and announcement. Defaults to false. This setting is read by
464 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
470 <term><varname>DNSOverTLS=
</varname></term>
472 <para>Takes a boolean or
<literal>opportunistic
</literal>. When true, enables
473 <ulink url=
"https://tools.ietf.org/html/rfc7858">DNS-over-TLS
</ulink> support on the link.
474 When set to
<literal>opportunistic
</literal>, compatibility with non-DNS-over-TLS servers is
475 increased, by automatically turning off DNS-over-TLS servers in this case. This option
476 defines a per-interface setting for
477 <citerefentry><refentrytitle>resolved.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>'s
478 global
<varname>DNSOverTLS=
</varname> option. Defaults to unset, and the global setting will
479 be used. This setting is read by
480 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
486 <term><varname>DNSSEC=
</varname></term>
488 <para>Takes a boolean or
<literal>allow-downgrade
</literal>. When true, enables
489 <ulink url=
"https://tools.ietf.org/html/rfc4033">DNSSEC
</ulink> DNS validation support on the
490 link. When set to
<literal>allow-downgrade
</literal>, compatibility with non-DNSSEC capable
491 networks is increased, by automatically turning off DNSSEC in this case. This option defines
492 a per-interface setting for
493 <citerefentry><refentrytitle>resolved.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>'s
494 global
<varname>DNSSEC=
</varname> option. Defaults to unset, and the global setting will be
495 used. This setting is read by
496 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
502 <term><varname>DNSSECNegativeTrustAnchors=
</varname></term>
504 <para>A space-separated list of DNSSEC negative trust anchor domains. If specified and DNSSEC
505 is enabled, look-ups done via the interface's DNS server will be subject to the list of
506 negative trust anchors, and not require authentication for the specified domains, or anything
507 below it. Use this to disable DNSSEC authentication for specific private domains, that cannot
508 be proven valid using the Internet DNS hierarchy. Defaults to the empty list. This setting is
510 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
516 <term><varname>LLDP=
</varname></term>
518 <para>Controls support for Ethernet LLDP packet reception. LLDP is a link-layer protocol
519 commonly implemented on professional routers and bridges which announces which physical port
520 a system is connected to, as well as other related data. Accepts a boolean or the special
521 value
<literal>routers-only
</literal>. When true, incoming LLDP packets are accepted and a
522 database of all LLDP neighbors maintained. If
<literal>routers-only
</literal> is set only
523 LLDP data of various types of routers is collected and LLDP data about other types of devices
524 ignored (such as stations, telephones and others). If false, LLDP reception is disabled.
525 Defaults to
<literal>routers-only
</literal>. Use
526 <citerefentry><refentrytitle>networkctl
</refentrytitle><manvolnum>1</manvolnum></citerefentry>
527 to query the collected neighbor data. LLDP is only available on Ethernet links. See
528 <varname>EmitLLDP=
</varname> below for enabling LLDP packet emission from the local system.
534 <term><varname>EmitLLDP=
</varname></term>
536 <para>Controls support for Ethernet LLDP packet emission. Accepts a boolean parameter or the
537 special values
<literal>nearest-bridge
</literal>,
<literal>non-tpmr-bridge
</literal> and
538 <literal>customer-bridge
</literal>. Defaults to false, which turns off LLDP packet emission.
539 If not false, a short LLDP packet with information about the local system is sent out in
540 regular intervals on the link. The LLDP packet will contain information about the local
541 hostname, the local machine ID (as stored in
542 <citerefentry><refentrytitle>machine-id
</refentrytitle><manvolnum>5</manvolnum></citerefentry>)
543 and the local interface name, as well as the pretty hostname of the system (as set in
544 <citerefentry><refentrytitle>machine-info
</refentrytitle><manvolnum>5</manvolnum></citerefentry>).
545 LLDP emission is only available on Ethernet links. Note that this setting passes data
546 suitable for identification of host to the network and should thus not be enabled on
547 untrusted networks, where such identification data should not be made available. Use this
548 option to permit other systems to identify on which interfaces they are connected to this
549 system. The three special values control propagation of the LLDP packets. The
550 <literal>nearest-bridge
</literal> setting permits propagation only to the nearest connected
551 bridge,
<literal>non-tpmr-bridge
</literal> permits propagation across Two-Port MAC Relays,
552 but not any other bridges, and
<literal>customer-bridge
</literal> permits propagation until
553 a customer bridge is reached. For details about these concepts, see
554 <ulink url=
"https://standards.ieee.org/findstds/standard/802.1AB-2016.html">IEEE
802.1AB-
2016</ulink>.
555 Note that configuring this setting to true is equivalent to
556 <literal>nearest-bridge
</literal>, the recommended and most restricted level of propagation.
557 See
<varname>LLDP=
</varname> above for an option to enable LLDP reception.
</para>
562 <term><varname>BindCarrier=
</varname></term>
564 <para>A link name or a list of link names. When set, controls the behavior of the current
565 link. When all links in the list are in an operational down state, the current link is
566 brought down. When at least one link has carrier, the current interface is brought up.
</para>
568 <para>This forces
<varname>ActivationPolicy=
</varname> to be set to
<literal>bound
</literal>.
574 <term><varname>Address=
</varname></term>
576 <para>A static IPv4 or IPv6 address and its prefix length, separated by a
577 <literal>/
</literal> character. Specify this key more than once to configure several
578 addresses. The format of the address must be as described in
579 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
580 This is a short-hand for an [Address] section only containing an Address key (see below).
581 This option may be specified more than once.
</para>
583 <para>If the specified address is
<literal>0.0.0.0</literal> (for IPv4) or
584 <literal>::
</literal> (for IPv6), a new address range of the requested size is automatically
585 allocated from a system-wide pool of unused ranges. Note that the prefix length must be equal
586 or larger than
8 for IPv4, and
64 for IPv6. The allocated range is checked against all
587 current network interfaces and all known network configuration files to avoid address range
588 conflicts. The default system-wide pool consists of
192.168.0.0/
16,
172.16.0.0/
12 and
589 10.0.0.0/
8 for IPv4, and fd00::/
8 for IPv6. This functionality is useful to manage a large
590 number of dynamically created network interfaces with the same network configuration and
591 automatic address range assignment.
</para>
596 <term><varname>Gateway=
</varname></term>
598 <para>The gateway address, which must be in the format described in
599 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
600 This is a short-hand for a [Route] section only containing a
<varname>Gateway=
</varname> key.
601 This option may be specified more than once.
</para>
606 <term><varname>DNS=
</varname></term>
608 <para>A DNS server address, which must be in the format described in
609 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
610 This option may be specified more than once. Each address can optionally take a port number
611 separated with
<literal>:
</literal>, a network interface name or index separated with
612 <literal>%
</literal>, and a Server Name Indication (SNI) separated with
<literal>#
</literal>.
613 When IPv6 address is specified with a port number, then the address must be in the square
614 brackets. That is, the acceptable full formats are
615 <literal>111.222.333.444:
9953%ifname#example.com
</literal> for IPv4 and
616 <literal>[
1111:
2222::
3333]:
9953%ifname#example.com
</literal> for IPv6. If an empty string is
617 assigned, then the all previous assignments are cleared. This setting is read by
618 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
624 <term><varname>Domains=
</varname></term>
626 <para>A whitespace-separated list of domains which should be resolved using the DNS servers
627 on this link. Each item in the list should be a domain name, optionally prefixed with a tilde
628 (
<literal>~
</literal>). The domains with the prefix are called
"routing-only domains". The
629 domains without the prefix are called
"search domains" and are first used as search suffixes
630 for extending single-label hostnames (hostnames containing no dots) to become fully qualified
631 domain names (FQDNs). If a single-label hostname is resolved on this interface, each of the
632 specified search domains are appended to it in turn, converting it into a fully qualified
633 domain name, until one of them may be successfully resolved.
</para>
635 <para>Both
"search" and
"routing-only" domains are used for routing of DNS queries: look-ups
636 for hostnames ending in those domains (hence also single label names, if any
"search domains"
637 are listed), are routed to the DNS servers configured for this interface. The domain routing
638 logic is particularly useful on multi-homed hosts with DNS servers serving particular private
639 DNS zones on each interface.
</para>
641 <para>The
"routing-only" domain
<literal>~.
</literal> (the tilde indicating definition of a
642 routing domain, the dot referring to the DNS root domain which is the implied suffix of all
643 valid DNS names) has special effect. It causes all DNS traffic which does not match another
644 configured domain routing entry to be routed to DNS servers specified for this interface.
645 This setting is useful to prefer a certain set of DNS servers if a link on which they are
646 connected is available.
</para>
648 <para>This setting is read by
649 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
650 "Search domains" correspond to the
<varname>domain
</varname> and
<varname>search
</varname>
652 <citerefentry project='man-pages'
><refentrytitle>resolv.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
653 Domain name routing has no equivalent in the traditional glibc API, which has no concept of
654 domain name servers limited to a specific link.
</para>
659 <term><varname>DNSDefaultRoute=
</varname></term>
661 <para>Takes a boolean argument. If true, this link's configured DNS servers are used for
662 resolving domain names that do not match any link's configured
<varname>Domains=
</varname>
663 setting. If false, this link's configured DNS servers are never used for such domains, and
664 are exclusively used for resolving names that match at least one of the domains configured on
665 this link. If not specified defaults to an automatic mode: queries not matching any link's
666 configured domains will be routed to this link if it has no routing-only domains configured.
672 <term><varname>NTP=
</varname></term>
674 <para>An NTP server address (either an IP address, or a hostname). This option may be
675 specified more than once. This setting is read by
676 <citerefentry><refentrytitle>systemd-timesyncd.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
682 <term><varname>IPForward=
</varname></term>
684 <para>Configures IP packet forwarding for the system. If enabled, incoming packets on any
685 network interface will be forwarded to any other interfaces according to the routing table.
686 Takes a boolean, or the values
<literal>ipv4
</literal> or
<literal>ipv6
</literal>, which only
687 enable IP packet forwarding for the specified address family. This controls the
688 <filename>net.ipv4.ip_forward
</filename> and
<filename>net.ipv6.conf.all.forwarding
</filename>
689 sysctl options of the network interface (see
690 <ulink url=
"https://docs.kernel.org/networking/ip-sysctl.html">IP Sysctl
</ulink>
691 for details about sysctl options). Defaults to
<literal>no
</literal>.
</para>
693 <para>Note: this setting controls a global kernel option, and does so one way only: if a
694 network that has this setting enabled is set up the global setting is turned on. However,
695 it is never turned off again, even after all networks with this setting enabled are shut
698 <para>To allow IP packet forwarding only between specific network interfaces use a firewall.
704 <term><varname>IPMasquerade=
</varname></term>
706 <para>Configures IP masquerading for the network interface. If enabled, packets forwarded
707 from the network interface will be appear as coming from the local host. Takes one of
708 <literal>ipv4
</literal>,
<literal>ipv6
</literal>,
<literal>both
</literal>, or
709 <literal>no
</literal>. Defaults to
<literal>no
</literal>. If enabled, this automatically sets
710 <varname>IPForward=
</varname> to one of
<literal>ipv4
</literal>,
<literal>ipv6
</literal> or
711 <literal>yes
</literal>.
</para>
712 <para>Note. Any positive boolean values such as
<literal>yes
</literal> or
713 <literal>true
</literal> are now deprecated. Please use one of the values in the above.
</para>
718 <term><varname>IPv6PrivacyExtensions=
</varname></term>
720 <para>Configures use of stateless temporary addresses that change over time (see
721 <ulink url=
"https://tools.ietf.org/html/rfc4941">RFC
4941</ulink>,
722 Privacy Extensions for Stateless Address Autoconfiguration in IPv6). Takes a boolean or the
723 special values
<literal>prefer-public
</literal> and
<literal>kernel
</literal>. When true,
724 enables the privacy extensions and prefers temporary addresses over public addresses. When
725 <literal>prefer-public
</literal>, enables the privacy extensions, but prefers public
726 addresses over temporary addresses. When false, the privacy extensions remain disabled. When
727 <literal>kernel
</literal>, the kernel's default setting will be left in place. Defaults to
728 <literal>no
</literal>.
</para>
733 <term><varname>IPv6AcceptRA=
</varname></term>
735 <para>Takes a boolean. Controls IPv6 Router Advertisement (RA) reception support for the
736 interface. If true, RAs are accepted; if false, RAs are ignored. When RAs are accepted, they
737 may trigger the start of the DHCPv6 client if the relevant flags are set in the RA data, or
738 if no routers are found on the link. The default is to disable RA reception for bridge
739 devices or when IP forwarding is enabled, and to enable it otherwise. Cannot be enabled on
740 bond devices and when link-local addressing is disabled.
</para>
742 <para>Further settings for the IPv6 RA support may be configured in the [IPv6AcceptRA]
743 section, see below.
</para>
746 <ulink url=
"https://docs.kernel.org/networking/ip-sysctl.html">IP Sysctl
</ulink>
747 in the kernel documentation regarding
<literal>accept_ra
</literal>, but note that systemd's
748 setting of
<constant>1</constant> (i.e. true) corresponds to kernel's setting of
749 <constant>2</constant>.
</para>
751 <para>Note that kernel's implementation of the IPv6 RA protocol is always disabled,
752 regardless of this setting. If this option is enabled, a userspace implementation of the IPv6
753 RA protocol is used, and the kernel's own implementation remains disabled, since
754 <command>systemd-networkd
</command> needs to know all details supplied in the advertisements,
755 and these are not available from the kernel if the kernel's own implementation is used.
761 <term><varname>IPv6DuplicateAddressDetection=
</varname></term>
763 <para>Configures the amount of IPv6 Duplicate Address Detection (DAD) probes to send. When
764 unset, the kernel's default will be used.
</para>
769 <term><varname>IPv6HopLimit=
</varname></term>
771 <para>Configures IPv6 Hop Limit. For each router that forwards the packet, the hop limit is
772 decremented by
1. When the hop limit field reaches zero, the packet is discarded. When unset,
773 the kernel's default will be used.
</para>
778 <term><varname>IPv4AcceptLocal=
</varname></term>
780 <para>Takes a boolean. Accept packets with local source addresses. In combination with
781 suitable routing, this can be used to direct packets between two local interfaces over the
782 wire and have them accepted properly. When unset, the kernel's default will be used.
</para>
787 <term><varname>IPv4RouteLocalnet=
</varname></term>
789 <para>Takes a boolean. When true, the kernel does not consider loopback addresses as martian
790 source or destination while routing. This enables the use of
127.0.0.0/
8 for local routing
791 purposes. When unset, the kernel's default will be used.
</para>
796 <term><varname>IPv4ProxyARP=
</varname></term>
798 <para>Takes a boolean. Configures proxy ARP for IPv4. Proxy ARP is the technique in which one
799 host, usually a router, answers ARP requests intended for another machine. By
"faking" its
800 identity, the router accepts responsibility for routing packets to the
"real" destination.
801 See
<ulink url=
"https://tools.ietf.org/html/rfc1027">RFC
1027</ulink>. When unset, the
802 kernel's default will be used.
</para>
807 <term><varname>IPv6ProxyNDP=
</varname></term>
809 <para>Takes a boolean. Configures proxy NDP for IPv6. Proxy NDP (Neighbor Discovery Protocol)
810 is a technique for IPv6 to allow routing of addresses to a different destination when peers
811 expect them to be present on a certain physical link. In this case a router answers Neighbour
812 Advertisement messages intended for another machine by offering its own MAC address as
813 destination. Unlike proxy ARP for IPv4, it is not enabled globally, but will only send
814 Neighbour Advertisement messages for addresses in the IPv6 neighbor proxy table, which can
815 also be shown by
<command>ip -
6 neighbour show proxy
</command>. systemd-networkd will control
816 the per-interface `proxy_ndp` switch for each configured interface depending on this option.
817 When unset, the kernel's default will be used.
</para>
822 <term><varname>IPv6ProxyNDPAddress=
</varname></term>
824 <para>An IPv6 address, for which Neighbour Advertisement messages will be proxied. This
825 option may be specified more than once. systemd-networkd will add the
826 <varname>IPv6ProxyNDPAddress=
</varname> entries to the kernel's IPv6 neighbor proxy table.
827 This setting implies
<varname>IPv6ProxyNDP=yes
</varname> but has no effect if
828 <varname>IPv6ProxyNDP=
</varname> has been set to false. When unset, the kernel's default will
834 <term><varname>IPv6SendRA=
</varname></term>
836 <para>Whether to enable or disable Router Advertisement sending on a link. Takes a boolean
837 value. When enabled, prefixes configured in [IPv6Prefix] sections and routes configured in
838 the [IPv6RoutePrefix] sections are distributed as defined in the [IPv6SendRA] section. If
839 <varname>DHCPPrefixDelegation=
</varname> is enabled, then the delegated prefixes are also
840 distributed. See
<varname>DCHPPrefixDelegation=
</varname> setting and the [IPv6SendRA],
841 [IPv6Prefix], [IPv6RoutePrefix], and [DHCPPrefixDelegation] sections for more configuration
847 <term><varname>DHCPPrefixDelegation=
</varname></term>
849 <para>Takes a boolean value. When enabled, requests subnet prefixes on another link via the DHCPv6
850 protocol or via the
6RD option in the DHCPv4 protocol. An address within each delegated prefix will
851 be assigned, and the prefixes will be announced through IPv6 Router Advertisement if
852 <varname>IPv6SendRA=
</varname> is enabled. This behaviour can be configured in the
853 [DHCPPrefixDelegation] section. Defaults to disabled.
</para>
858 <term><varname>IPv6MTUBytes=
</varname></term>
860 <para>Configures IPv6 maximum transmission unit (MTU). An integer greater than or equal to
861 1280 bytes. When unset, the kernel's default will be used.
</para>
866 <term><varname>KeepMaster=
</varname></term>
868 <para>Takes a boolean value. When enabled, the current master interface index will not be
869 changed, and
<varname>BatmanAdvanced=
</varname>,
<varname>Bond=
</varname>,
870 <varname>Bridge=
</varname>, and
<varname>VRF=
</varname> settings are ignored. This may be
871 useful when a netdev with a master interface is created by another program, e.g.
872 <citerefentry><refentrytitle>systemd-nspawn
</refentrytitle><manvolnum>1</manvolnum></citerefentry>.
873 Defaults to false.
</para>
878 <term><varname>BatmanAdvanced=
</varname></term>
879 <term><varname>Bond=
</varname></term>
880 <term><varname>Bridge=
</varname></term>
881 <term><varname>VRF=
</varname></term>
883 <para>The name of the B.A.T.M.A.N. Advanced, bond, bridge, or VRF interface to add the link
885 <citerefentry><refentrytitle>systemd.netdev
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
891 <term><varname>IPoIB=
</varname></term>
892 <term><varname>IPVLAN=
</varname></term>
893 <term><varname>IPVTAP=
</varname></term>
894 <term><varname>MACsec=
</varname></term>
895 <term><varname>MACVLAN=
</varname></term>
896 <term><varname>MACVTAP=
</varname></term>
897 <term><varname>Tunnel=
</varname></term>
898 <term><varname>VLAN=
</varname></term>
899 <term><varname>VXLAN=
</varname></term>
900 <term><varname>Xfrm=
</varname></term>
902 <para>The name of an IPoIB, IPVLAN, IPVTAP, MACsec, MACVLAN, MACVTAP, tunnel, VLAN,
903 VXLAN, or Xfrm to be created on the link. See
904 <citerefentry><refentrytitle>systemd.netdev
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
905 This option may be specified more than once.
</para>
910 <term><varname>ActiveSlave=
</varname></term>
912 <para>Takes a boolean. Specifies the new active slave. The
<literal>ActiveSlave=
</literal>
913 option is only valid for following modes:
<literal>active-backup
</literal>,
914 <literal>balance-alb
</literal>, and
<literal>balance-tlb
</literal>. Defaults to false.
</para>
919 <term><varname>PrimarySlave=
</varname></term>
921 <para>Takes a boolean. Specifies which slave is the primary device. The specified device will
922 always be the active slave while it is available. Only when the primary is off-line will
923 alternate devices be used. This is useful when one slave is preferred over another, e.g.
924 when one slave has higher throughput than another. The
<literal>PrimarySlave=
</literal>
925 option is only valid for following modes:
<literal>active-backup
</literal>,
926 <literal>balance-alb
</literal>, and
<literal>balance-tlb
</literal>. Defaults to false.
</para>
931 <term><varname>ConfigureWithoutCarrier=
</varname></term>
933 <para>Takes a boolean. Allows networkd to configure a specific link even if it has no
934 carrier. Defaults to false. If enabled, and the
<varname>IgnoreCarrierLoss=
</varname> setting
935 is not explicitly set, then it is enabled as well.
</para>
940 <term><varname>IgnoreCarrierLoss=
</varname></term>
942 <para>Takes a boolean or a timespan. When true,
<command>systemd-networkd
</command> retains
943 both the static and dynamic configuration of the interface even if its carrier is lost. When
944 false,
<command>systemd-networkd
</command> drops both the static and dynamic configuration of
945 the interface. When a timespan is specified,
<command>systemd-networkd
</command> waits for
946 the specified timespan, and ignores the carrier loss if the link regain its carrier within
947 the timespan. Setting
0 seconds is equivalent to
<literal>no
</literal>, and
948 <literal>infinite
</literal> is equivalent to
<literal>yes
</literal>.
</para>
950 <para>Setting a finite timespan may be useful when e.g. in the following cases:
953 <para>A wireless interface connecting to a network which has multiple access points with
954 the same SSID.
</para>
957 <para>Enslaving a wireless interface to a bond interface, which may disconnect from the
958 connected access point and causes its carrier to be lost.
</para>
961 <para>The driver of the interface resets when the MTU is changed.
</para>
966 <para>When
<varname>Bond=
</varname> is specified to a wireless interface, defaults to
3
967 seconds. When the DHCPv4 client is enabled and
<varname>UseMTU=
</varname> in the [DHCPv4]
968 section enabled, defaults to
5 seconds. Otherwise, defaults to the value specified with
969 <varname>ConfigureWithoutCarrier=
</varname>. When
<varname>ActivationPolicy=
</varname> is set
970 to
<literal>always-up
</literal>, this is forced to
<literal>yes
</literal>, and ignored any
971 user specified values.
</para>
976 <term><varname>KeepConfiguration=
</varname></term>
978 <para>Takes a boolean or one of
<literal>static
</literal>,
<literal>dhcp-on-stop
</literal>,
979 <literal>dhcp
</literal>. When
<literal>static
</literal>,
<command>systemd-networkd
</command>
980 will not drop static addresses and routes on starting up process. When set to
981 <literal>dhcp-on-stop
</literal>,
<command>systemd-networkd
</command> will not drop addresses
982 and routes on stopping the daemon. When
<literal>dhcp
</literal>,
983 the addresses and routes provided by a DHCP server will never be dropped even if the DHCP
984 lease expires. This is contrary to the DHCP specification, but may be the best choice if,
985 e.g., the root filesystem relies on this connection. The setting
<literal>dhcp
</literal>
986 implies
<literal>dhcp-on-stop
</literal>, and
<literal>yes
</literal> implies
987 <literal>dhcp
</literal> and
<literal>static
</literal>. Defaults to
988 <literal>dhcp-on-stop
</literal> when
<command>systemd-networkd
</command> is running in
989 initrd,
<literal>yes
</literal> when the root filesystem is a network filesystem, and
990 <literal>no
</literal> otherwise.
</para>
997 <title>[Address] Section Options
</title>
999 <para>An [Address] section accepts the following keys. Specify several [Address] sections to
1000 configure several addresses.
</para>
1002 <variablelist class='network-directives'
>
1004 <term><varname>Address=
</varname></term>
1006 <para>As in the [Network] section. This setting is mandatory. Each [Address] section can
1007 contain one
<varname>Address=
</varname> setting.
</para>
1012 <term><varname>Peer=
</varname></term>
1014 <para>The peer address in a point-to-point connection. Accepts the same format as the
1015 <varname>Address=
</varname> setting.
</para>
1020 <term><varname>Broadcast=
</varname></term>
1022 <para>Takes an IPv4 address or boolean value. The address must be in the format described in
1023 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
1024 If set to true, then the IPv4 broadcast address will be derived from the
1025 <varname>Address=
</varname> setting. If set to false, then the broadcast address will not be
1026 set. Defaults to true, except for wireguard interfaces, where it default to false.
</para>
1031 <term><varname>Label=
</varname></term>
1033 <para>Specifies the label for the IPv4 address. The label must be a
7-bit ASCII string with
1034 a length of
1…
15 characters. Defaults to unset.
</para>
1039 <term><varname>PreferredLifetime=
</varname></term>
1041 <para>Allows the default
"preferred lifetime" of the address to be overridden. Only three
1042 settings are accepted:
<literal>forever
</literal>,
<literal>infinity
</literal>, which is the
1043 default and means that the address never expires, and
<literal>0</literal>, which means that
1044 the address is considered immediately
"expired" and will not be used, unless explicitly
1045 requested. A setting of
<option>PreferredLifetime=
0</option> is useful for addresses which
1046 are added to be used only by a specific application, which is then configured to use them
1052 <term><varname>Scope=
</varname></term>
1054 <para>The scope of the address, which can be
<literal>global
</literal> (valid everywhere on
1055 the network, even through a gateway),
<literal>link
</literal> (only valid on this device,
1056 will not traverse a gateway) or
<literal>host
</literal> (only valid within the device itself,
1057 e.g.
127.0.0.1) or an integer in the range
0…
255. Defaults to
<literal>global
</literal>.
1063 <term><varname>RouteMetric=
</varname></term>
1065 <para>The metric of the prefix route, which is pointing to the subnet of the configured IP
1066 address, taking the configured prefix length into account. Takes an unsigned integer in the
1067 range
0…
4294967295. When unset or set to
0, the kernel's default value is used. This
1068 setting will be ignored when
<varname>AddPrefixRoute=
</varname> is false.
</para>
1073 <term><varname>HomeAddress=
</varname></term>
1075 <para>Takes a boolean. Designates this address the
"home address" as defined in
1076 <ulink url=
"https://tools.ietf.org/html/rfc6275">RFC
6275</ulink>. Supported only on IPv6.
1077 Defaults to false.
</para>
1082 <term><varname>DuplicateAddressDetection=
</varname></term>
1084 <para>Takes one of
<literal>ipv4
</literal>,
<literal>ipv6
</literal>,
<literal>both
</literal>,
1085 or
<literal>none
</literal>. When
<literal>ipv4
</literal>, performs IPv4 Address Conflict
1086 Detection. See
<ulink url=
"https://tools.ietf.org/html/rfc5227">RFC
5227</ulink>.
1087 When
<literal>ipv6
</literal>, performs IPv6 Duplicate Address Detection. See
1088 <ulink url=
"https://tools.ietf.org/html/rfc4862">RFC
4862</ulink>. Defaults to
1089 <literal>ipv4
</literal> for IPv4 link-local addresses,
<literal>ipv6
</literal> for IPv6
1090 addresses, and
<literal>none
</literal> otherwise.
</para>
1095 <term><varname>ManageTemporaryAddress=
</varname></term>
1097 <para>Takes a boolean. If true the kernel manage temporary addresses created from this one as
1098 template on behalf of Privacy Extensions
1099 <ulink url=
"https://tools.ietf.org/html/rfc3041">RFC
3041</ulink>. For this to become active,
1100 the use_tempaddr sysctl setting has to be set to a value greater than zero. The given address
1101 needs to have a prefix length of
64. This flag allows using privacy extensions in a manually
1102 configured network, just like if stateless auto-configuration was active. Defaults to false.
1108 <term><varname>AddPrefixRoute=
</varname></term>
1110 <para>Takes a boolean. When true, the prefix route for the address is automatically added.
1111 Defaults to true.
</para>
1116 <term><varname>AutoJoin=
</varname></term>
1118 <para>Takes a boolean. Joining multicast group on ethernet level via
1119 <command>ip maddr
</command> command would not work if we have an Ethernet switch that does
1120 IGMP snooping since the switch would not replicate multicast packets on ports that did not
1121 have IGMP reports for the multicast addresses. Linux vxlan interfaces created via
1122 <command>ip link add vxlan
</command> or networkd's netdev kind vxlan have the group option
1123 that enables them to do the required join. By extending
<command>ip address
</command> command
1124 with option
<literal>autojoin
</literal> we can get similar functionality for openvswitch (OVS)
1125 vxlan interfaces as well as other tunneling mechanisms that need to receive multicast traffic.
1126 Defaults to
<literal>no
</literal>.
</para>
1131 <term><varname>NetLabel=
</varname><replaceable>label
</replaceable></term>
1134 <para>This setting provides a method for integrating static and dynamic network configuration into
1135 Linux
<ulink url=
"https://docs.kernel.org/netlabel/index.html">NetLabel
</ulink> subsystem rules,
1136 used by
<ulink url=
"https://en.wikipedia.org/wiki/Linux_Security_Modules">Linux Security Modules
1137 (LSMs)
</ulink> for network access control. The label, with suitable LSM rules, can be used to
1138 control connectivity of (for example) a service with peers in the local network. At least with
1139 SELinux, only the ingress can be controlled but not egress. The benefit of using this setting is
1140 that it may be possible to apply interface independent part of NetLabel configuration at very early
1141 stage of system boot sequence, at the time when the network interfaces are not available yet, with
1143 project='man-pages'
><refentrytitle>netlabelctl
</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
1144 and the per-interface configuration with
<command>systemd-networkd
</command> once the interfaces
1145 appear later. Currently this feature is only implemented for SELinux.
</para>
1147 <para>The option expects a single NetLabel label. The label must conform to lexical restrictions of
1148 LSM labels. When an interface is configured with IP addresses, the addresses and subnetwork masks
1149 will be appended to the
<ulink
1150 url=
"https://github.com/SELinuxProject/selinux-notebook/blob/main/src/network_support.md">NetLabel
1151 Fallback Peer Labeling
</ulink> rules. They will be removed when the interface is
1152 deconfigured. Failures to manage the labels will be ignored.
</para>
1154 <para>Warning: Once labeling is enabled for network traffic, a lot of LSM access control points in
1155 Linux networking stack go from dormant to active. Care should be taken to avoid getting into a
1156 situation where for example remote connectivity is broken, when the security policy hasn't been
1157 updated to consider LSM per-packet access controls and no rules would allow any network
1158 traffic. Also note that additional configuration with
<citerefentry
1159 project='man-pages'
><refentrytitle>netlabelctl
</refentrytitle><manvolnum>8</manvolnum></citerefentry>
1163 <programlisting>[Address]
1164 NetLabel=system_u:object_r:localnet_peer_t:s0
</programlisting>
1166 With the example rules applying for interface
<literal>eth0
</literal>, when the interface is
1167 configured with an IPv4 address of
10.0.0.123/
8,
<command>systemd-networkd
</command> performs the
1168 equivalent of
<command>netlabelctl
</command> operation
1170 <programlisting>netlabelctl unlbl add interface eth0 address:
10.0.0.0/
8 label:system_u:object_r:localnet_peer_t:s0
</programlisting>
1172 and the reverse operation when the IPv4 address is deconfigured. The configuration can be used with
1173 LSM rules; in case of SELinux to allow a SELinux domain to receive data from objects of SELinux
1174 <literal>peer
</literal> class. For example:
1176 <programlisting>type localnet_peer_t;
1177 allow my_server_t localnet_peer_t:peer recv;
</programlisting>
1179 The effect of the above configuration and rules (in absence of other rules as may be the case) is
1180 to only allow
<literal>my_server_t
</literal> (and nothing else) to receive data from local subnet
1181 10.0.0.0/
8 of interface
<literal>eth0
</literal>.
1189 <title>[Neighbor] Section Options
</title>
1191 <para>A [Neighbor] section accepts the following keys. The neighbor section adds a permanent,
1192 static entry to the neighbor table (IPv6) or ARP table (IPv4) for the given hardware address on the
1193 links matched for the network. Specify several [Neighbor] sections to configure several static
1196 <variablelist class='network-directives'
>
1198 <term><varname>Address=
</varname></term>
1200 <para>The IP address of the neighbor.
</para>
1205 <term><varname>LinkLayerAddress=
</varname></term>
1207 <para>The link layer address (MAC address or IP address) of the neighbor.
</para>
1214 <title>[IPv6AddressLabel] Section Options
</title>
1216 <para>An [IPv6AddressLabel] section accepts the following keys. Specify several [IPv6AddressLabel]
1217 sections to configure several address labels. IPv6 address labels are used for address selection.
1218 See
<ulink url=
"https://tools.ietf.org/html/rfc3484">RFC
3484</ulink>. Precedence is managed by
1219 userspace, and only the label itself is stored in the kernel.
</para>
1221 <variablelist class='network-directives'
>
1223 <term><varname>Label=
</varname></term>
1225 <para>The label for the prefix, an unsigned integer in the range
0…
4294967294.
0xffffffff is
1226 reserved. This setting is mandatory.
</para>
1231 <term><varname>Prefix=
</varname></term>
1233 <para>IPv6 prefix is an address with a prefix length, separated by a slash
1234 <literal>/
</literal> character. This setting is mandatory.
</para>
1241 <title>[RoutingPolicyRule] Section Options
</title>
1243 <para>An [RoutingPolicyRule] section accepts the following settings. Specify several
1244 [RoutingPolicyRule] sections to configure several rules.
</para>
1246 <variablelist class='network-directives'
>
1248 <term><varname>TypeOfService=
</varname></term>
1250 <para>Takes a number between
0 and
255 that specifies the type of service to match.
</para>
1255 <term><varname>From=
</varname></term>
1257 <para>Specifies the source address prefix to match. Possibly followed by a slash and the
1258 prefix length.
</para>
1263 <term><varname>To=
</varname></term>
1265 <para>Specifies the destination address prefix to match. Possibly followed by a slash and the
1266 prefix length.
</para>
1271 <term><varname>FirewallMark=
</varname></term>
1273 <para>Specifies the iptables firewall mark value to match (a number in the range
1274 1…
4294967295). Optionally, the firewall mask (also a number between
1…
4294967295) can be
1275 suffixed with a slash (
<literal>/
</literal>), e.g.,
<literal>7/
255</literal>.
</para>
1280 <term><varname>Table=
</varname></term>
1282 <para>Specifies the routing table identifier to lookup if the rule selector matches. Takes
1283 one of predefined names
<literal>default
</literal>,
<literal>main
</literal>, and
1284 <literal>local
</literal>, and names defined in
<varname>RouteTable=
</varname> in
1285 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1286 or a number between
1 and
4294967295. Defaults to
<literal>main
</literal>.
</para>
1291 <term><varname>Priority=
</varname></term>
1293 <para>Specifies the priority of this rule.
<varname>Priority=
</varname> is an integer in the
1294 range
0…
4294967295. Higher number means lower priority, and rules get processed in order of
1295 increasing number. Defaults to unset, and the kernel will pick a value dynamically.
</para>
1300 <term><varname>IncomingInterface=
</varname></term>
1302 <para>Specifies incoming device to match. If the interface is loopback, the rule only matches
1303 packets originating from this host.
</para>
1308 <term><varname>OutgoingInterface=
</varname></term>
1310 <para>Specifies the outgoing device to match. The outgoing interface is only available for
1311 packets originating from local sockets that are bound to a device.
</para>
1316 <term><varname>SourcePort=
</varname></term>
1318 <para>Specifies the source IP port or IP port range match in forwarding information base
1319 (FIB) rules. A port range is specified by the lower and upper port separated by a dash.
1320 Defaults to unset.
</para>
1325 <term><varname>DestinationPort=
</varname></term>
1327 <para>Specifies the destination IP port or IP port range match in forwarding information base
1328 (FIB) rules. A port range is specified by the lower and upper port separated by a dash.
1329 Defaults to unset.
</para>
1334 <term><varname>IPProtocol=
</varname></term>
1336 <para>Specifies the IP protocol to match in forwarding information base (FIB) rules. Takes IP
1337 protocol name such as
<literal>tcp
</literal>,
<literal>udp
</literal> or
1338 <literal>sctp
</literal>, or IP protocol number such as
<literal>6</literal> for
1339 <literal>tcp
</literal> or
<literal>17</literal> for
<literal>udp
</literal>. Defaults to unset.
1345 <term><varname>InvertRule=
</varname></term>
1347 <para>A boolean. Specifies whether the rule is to be inverted. Defaults to false.
</para>
1352 <term><varname>Family=
</varname></term>
1354 <para>Takes a special value
<literal>ipv4
</literal>,
<literal>ipv6
</literal>, or
1355 <literal>both
</literal>. By default, the address family is determined by the address
1356 specified in
<varname>To=
</varname> or
<varname>From=
</varname>. If neither
1357 <varname>To=
</varname> nor
<varname>From=
</varname> are specified, then defaults to
1358 <literal>ipv4
</literal>.
</para>
1363 <term><varname>User=
</varname></term>
1365 <para>Takes a username, a user ID, or a range of user IDs separated by a dash. Defaults to
1371 <term><varname>SuppressPrefixLength=
</varname></term>
1373 <para>Takes a number
<replaceable>N
</replaceable> in the range
0…
128 and rejects routing
1374 decisions that have a prefix length of
<replaceable>N
</replaceable> or less. Defaults to
1380 <term><varname>SuppressInterfaceGroup=
</varname></term>
1382 <para>Takes an integer in the range
0…
2147483647 and rejects routing decisions that have
1383 an interface with the same group id. It has the same meaning as
1384 <option>suppress_ifgroup
</option> in
<command>ip rule
</command>. Defaults to unset.
</para>
1389 <term><varname>Type=
</varname></term>
1391 <para>Specifies Routing Policy Database (RPDB) rule type. Takes one of
1392 <literal>blackhole
</literal>,
<literal>unreachable
</literal> or
<literal>prohibit
</literal>.
1400 <title>[NextHop] Section Options
</title>
1402 <para>The [NextHop] section is used to manipulate entries in the kernel's
"nexthop" tables. The
1403 [NextHop] section accepts the following settings. Specify several [NextHop] sections to configure
1404 several hops.
</para>
1406 <variablelist class='network-directives'
>
1408 <term><varname>Id=
</varname></term>
1410 <para>The id of the next hop. Takes an integer in the range
1…
4294967295. If unspecified,
1411 then automatically chosen by kernel.
</para>
1416 <term><varname>Gateway=
</varname></term>
1418 <para>As in the [Network] section.
</para>
1423 <term><varname>Family=
</varname></term>
1425 <para>Takes one of the special values
<literal>ipv4
</literal> or
<literal>ipv6
</literal>.
1426 By default, the family is determined by the address specified in
1427 <varname>Gateway=
</varname>. If
<varname>Gateway=
</varname> is not specified, then defaults
1428 to
<literal>ipv4
</literal>.
</para>
1433 <term><varname>OnLink=
</varname></term>
1435 <para>Takes a boolean. If set to true, the kernel does not have to check if the gateway is
1436 reachable directly by the current machine (i.e., attached to the local network), so that we
1437 can insert the nexthop in the kernel table without it being complained about. Defaults to
1438 <literal>no
</literal>.
</para>
1443 <term><varname>Blackhole=
</varname></term>
1445 <para>Takes a boolean. If enabled, packets to the corresponding routes are discarded
1446 silently, and
<varname>Gateway=
</varname> cannot be specified. Defaults to
1447 <literal>no
</literal>.
</para>
1452 <term><varname>Group=
</varname></term>
1454 <para>Takes a whitespace separated list of nexthop IDs. Each ID must be in the range
1455 1…
4294967295. Optionally, each nexthop ID can take a weight after a colon
1456 (
<literal><replaceable>id
</replaceable><optional>:
<replaceable>weight
</replaceable></optional></literal>).
1457 The weight must be in the range
1…
255. If the weight is not specified, then it is assumed
1458 that the weight is
1. This setting cannot be specified with
<varname>Gateway=
</varname>,
1459 <varname>Family=
</varname>,
<varname>Blackhole=
</varname>. This setting can be specified
1460 multiple times. If an empty string is assigned, then the all previous assignments are
1461 cleared. Defaults to unset.
</para>
1468 <title>[Route] Section Options
</title>
1470 <para>The [Route] section accepts the following settings. Specify several [Route] sections to
1471 configure several routes.
</para>
1473 <variablelist class='network-directives'
>
1475 <term><varname>Gateway=
</varname></term>
1477 <para>Takes the gateway address or the special values
<literal>_dhcp4
</literal> and
1478 <literal>_ipv6ra
</literal>. If
<literal>_dhcp4
</literal> or
<literal>_ipv6ra
</literal> is
1479 set, then the gateway address provided by DHCPv4 or IPv6 RA is used.
</para>
1484 <term><varname>GatewayOnLink=
</varname></term>
1486 <para>Takes a boolean. If set to true, the kernel does not have to check if the gateway is
1487 reachable directly by the current machine (i.e., attached to the local network), so that we
1488 can insert the route in the kernel table without it being complained about. Defaults to
1489 <literal>no
</literal>.
</para>
1494 <term><varname>Destination=
</varname></term>
1496 <para>The destination prefix of the route. Possibly followed by a slash and the prefix
1497 length. If omitted, a full-length host route is assumed.
</para>
1502 <term><varname>Source=
</varname></term>
1504 <para>The source prefix of the route. Possibly followed by a slash and the prefix length. If
1505 omitted, a full-length host route is assumed.
</para>
1510 <term><varname>Metric=
</varname></term>
1512 <para>The metric of the route. Takes an unsigned integer in the range
0…
4294967295. Defaults
1513 to unset, and the kernel's default will be used.
</para>
1518 <term><varname>IPv6Preference=
</varname></term>
1520 <para>Specifies the route preference as defined in
1521 <ulink url=
"https://tools.ietf.org/html/rfc4191">RFC
4191</ulink> for Router Discovery
1522 messages. Which can be one of
<literal>low
</literal> the route has a lowest priority,
1523 <literal>medium
</literal> the route has a default priority or
<literal>high
</literal> the
1524 route has a highest priority.
</para>
1529 <term><varname>Scope=
</varname></term>
1531 <para>The scope of the IPv4 route, which can be
<literal>global
</literal>,
1532 <literal>site
</literal>,
<literal>link
</literal>,
<literal>host
</literal>, or
1533 <literal>nowhere
</literal>:
</para>
1536 <para><literal>global
</literal> means the route can reach hosts more than one hop away.
1541 <para><literal>site
</literal> means an interior route in the local autonomous system.
1546 <para><literal>link
</literal> means the route can only reach hosts on the local network
1547 (one hop away).
</para>
1551 <para><literal>host
</literal> means the route will not leave the local machine (used for
1552 internal addresses like
127.0.0.1).
</para>
1556 <para><literal>nowhere
</literal> means the destination doesn't exist.
</para>
1560 <para>For IPv4 route, defaults to
<literal>host
</literal> if
<varname>Type=
</varname> is
1561 <literal>local
</literal> or
<literal>nat
</literal>, and
<literal>link
</literal> if
1562 <varname>Type=
</varname> is
<literal>broadcast
</literal>,
<literal>multicast
</literal>,
1563 <literal>anycast
</literal>, or
<literal>unicast
</literal>. In other cases,
1564 defaults to
<literal>global
</literal>. The value is not used for IPv6.
</para>
1569 <term><varname>PreferredSource=
</varname></term>
1571 <para>The preferred source address of the route. The address must be in the format described
1573 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
1579 <term><varname>Table=
</varname></term>
1581 <para>The table identifier for the route. Takes one of predefined names
1582 <literal>default
</literal>,
<literal>main
</literal>, and
<literal>local
</literal>, and names
1583 defined in
<varname>RouteTable=
</varname> in
1584 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
1585 or a number between
1 and
4294967295. The table can be retrieved using
1586 <command>ip route show table
<replaceable>num
</replaceable></command>. If unset and
1587 <varname>Type=
</varname> is
<literal>local
</literal>,
<literal>broadcast
</literal>,
1588 <literal>anycast
</literal>, or
<literal>nat
</literal>, then
<literal>local
</literal> is used.
1589 In other cases, defaults to
<literal>main
</literal>.
</para>
1594 <term><varname>Protocol=
</varname></term>
1596 <para>The protocol identifier for the route. Takes a number between
0 and
255 or the special
1597 values
<literal>kernel
</literal>,
<literal>boot
</literal>,
<literal>static
</literal>,
1598 <literal>ra
</literal> and
<literal>dhcp
</literal>. Defaults to
<literal>static
</literal>.
1604 <term><varname>Type=
</varname></term>
1606 <para>Specifies the type for the route. Takes one of
<literal>unicast
</literal>,
1607 <literal>local
</literal>,
<literal>broadcast
</literal>,
<literal>anycast
</literal>,
1608 <literal>multicast
</literal>,
<literal>blackhole
</literal>,
<literal>unreachable
</literal>,
1609 <literal>prohibit
</literal>,
<literal>throw
</literal>,
<literal>nat
</literal>, and
1610 <literal>xresolve
</literal>. If
<literal>unicast
</literal>, a regular route is defined, i.e.
1611 a route indicating the path to take to a destination network address. If
1612 <literal>blackhole
</literal>, packets to the defined route are discarded silently. If
1613 <literal>unreachable
</literal>, packets to the defined route are discarded and the ICMP
1614 message
"Host Unreachable" is generated. If
<literal>prohibit
</literal>, packets to the
1615 defined route are discarded and the ICMP message
"Communication Administratively Prohibited"
1616 is generated. If
<literal>throw
</literal>, route lookup in the current routing table will
1617 fail and the route selection process will return to Routing Policy Database (RPDB). Defaults
1618 to
<literal>unicast
</literal>.
</para>
1623 <term><varname>InitialCongestionWindow=
</varname></term>
1625 <para>The TCP initial congestion window is used during the start of a TCP connection.
1626 During the start of a TCP session, when a client requests a resource, the server's initial
1627 congestion window determines how many packets will be sent during the initial burst of data
1628 without waiting for acknowledgement. Takes a number between
1 and
1023. Note that
100 is
1629 considered an extremely large value for this option. When unset, the kernel's default
1630 (typically
10) will be used.
</para>
1635 <term><varname>InitialAdvertisedReceiveWindow=
</varname></term>
1637 <para>The TCP initial advertised receive window is the amount of receive data (in bytes)
1638 that can initially be buffered at one time on a connection. The sending host can send only
1639 that amount of data before waiting for an acknowledgment and window update from the
1640 receiving host. Takes a number between
1 and
1023. Note that
100 is considered an extremely
1641 large value for this option. When unset, the kernel's default will be used.
</para>
1646 <term><varname>QuickAck=
</varname></term>
1648 <para>Takes a boolean. When true enables TCP quick ack mode for the route. When unset, the
1649 kernel's default will be used.
</para>
1654 <term><varname>FastOpenNoCookie=
</varname></term>
1656 <para>Takes a boolean. When true enables TCP fastopen without a cookie on a per-route basis.
1657 When unset, the kernel's default will be used.
</para>
1662 <term><varname>TTLPropagate=
</varname></term>
1664 <para>Takes a boolean. When true enables TTL propagation at Label Switched Path (LSP) egress.
1665 When unset, the kernel's default will be used.
</para>
1670 <term><varname>MTUBytes=
</varname></term>
1672 <para>The maximum transmission unit in bytes to set for the route. The usual suffixes K, M,
1673 G, are supported and are understood to the base of
1024.
</para>
1678 <term><varname>TCPAdvertisedMaximumSegmentSize=
</varname></term>
1680 <para>Specifies the Path MSS (in bytes) hints given on TCP layer. The usual suffixes K, M, G,
1681 are supported and are understood to the base of
1024. An unsigned integer in the range
1682 1…
4294967294. When unset, the kernel's default will be used.
</para>
1687 <term><varname>TCPCongestionControlAlgorithm=
</varname></term>
1689 <para>Specifies the TCP congestion control algorithm for the route. Takes a name of the algorithm,
1690 e.g.
<literal>bbr
</literal>,
<literal>dctcp
</literal>, or
<literal>vegas
</literal>. When unset,
1691 the kernel's default will be used.
</para>
1696 <term><varname>MultiPathRoute=
<replaceable>address
</replaceable>[@
<replaceable>name
</replaceable>] [
<replaceable>weight
</replaceable>]
</varname></term>
1698 <para>Configures multipath route. Multipath routing is the technique of using multiple
1699 alternative paths through a network. Takes gateway address. Optionally, takes a network
1700 interface name or index separated with
<literal>@
</literal>, and a weight in
1.
.256 for this
1701 multipath route separated with whitespace. This setting can be specified multiple times. If
1702 an empty string is assigned, then the all previous assignments are cleared.
</para>
1707 <term><varname>NextHop=
</varname></term>
1709 <para>Specifies the nexthop id. Takes an unsigned integer in the range
1…
4294967295. If set,
1710 the corresponding [NextHop] section must be configured. Defaults to unset.
</para>
1717 <title>[DHCPv4] Section Options
</title>
1719 <para>The [DHCPv4] section configures the DHCPv4 client, if it is enabled with the
1720 <varname>DHCP=
</varname> setting described above:
</para>
1722 <variablelist class='network-directives'
>
1724 <!-- DHCP packet contents -->
1727 <term><varname>SendHostname=
</varname></term>
1729 <para>When true (the default), the machine's hostname (or the value specified with
1730 <varname>Hostname=
</varname>, described below) will be sent to the DHCP server. Note that the
1731 hostname must consist only of
7-bit ASCII lower-case characters and no spaces or dots, and be
1732 formatted as a valid DNS domain name. Otherwise, the hostname is not sent even if this option
1738 <term><varname>Hostname=
</varname></term>
1740 <para>Use this value for the hostname which is sent to the DHCP server, instead of machine's
1741 hostname. Note that the specified hostname must consist only of
7-bit ASCII lower-case
1742 characters and no spaces or dots, and be formatted as a valid DNS domain name.
</para>
1747 <term><varname>MUDURL=
</varname></term>
1749 <para>When configured, the specified Manufacturer Usage Description (MUD) URL will be sent
1750 to the DHCPv4 server. Takes a URL of length up to
255 characters. A superficial verification
1751 that the string is a valid URL will be performed. DHCPv4 clients are intended to have at most
1752 one MUD URL associated with them. See
1753 <ulink url=
"https://tools.ietf.org/html/rfc8520">RFC
8520</ulink>.
</para>
1755 <para>MUD is an embedded software standard defined by the IETF that allows IoT device makers
1756 to advertise device specifications, including the intended communication patterns for their
1757 device when it connects to the network. The network can then use this to author a
1758 context-specific access policy, so the device functions only within those parameters.
</para>
1763 <term><varname>ClientIdentifier=
</varname></term>
1765 <para>The DHCPv4 client identifier to use. Takes one of
<option>mac
</option>,
1766 <option>duid
</option> or
<option>duid-only
</option>. If set to
<option>mac
</option>, the
1767 MAC address of the link is used. If set to
<option>duid
</option>, an RFC4361-compliant Client
1768 ID, which is the combination of IAID and DUID (see below), is used. If set to
1769 <option>duid-only
</option>, only DUID is used, this may not be RFC compliant, but some setups
1770 may require to use this. Defaults to
<option>duid
</option>.
</para>
1775 <term><varname>VendorClassIdentifier=
</varname></term>
1777 <para>The vendor class identifier used to identify vendor type and configuration.
</para>
1782 <term><varname>UserClass=
</varname></term>
1784 <para>A DHCPv4 client can use UserClass option to identify the type or category of user or
1785 applications it represents. The information contained in this option is a string that
1786 represents the user class of which the client is a member. Each class sets an identifying
1787 string of information to be used by the DHCP service to classify clients. Takes a
1788 whitespace-separated list of strings.
</para>
1793 <term><varname>DUIDType=
</varname></term>
1795 <para>Override the global
<varname>DUIDType=
</varname> setting for this network. See
1796 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
1797 for a description of possible values.
</para>
1802 <term><varname>DUIDRawData=
</varname></term>
1804 <para>Override the global
<varname>DUIDRawData=
</varname> setting for this network. See
1805 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>
1806 for a description of possible values.
</para>
1811 <term><varname>IAID=
</varname></term>
1813 <para>The DHCP Identity Association Identifier (IAID) for the interface, a
32-bit unsigned
1819 <term><varname>Anonymize=
</varname></term>
1821 <para>Takes a boolean. When true, the options sent to the DHCP server will follow the
1822 <ulink url=
"https://tools.ietf.org/html/rfc7844">RFC
7844</ulink> (Anonymity Profiles for
1823 DHCP Clients) to minimize disclosure of identifying information. Defaults to false.
</para>
1825 <para>This option should only be set to true when
<varname>MACAddressPolicy=
</varname> is set
1826 to
<option>random
</option> (see
1827 <citerefentry><refentrytitle>systemd.link
</refentrytitle><manvolnum>5</manvolnum></citerefentry>).
1830 <para>When true,
<varname>SendHostname=
</varname>,
<varname>ClientIdentifier=
</varname>,
1831 <varname>VendorClassIdentifier=
</varname>,
<varname>UserClass=
</varname>,
1832 <varname>RequestOptions=
</varname>,
<varname>SendOption=
</varname>,
1833 <varname>SendVendorOption=
</varname>, and
<varname>MUDURL=
</varname> are ignored.
</para>
1835 <para>With this option enabled DHCP requests will mimic those generated by Microsoft
1836 Windows, in order to reduce the ability to fingerprint and recognize installations. This
1837 means DHCP request sizes will grow and lease data will be more comprehensive than normally,
1838 though most of the requested data is not actually used.
</para>
1843 <term><varname>RequestOptions=
</varname></term>
1845 <para>Sets request options to be sent to the server in the DHCPv4 request options list. A
1846 whitespace-separated list of integers in the range
1…
254. Defaults to unset.
</para>
1851 <term><varname>SendOption=
</varname></term>
1853 <para>Send an arbitrary raw option in the DHCPv4 request. Takes a DHCP option number, data
1854 type and data separated with a colon
1855 (
<literal><replaceable>option
</replaceable>:
<replaceable>type
</replaceable>:
<replaceable>value
</replaceable></literal>).
1856 The option number must be an integer in the range
1…
254. The type takes one of
1857 <literal>uint8
</literal>,
<literal>uint16
</literal>,
<literal>uint32
</literal>,
1858 <literal>ipv4address
</literal>, or
<literal>string
</literal>. Special characters in the data
1859 string may be escaped using
1860 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
1861 escapes
</ulink>. This setting can be specified multiple times. If an empty string is
1862 specified, then all options specified earlier are cleared. Defaults to unset.
</para>
1867 <term><varname>SendVendorOption=
</varname></term>
1869 <para>Send an arbitrary vendor option in the DHCPv4 request. Takes a DHCP option number, data
1870 type and data separated with a colon
1871 (
<literal><replaceable>option
</replaceable>:
<replaceable>type
</replaceable>:
<replaceable>value
</replaceable></literal>).
1872 The option number must be an integer in the range
1…
254. The type takes one of
1873 <literal>uint8
</literal>,
<literal>uint16
</literal>,
<literal>uint32
</literal>,
1874 <literal>ipv4address
</literal>, or
<literal>string
</literal>. Special characters in the data
1875 string may be escaped using
1876 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
1877 escapes
</ulink>. This setting can be specified multiple times. If an empty string is specified,
1878 then all options specified earlier are cleared. Defaults to unset.
</para>
1883 <term><varname>IPServiceType=
</varname></term>
1885 <para>Takes one of the special values
<literal>none
</literal>,
<literal>CS6
</literal>, or
1886 <literal>CS4
</literal>. When
<literal>none
</literal> no IP service type is set to the packet
1887 sent from the DHCPv4 client. When
<literal>CS6
</literal> (network control) or
1888 <literal>CS4
</literal> (realtime), the corresponding service type will be set. Defaults to
1889 <literal>CS6
</literal>.
</para>
1893 <!-- How to use the DHCP lease -->
1896 <term><varname>Label=
</varname></term>
1898 <para>Specifies the label for the IPv4 address received from the DHCP server. The label must
1899 be a
7-bit ASCII string with a length of
1…
15 characters. Defaults to unset.
</para>
1904 <term><varname>UseDNS=
</varname></term>
1906 <para>When true (the default), the DNS servers received from the DHCP server will be used.
1909 <para>This corresponds to the
<option>nameserver
</option> option in
1910 <citerefentry project='man-pages'
><refentrytitle>resolv.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
1916 <term><varname>RoutesToDNS=
</varname></term>
1918 <para>When true, the routes to the DNS servers received from the DHCP server will be
1919 configured. When
<varname>UseDNS=
</varname> is disabled, this setting is ignored. Defaults to
1925 <term><varname>UseNTP=
</varname></term>
1927 <para>When true (the default), the NTP servers received from the DHCP server will be used by
1928 <filename>systemd-timesyncd.service
</filename>.
</para>
1933 <term><varname>RoutesToNTP=
</varname></term>
1935 <para>When true, the routes to the NTP servers received from the DHCP server will be
1936 configured. When
<varname>UseNTP=
</varname> is disabled, this setting is ignored. Defaults to
1942 <term><varname>UseSIP=
</varname></term>
1944 <para>When true (the default), the SIP servers received from the DHCP server will be collected
1945 and made available to client programs.
</para>
1950 <term><varname>UseMTU=
</varname></term>
1952 <para>When true, the interface maximum transmission unit from the DHCP server will be used on
1953 the current link. If
<varname>MTUBytes=
</varname> is set, then this setting is ignored.
1954 Defaults to false.
</para>
1956 <para>Note, some drivers will reset the interfaces if the MTU is changed. For such
1957 interfaces, please try to use
<varname>IgnoreCarrierLoss=
</varname> with a short timespan,
1958 e.g.
<literal>3 seconds
</literal>.
</para>
1963 <term><varname>UseHostname=
</varname></term>
1965 <para>When true (the default), the hostname received from the DHCP server will be set as the
1966 transient hostname of the system.
</para>
1971 <term><varname>UseDomains=
</varname></term>
1973 <para>Takes a boolean, or the special value
<option>route
</option>. When true, the domain name
1974 received from the DHCP server will be used as DNS search domain over this link, similarly to the
1975 effect of the
<option>Domains=
</option> setting. If set to
<option>route
</option>, the domain name
1976 received from the DHCP server will be used for routing DNS queries only, but not for searching,
1977 similarly to the effect of the
<option>Domains=
</option> setting when the argument is prefixed with
1978 <literal>~
</literal>. Defaults to false.
</para>
1980 <para>It is recommended to enable this option only on trusted networks, as setting this
1981 affects resolution of all hostnames, in particular of single-label names. It is generally
1982 safer to use the supplied domain only as routing domain, rather than as search domain, in
1983 order to not have it affect local resolution of single-label names.
</para>
1985 <para>When set to true, this setting corresponds to the
<option>domain
</option> option in
1986 <citerefentry project='man-pages'
><refentrytitle>resolv.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
1992 <term><varname>UseRoutes=
</varname></term>
1994 <para>When true (the default), the static routes will be requested from the DHCP server and
1995 added to the routing table with a metric of
1024, and a scope of
<option>global
</option>,
1996 <option>link
</option> or
<option>host
</option>, depending on the route's destination and
1997 gateway. If the destination is on the local host, e.g.,
127.x.x.x, or the same as the link's
1998 own address, the scope will be set to
<option>host
</option>. Otherwise if the gateway is null
1999 (a direct route), a
<option>link
</option> scope will be used. For anything else, scope
2000 defaults to
<option>global
</option>.
</para>
2005 <term><varname>RouteMetric=
</varname></term>
2007 <para>Set the routing metric for routes specified by the DHCP server (including the prefix
2008 route added for the specified prefix). Takes an unsigned integer in the range
0…
4294967295.
2009 Defaults to
1024.
</para>
2014 <term><varname>RouteTable=
<replaceable>num
</replaceable></varname></term>
2016 <para>The table identifier for DHCP routes. Takes one of predefined names
2017 <literal>default
</literal>,
<literal>main
</literal>, and
<literal>local
</literal>, and names
2018 defined in
<varname>RouteTable=
</varname> in
2019 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
2020 or a number between
1…
4294967295.
</para>
2022 <para>When used in combination with
<varname>VRF=
</varname>, the VRF's routing table is
2023 used when this parameter is not specified.
</para>
2028 <term><varname>RouteMTUBytes=
</varname></term>
2030 <para>Specifies the MTU for the DHCP routes. Please see the [Route] section for further
2036 <term><varname>UseGateway=
</varname></term>
2038 <para>When true, the gateway will be requested from the DHCP server and added to the routing
2039 table with a metric of
1024, and a scope of
<option>link
</option>. When unset, the value
2040 specified with
<varname>UseRoutes=
</varname> is used.
</para>
2045 <term><varname>UseTimezone=
</varname></term>
2046 <listitem><para>When true, the timezone received from the DHCP server will be set as timezone
2047 of the local system. Defaults to false.
</para></listitem>
2051 <term><varname>Use6RD=
</varname></term>
2053 <para>When true, subnets of the received IPv6 prefix are assigned to downstream interfaces
2054 which enables
<varname>DHCPPrefixDelegation=
</varname>. See also
2055 <varname>DHCPPrefixDelegation=
</varname> in the [Network] section, the [DHCPPrefixDelegation]
2056 section, and
<ulink url=
"https://tools.ietf.org/html/rfc5969">RFC
5969</ulink>. Defaults to
2062 <term><varname>FallbackLeaseLifetimeSec=
</varname></term>
2064 <para>Allows one to set DHCPv4 lease lifetime when DHCPv4 server does not send the lease
2065 lifetime. Takes one of
<literal>forever
</literal> or
<literal>infinity
</literal>. If
2066 specified, the acquired address never expires. Defaults to unset.
</para>
2070 <!-- How to communicate with the server -->
2073 <term><varname>RequestBroadcast=
</varname></term>
2075 <para>Request the server to use broadcast messages before the IP address has been configured.
2076 This is necessary for devices that cannot receive RAW packets, or that cannot receive packets
2077 at all before an IP address has been configured. On the other hand, this must not be enabled
2078 on networks where broadcasts are filtered out.
</para>
2083 <term><varname>MaxAttempts=
</varname></term>
2085 <para>Specifies how many times the DHCPv4 client configuration should be attempted. Takes a
2086 number or
<literal>infinity
</literal>. Defaults to
<literal>infinity
</literal>. Note that the
2087 time between retries is increased exponentially, up to approximately one per minute, so the
2088 network will not be overloaded even if this number is high. The default is suitable in most
2089 circumstances.
</para>
2094 <term><varname>ListenPort=
</varname></term>
2096 <para>Set the port from which the DHCP client packets originate.
</para>
2101 <term><varname>DenyList=
</varname></term>
2103 <para>A whitespace-separated list of IPv4 addresses. Each address can optionally take a
2104 prefix length after
<literal>/
</literal>. DHCP offers from servers in the list are rejected.
2105 Note that if
<varname>AllowList=
</varname> is configured then
<varname>DenyList=
</varname> is
2111 <term><varname>AllowList=
</varname></term>
2113 <para>A whitespace-separated list of IPv4 addresses. Each address can optionally take a
2114 prefix length after
<literal>/
</literal>. DHCP offers from servers in the list are accepted.
2120 <term><varname>SendRelease=
</varname></term>
2122 <para>When true, the DHCPv4 client sends a DHCP release packet when it stops. Defaults to
2128 <term><varname>SendDecline=
</varname></term>
2130 <para>A boolean. When true,
<command>systemd-networkd
</command> performs IPv4 Duplicate
2131 Address Detection to the acquired address by the DHCPv4 client. If duplicate is detected,
2132 the DHCPv4 client rejects the address by sending a
<constant>DHCPDECLINE
</constant> packet to
2133 the DHCP server, and tries to obtain an IP address again. See
2134 <ulink url=
"https://tools.ietf.org/html/rfc5227">RFC
5227</ulink>. Defaults to false.
</para>
2139 <term><varname>NetLabel=
</varname></term>
2141 <para>This applies the NetLabel for the addresses received with DHCP, like
2142 <varname>NetLabel=
</varname> in [Address] section applies it to statically configured
2143 addresses. See
<varname>NetLabel=
</varname> in [Address] section for more details.
</para>
2150 <title>[DHCPv6] Section Options
</title>
2152 <para>The [DHCPv6] section configures the DHCPv6 client, if it is enabled with the
2153 <varname>DHCP=
</varname> setting described above, or invoked by the IPv6 Router Advertisement:
2156 <variablelist class='network-directives'
>
2158 <!-- DHCP packet contents -->
2161 <term><varname>MUDURL=
</varname></term>
2162 <term><varname>IAID=
</varname></term>
2163 <term><varname>DUIDType=
</varname></term>
2164 <term><varname>DUIDRawData=
</varname></term>
2165 <term><varname>RequestOptions=
</varname></term>
2167 <para>As in the [DHCPv4] section.
</para>
2172 <term><varname>SendOption=
</varname></term>
2174 <para>As in the [DHCPv4] section, however because DHCPv6 uses
16-bit fields to store option
2175 numbers, the option number is an integer in the range
1…
65536.
</para>
2180 <term><varname>SendVendorOption=
</varname></term>
2182 <para>Send an arbitrary vendor option in the DHCPv6 request. Takes an enterprise identifier,
2183 DHCP option number, data type, and data separated with a colon
2184 (
<literal><replaceable>enterprise identifier
</replaceable>:
<replaceable>option
</replaceable>:
<replaceable>type
</replaceable>:
<replaceable>value
</replaceable></literal>).
2185 Enterprise identifier is an unsigned integer in the range
1…
4294967294. The option number
2186 must be an integer in the range
1…
254. Data type takes one of
<literal>uint8
</literal>,
2187 <literal>uint16
</literal>,
<literal>uint32
</literal>,
<literal>ipv4address
</literal>,
2188 <literal>ipv6address
</literal>, or
<literal>string
</literal>. Special characters in the data
2189 string may be escaped using
2190 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
2191 escapes
</ulink>. This setting can be specified multiple times. If an empty string is
2192 specified, then all options specified earlier are cleared. Defaults to unset.
</para>
2197 <term><varname>UserClass=
</varname></term>
2199 <para>A DHCPv6 client can use User Class option to identify the type or category of user or
2200 applications it represents. The information contained in this option is a string that
2201 represents the user class of which the client is a member. Each class sets an identifying
2202 string of information to be used by the DHCP service to classify clients. Special characters
2203 in the data string may be escaped using
2204 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
2205 escapes
</ulink>. This setting can be specified multiple times. If an empty string is
2206 specified, then all options specified earlier are cleared. Takes a whitespace-separated list
2207 of strings. Note that currently
<constant>NUL
</constant> bytes are not allowed.
</para>
2212 <term><varname>VendorClass=
</varname></term>
2214 <para>A DHCPv6 client can use VendorClass option to identify the vendor that manufactured the
2215 hardware on which the client is running. The information contained in the data area of this
2216 option is contained in one or more opaque fields that identify details of the hardware
2217 configuration. Takes a whitespace-separated list of strings.
</para>
2222 <term><varname>PrefixDelegationHint=
</varname></term>
2224 <para>Takes an IPv6 address with prefix length in the same format as the
2225 <varname>Address=
</varname> in the [Network] section. The DHCPv6 client will include a prefix
2226 hint in the DHCPv6 solicitation sent to the server. The prefix length must be in the range
2227 1…
128. Defaults to unset.
</para>
2232 <term><varname>RapidCommit=
</varname></term>
2234 <para>Takes a boolean. The DHCPv6 client can obtain configuration parameters from a DHCPv6 server
2235 through a rapid two-message exchange (solicit and reply). When the rapid commit option is set by
2236 both the DHCPv6 client and the DHCPv6 server, the two-message exchange is used. Otherwise, the
2237 four-message exchange (solicit, advertise, request, and reply) is used. The two-message exchange
2238 provides faster client configuration. See
2239 <ulink url=
"https://tools.ietf.org/html/rfc3315#section-17.2.1">RFC
3315</ulink> for details.
2240 Defaults to true, and the two-message exchange will be used if the server support it.
</para>
2244 <!-- How to use the DHCP lease -->
2247 <term><varname>UseAddress=
</varname></term>
2249 <para>When true (the default), the IP addresses provided by the DHCPv6 server will be
2255 <term><varname>UseDelegatedPrefix=
</varname></term>
2257 <para>When true (the default), the client will request the DHCPv6 server to delegate
2258 prefixes. If the server provides prefixes to be delegated, then subnets of the prefixes are
2259 assigned to the interfaces that have
<varname>DHCPPrefixDelegation=yes
</varname>.
2260 See also the
<varname>DHCPPrefixDelegation=
</varname> setting in the [Network] section,
2261 settings in the [DHCPPrefixDelegation] section, and
2262 <ulink url=
"https://www.rfc-editor.org/rfc/rfc8415.html#section-6.3">RFC
8415</ulink>.
2268 <term><varname>UseDNS=
</varname></term>
2269 <term><varname>UseNTP=
</varname></term>
2270 <term><varname>UseHostname=
</varname></term>
2271 <term><varname>UseDomains=
</varname></term>
2272 <term><varname>NetLabel=
</varname></term>
2274 <para>As in the [DHCPv4] section.
</para>
2278 <!-- How to communicate with the server -->
2281 <term><varname>WithoutRA=
</varname></term>
2283 <para>Allows DHCPv6 client to start without router advertisements's
2284 <literal>managed
</literal> or
<literal>other configuration
</literal> flag. Takes one of
2285 <literal>no
</literal>,
<literal>solicit
</literal>, or
2286 <literal>information-request
</literal>. If this is not specified,
2287 <literal>solicit
</literal> is used when
<varname>DHCPPrefixDelegation=
</varname> is enabled
2288 and
<varname>UplinkInterface=:self
</varname> is specified in the [DHCPPrefixDelegation]
2289 section. Otherwise, defaults to
<literal>no
</literal>, and the DHCPv6 client will be started
2290 when an RA is received. See also the
<varname>DHCPv6Client=
</varname> setting in the
2291 [IPv6AcceptRA] section.
</para>
2298 <title>[DHCPPrefixDelegation] Section Options
</title>
2299 <para>The [DHCPPrefixDelegation] section configures subnet prefixes of the delegated prefixes
2300 acquired by a DHCPv6 client, or by a DHCPv4 client through the
6RD option on another interface.
2301 The settings in this section are used only when the
<varname>DHCPPrefixDelegation=
</varname>
2302 setting in the [Network] section is enabled.
</para>
2304 <variablelist class='network-directives'
>
2306 <term><varname>UplinkInterface=
</varname></term>
2308 <para>Specifies the name or the index of the uplink interface, or one of the special values
2309 <literal>:self
</literal> and
<literal>:auto
</literal>. When
<literal>:self
</literal>, the
2310 interface itself is considered the uplink interface, and
2311 <varname>WithoutRA=solicit
</varname> is implied if the setting is not explicitly specified.
2312 When
<literal>:auto
</literal>, the first link which acquired prefixes to be delegated from
2313 the DHCPv6 or DHCPv4 server is selected. Defaults to
<literal>:auto
</literal>.
</para>
2318 <term><varname>SubnetId=
</varname></term>
2320 <para>Configure a specific subnet ID on the interface from a (previously) received prefix
2321 delegation. You can either set
"auto" (the default) or a specific subnet ID (as defined in
2322 <ulink url=
"https://tools.ietf.org/html/rfc4291#section-2.5.4">RFC
4291</ulink>, section
2323 2.5.4), in which case the allowed value is hexadecimal, from
0 to
0x7fffffffffffffff
2329 <term><varname>Announce=
</varname></term>
2331 <para>Takes a boolean. When enabled, and
<varname>IPv6SendRA=
</varname> in [Network] section
2332 is enabled, the delegated prefixes are distributed through the IPv6 Router Advertisement.
2333 This setting will be ignored when the
<varname>DHCPPrefixDelegation=
</varname> setting is
2334 enabled on the upstream interface. Defaults to yes.
</para>
2339 <term><varname>Assign=
</varname></term>
2341 <para>Takes a boolean. Specifies whether to add an address from the delegated prefixes which
2342 are received from the WAN interface by the DHCPv6 Prefix Delegation. When true (on LAN
2343 interfce), the EUI-
64 algorithm will be used by default to form an interface identifier from
2344 the delegated prefixes. See also
<varname>Token=
</varname> setting below. Defaults to yes.
2350 <term><varname>Token=
</varname></term>
2352 <para>Specifies an optional address generation mode for assigning an address in each
2353 delegated prefix. This accepts the same syntax as
<varname>Token=
</varname> in the
2354 [IPv6AcceptRA] section. If
<varname>Assign=
</varname> is set to false, then this setting will
2355 be ignored. Defaults to unset, which means the EUI-
64 algorithm will be used.
</para>
2360 <term><varname>ManageTemporaryAddress=
</varname></term>
2362 <para>As in the [Address] section, but defaults to true.
</para>
2367 <term><varname>RouteMetric=
</varname></term>
2369 <para>The metric of the route to the delegated prefix subnet. Takes an unsigned integer in
2370 the range
0…
4294967295. When set to
0, the kernel's default value is used. Defaults to
256.
2376 <term><varname>NetLabel=
</varname></term>
2378 <para>This applies the NetLabel for the addresses received with DHCP, like
2379 <varname>NetLabel=
</varname> in [Address] section applies it to statically configured
2380 addresses. See
<varname>NetLabel=
</varname> in [Address] section for more details.
</para>
2387 <title>[IPv6AcceptRA] Section Options
</title>
2388 <para>The [IPv6AcceptRA] section configures the IPv6 Router Advertisement (RA) client, if it is enabled
2389 with the
<varname>IPv6AcceptRA=
</varname> setting described above:
</para>
2391 <variablelist class='network-directives'
>
2393 <term><varname>Token=
</varname></term>
2395 <para>Specifies an optional address generation mode for the Stateless Address
2396 Autoconfiguration (SLAAC). The following values are supported:
</para>
2400 <term><option>eui64
</option></term>
2403 The EUI-
64 algorithm will be used to generate an address for that prefix. Only
2404 supported by Ethernet or InfiniBand interfaces.
2409 <term><option>static:
<replaceable>ADDRESS
</replaceable></option></term>
2412 An IPv6 address must be specified after a colon (
<literal>:
</literal>), and the
2413 lower bits of the supplied address are combined with the upper bits of a prefix
2414 received in a Router Advertisement (RA) message to form a complete address. Note
2415 that if multiple prefixes are received in an RA message, or in multiple RA messages,
2416 addresses will be formed from each of them using the supplied address. This mode
2417 implements SLAAC but uses a static interface identifier instead of an identifier
2418 generated by using the EUI-
64 algorithm. Because the interface identifier is static,
2419 if Duplicate Address Detection detects that the computed address is a duplicate
2420 (in use by another node on the link), then this mode will fail to provide an address
2421 for that prefix. If an IPv6 address without mode is specified, then
2422 <literal>static
</literal> mode is assumed.
2427 <term><option>prefixstable[:
<replaceable>ADDRESS
</replaceable>][,
<replaceable>UUID
</replaceable>]
</option></term>
2430 The algorithm specified in
2431 <ulink url=
"https://tools.ietf.org/html/rfc7217">RFC
7217</ulink> will be used to
2432 generate interface identifiers. This mode can optionally take an IPv6 address
2433 separated with a colon (
<literal>:
</literal>). If an IPv6 address is specified,
2434 then an interface identifier is generated only when a prefix received in an RA
2435 message matches the supplied address.
2438 This mode can also optionally take a non-null UUID in the format which
2439 <function>sd_id128_from_string()
</function> accepts, e.g.
2440 <literal>86b123b969ba4b7eb8b3d8605123525a
</literal> or
2441 <literal>86b123b9-
69ba-
4b7e-b8b3-d8605123525a
</literal>. If a UUID is specified, the
2442 value is used as the secret key to generate interface identifiers. If not specified,
2443 then an application specific ID generated with the system's machine-ID will be used
2444 as the secret key. See
2445 <citerefentry><refentrytitle>sd-id128
</refentrytitle><manvolnum>3</manvolnum></citerefentry>,
2446 <citerefentry><refentrytitle>sd_id128_from_string
</refentrytitle><manvolnum>3</manvolnum></citerefentry>,
2448 <citerefentry><refentrytitle>sd_id128_get_machine
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
2451 Note that the
<literal>prefixstable
</literal> algorithm uses both the interface
2452 name and MAC address as input to the hash to compute the interface identifier, so
2453 if either of those are changed the resulting interface identifier (and address)
2454 will be changed, even if the prefix received in the RA message has not been
2461 <para>If no address generation mode is specified (which is the default), or a received
2462 prefix does not match any of the addresses provided in
<literal>prefixstable
</literal>
2463 mode, then the EUI-
64 algorithm will be used for Ethernet or InfiniBand interfaces,
2464 otherwise
<literal>prefixstable
</literal> will be used to form an interface identifier for
2467 <para>This setting can be specified multiple times. If an empty string is assigned, then
2468 the all previous assignments are cleared.
</para>
2471 <programlisting>Token=eui64
2473 Token=static:::
1a:
2b:
3c:
4d
2475 Token=prefixstable:
2002:da8:
1::
</programlisting></para>
2480 <term><varname>UseDNS=
</varname></term>
2482 <para>When true (the default), the DNS servers received in the Router Advertisement will be used.
</para>
2484 <para>This corresponds to the
<option>nameserver
</option> option in
<citerefentry
2485 project='man-pages'
><refentrytitle>resolv.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para>
2490 <term><varname>UseDomains=
</varname></term>
2492 <para>Takes a boolean, or the special value
<literal>route
</literal>. When true, the domain name
2493 received via IPv6 Router Advertisement (RA) will be used as DNS search domain over this link,
2494 similarly to the effect of the
<option>Domains=
</option> setting. If set to
2495 <literal>route
</literal>, the domain name received via IPv6 RA will be used for routing DNS queries
2496 only, but not for searching, similarly to the effect of the
<option>Domains=
</option> setting when
2497 the argument is prefixed with
<literal>~
</literal>. Defaults to false.
</para>
2499 <para>It is recommended to enable this option only on trusted networks, as setting this affects resolution
2500 of all hostnames, in particular of single-label names. It is generally safer to use the supplied domain
2501 only as routing domain, rather than as search domain, in order to not have it affect local resolution of
2502 single-label names.
</para>
2504 <para>When set to true, this setting corresponds to the
<option>domain
</option> option in
<citerefentry
2505 project='man-pages'
><refentrytitle>resolv.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para>
2510 <term><varname>RouteTable=
<replaceable>num
</replaceable></varname></term>
2512 <para>The table identifier for the routes received in the Router Advertisement. Takes one of
2513 predefined names
<literal>default
</literal>,
<literal>main
</literal>, and
<literal>local
</literal>,
2514 and names defined in
<varname>RouteTable=
</varname> in
2515 <citerefentry><refentrytitle>networkd.conf
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
2516 or a number between
1…
4294967295.
</para>
2518 <para>When used in combination with
<varname>VRF=
</varname>, the VRF's routing table is
2519 used when this parameter is not specified.
</para>
2524 <term><varname>RouteMetric=
</varname></term>
2526 <para>Set the routing metric for the routes received in the Router Advertisement. Takes an
2527 unsigned integer in the range
0…
4294967295. Defaults to
1024.
</para>
2532 <term><varname>UseMTU=
</varname></term>
2534 <para>Takes a boolean. When true, the MTU received in the Router Advertisement will be
2535 used. Defaults to true.
</para>
2540 <term><varname>UseGateway=
</varname></term>
2542 <para>When true (the default), the router address will be configured as the default gateway.
2548 <term><varname>UseRoutePrefix=
</varname></term>
2550 <para>When true (the default), the routes corresponding to the route prefixes received in
2551 the Router Advertisement will be configured.
</para>
2556 <term><varname>UseAutonomousPrefix=
</varname></term>
2558 <para>When true (the default), the autonomous prefix received in the Router Advertisement will be used and take
2559 precedence over any statically configured ones.
</para>
2564 <term><varname>UseOnLinkPrefix=
</varname></term>
2566 <para>When true (the default), the onlink prefix received in the Router Advertisement will be
2567 used and takes precedence over any statically configured ones.
</para>
2572 <term><varname>RouterDenyList=
</varname></term>
2574 <para>A whitespace-separated list of IPv6 router addresses. Each address can optionally
2575 take a prefix length after
<literal>/
</literal>. Any information advertised by the listed
2576 router is ignored.
</para>
2581 <term><varname>RouterAllowList=
</varname></term>
2583 <para>A whitespace-separated list of IPv6 router addresses. Each address can optionally
2584 take a prefix length after
<literal>/
</literal>. Only information advertised by the listed
2585 router is accepted. Note that if
<varname>RouterAllowList=
</varname> is configured then
2586 <varname>RouterDenyList=
</varname> is ignored.
</para>
2591 <term><varname>PrefixDenyList=
</varname></term>
2593 <para>A whitespace-separated list of IPv6 prefixes. Each prefix can optionally take its
2594 prefix length after
<literal>/
</literal>. IPv6 prefixes supplied via router advertisements
2595 in the list are ignored.
</para>
2600 <term><varname>PrefixAllowList=
</varname></term>
2602 <para>A whitespace-separated list of IPv6 prefixes. Each prefix can optionally take its
2603 prefix length after
<literal>/
</literal>. IPv6 prefixes supplied via router advertisements
2604 in the list are allowed. Note that if
<varname>PrefixAllowList=
</varname> is configured
2605 then
<varname>PrefixDenyList=
</varname> is ignored.
</para>
2610 <term><varname>RouteDenyList=
</varname></term>
2612 <para>A whitespace-separated list of IPv6 route prefixes. Each prefix can optionally take
2613 its prefix length after
<literal>/
</literal>. IPv6 route prefixes supplied via router
2614 advertisements in the list are ignored.
</para>
2619 <term><varname>RouteAllowList=
</varname></term>
2621 <para>A whitespace-separated list of IPv6 route prefixes. Each prefix can optionally take
2622 its prefix length after
<literal>/
</literal>. IPv6 route prefixes supplied via router
2623 advertisements in the list are allowed. Note that if
<varname>RouteAllowList=
</varname> is
2624 configured then
<varname>RouteDenyList=
</varname> is ignored.
</para>
2629 <term><varname>DHCPv6Client=
</varname></term>
2631 <para>Takes a boolean, or the special value
<literal>always
</literal>. When true, the
2632 DHCPv6 client will be started in
<literal>solicit
</literal> mode if the RA has the
2633 <literal>managed
</literal> flag or
<literal>information-request
</literal> mode if the RA
2634 lacks the
<literal>managed
</literal> flag but has the
2635 <literal>other configuration
</literal> flag. If set to
<literal>always
</literal>, the
2636 DHCPv6 client will be started in
<literal>solicit
</literal> mode when an RA is received,
2637 even if neither the
<literal>managed
</literal> nor the
2638 <literal>other configuration
</literal> flag is set in the RA. This will be ignored when
2639 <varname>WithoutRA=
</varname> in the [DHCPv6] section is enabled, or
2640 <varname>UplinkInterface=:self
</varname> in the [DHCPPrefixDelegation] section is
2641 specified. Defaults to true.
</para>
2646 <term><varname>NetLabel=
</varname></term>
2648 <para>This applies the NetLabel for the addresses received with RA, like
2649 <varname>NetLabel=
</varname> in [Address] section applies it to statically configured
2650 addresses. See
<varname>NetLabel=
</varname> in [Address] section for more details.
</para>
2657 <title>[DHCPServer] Section Options
</title>
2658 <para>The [DHCPServer] section contains settings for the DHCP server, if enabled via the
2659 <varname>DHCPServer=
</varname> option described above:
</para>
2661 <variablelist class='network-directives'
>
2664 <term><varname>ServerAddress=
</varname></term>
2665 <listitem><para>Specifies server address for the DHCP server. Takes an IPv4 address with prefix
2666 length, for example
192.168.0.1/
24. This setting may be useful when the link on
2667 which the DHCP server is running has multiple static addresses. When unset, one of static addresses
2668 in the link will be automatically selected. Defaults to unset.
</para></listitem>
2672 <term><varname>PoolOffset=
</varname></term>
2673 <term><varname>PoolSize=
</varname></term>
2675 <listitem><para>Configures the pool of addresses to hand out. The pool
2676 is a contiguous sequence of IP addresses in the subnet configured for
2677 the server address, which does not include the subnet nor the broadcast
2678 address.
<varname>PoolOffset=
</varname> takes the offset of the pool
2679 from the start of subnet, or zero to use the default value.
2680 <varname>PoolSize=
</varname> takes the number of IP addresses in the
2681 pool or zero to use the default value. By default, the pool starts at
2682 the first address after the subnet address and takes up the rest of
2683 the subnet, excluding the broadcast address. If the pool includes
2684 the server address (the default), this is reserved and not handed
2685 out to clients.
</para></listitem>
2689 <term><varname>DefaultLeaseTimeSec=
</varname></term>
2690 <term><varname>MaxLeaseTimeSec=
</varname></term>
2692 <listitem><para>Control the default and maximum DHCP lease
2693 time to pass to clients. These settings take time values in seconds or
2694 another common time unit, depending on the suffix. The default
2695 lease time is used for clients that did not ask for a specific
2696 lease time. If a client asks for a lease time longer than the
2697 maximum lease time, it is automatically shortened to the
2698 specified time. The default lease time defaults to
1h, the
2699 maximum lease time to
12h. Shorter lease times are beneficial
2700 if the configuration data in DHCP leases changes frequently
2701 and clients shall learn the new settings with shorter
2702 latencies. Longer lease times reduce the generated DHCP
2703 network traffic.
</para></listitem>
2707 <term><varname>UplinkInterface=
</varname></term>
2708 <listitem><para>Specifies the name or the index of the uplink interface, or one of the special
2709 values
<literal>:none
</literal> and
<literal>:auto
</literal>. When emitting DNS, NTP, or SIP
2710 servers is enabled but no servers are specified, the servers configured in the uplink interface
2711 will be emitted. When
<literal>:auto
</literal>, the link which has a default gateway with the
2712 highest priority will be automatically selected. When
<literal>:none
</literal>, no uplink
2713 interface will be selected. Defaults to
<literal>:auto
</literal>.
</para></listitem>
2717 <term><varname>EmitDNS=
</varname></term>
2718 <term><varname>DNS=
</varname></term>
2720 <listitem><para><varname>EmitDNS=
</varname> takes a boolean. Configures whether the DHCP leases
2721 handed out to clients shall contain DNS server information. Defaults to
<literal>yes
</literal>.
2722 The DNS servers to pass to clients may be configured with the
<varname>DNS=
</varname> option,
2723 which takes a list of IPv4 addresses, or special value
<literal>_server_address
</literal> which
2724 will be converted to the address used by the DHCP server.
</para>
2726 <para>If the
<varname>EmitDNS=
</varname> option is enabled but no servers configured, the
2727 servers are automatically propagated from an
"uplink" interface that has appropriate servers
2728 set. The
"uplink" interface is determined by the default route of the system with the highest
2729 priority. Note that this information is acquired at the time the lease is handed out, and does
2730 not take uplink interfaces into account that acquire DNS server information at a later point.
2731 If no suitable uplink interface is found the DNS server data from
2732 <filename>/etc/resolv.conf
</filename> is used. Also, note that the leases are not refreshed if
2733 the uplink network configuration changes. To ensure clients regularly acquire the most current
2734 uplink DNS server information, it is thus advisable to shorten the DHCP lease time via
2735 <varname>MaxLeaseTimeSec=
</varname> described above.
</para>
2737 <para>This setting can be specified multiple times. If an empty string is specified, then all
2738 DNS servers specified earlier are cleared.
</para></listitem>
2742 <term><varname>EmitNTP=
</varname></term>
2743 <term><varname>NTP=
</varname></term>
2744 <term><varname>EmitSIP=
</varname></term>
2745 <term><varname>SIP=
</varname></term>
2746 <term><varname>EmitPOP3=
</varname></term>
2747 <term><varname>POP3=
</varname></term>
2748 <term><varname>EmitSMTP=
</varname></term>
2749 <term><varname>SMTP=
</varname></term>
2750 <term><varname>EmitLPR=
</varname></term>
2751 <term><varname>LPR=
</varname></term>
2753 <listitem><para>Similar to the
<varname>EmitDNS=
</varname> and
<varname>DNS=
</varname> settings
2754 described above, these settings configure whether and what server information for the indicate
2755 protocol shall be emitted as part of the DHCP lease. The same syntax, propagation semantics and
2756 defaults apply as for
<varname>EmitDNS=
</varname> and
<varname>DNS=
</varname>.
</para></listitem>
2760 <term><varname>EmitRouter=
</varname></term>
2761 <term><varname>Router=
</varname></term>
2763 <listitem><para>The
<varname>EmitRouter=
</varname> setting takes a boolean value, and configures
2764 whether the DHCP lease should contain the router option. The
<varname>Router=
</varname> setting
2765 takes an IPv4 address, and configures the router address to be emitted. When the
2766 <varname>Router=
</varname> setting is not specified, then the server address will be used for
2767 the router option. When the
<varname>EmitRouter=
</varname> setting is disabled, the
2768 <varname>Router=
</varname> setting will be ignored. The
<varname>EmitRouter=
</varname> setting
2769 defaults to true, and the
<varname>Router=
</varname> setting defaults to unset.
2774 <term><varname>EmitTimezone=
</varname></term>
2775 <term><varname>Timezone=
</varname></term>
2777 <listitem><para>Takes a boolean. Configures whether the DHCP leases handed out
2778 to clients shall contain timezone information. Defaults to
<literal>yes
</literal>. The
2779 <varname>Timezone=
</varname> setting takes a timezone string
2780 (such as
<literal>Europe/Berlin
</literal> or
2781 <literal>UTC
</literal>) to pass to clients. If no explicit
2782 timezone is set, the system timezone of the local host is
2783 propagated, as determined by the
2784 <filename>/etc/localtime
</filename> symlink.
</para></listitem>
2788 <term><varname>BootServerAddress=
</varname></term>
2791 <para>Takes an IPv4 address of the boot server used by e.g. PXE boot systems. When specified, this
2792 address is sent in the
<option>siaddr
</option> field of the DHCP message header. See
<ulink
2793 url=
"https://www.rfc-editor.org/rfc/rfc2131.html">RFC
2131</ulink> for more details. Defaults to
2799 <term><varname>BootServerName=
</varname></term>
2802 <para>Takes a name of the boot server used by e.g. PXE boot systems. When specified, this name is
2803 sent in the DHCP option
66 (
"TFTP server name"). See
<ulink
2804 url=
"https://www.rfc-editor.org/rfc/rfc2132.html">RFC
2132</ulink> for more details. Defaults to
2807 <para>Note that typically setting one of
<varname>BootServerName=
</varname> or
2808 <varname>BootServerAddress=
</varname> is sufficient, but both can be set too, if desired.
</para>
2813 <term><varname>BootFilename=
</varname></term>
2816 <para>Takes a path or URL to a file loaded by e.g. a PXE boot loader. When specified, this path is
2817 sent in the DHCP option
67 (
"Bootfile name"). See
<ulink
2818 url=
"https://www.rfc-editor.org/rfc/rfc2132.html">RFC
2132</ulink> for more details. Defaults to
2824 <term><varname>SendOption=
</varname></term>
2826 <para>Send a raw option with value via DHCPv4 server. Takes a DHCP option number, data type
2827 and data (
<literal><replaceable>option
</replaceable>:
<replaceable>type
</replaceable>:
<replaceable>value
</replaceable></literal>).
2828 The option number is an integer in the range
1…
254. The type takes one of
<literal>uint8
</literal>,
2829 <literal>uint16
</literal>,
<literal>uint32
</literal>,
<literal>ipv4address
</literal>,
<literal>ipv6address
</literal>, or
2830 <literal>string
</literal>. Special characters in the data string may be escaped using
2831 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
2832 escapes
</ulink>. This setting can be specified multiple times. If an empty string is specified,
2833 then all options specified earlier are cleared. Defaults to unset.
</para>
2838 <term><varname>SendVendorOption=
</varname></term>
2840 <para>Send a vendor option with value via DHCPv4 server. Takes a DHCP option number, data type
2841 and data (
<literal><replaceable>option
</replaceable>:
<replaceable>type
</replaceable>:
<replaceable>value
</replaceable></literal>).
2842 The option number is an integer in the range
1…
254. The type takes one of
<literal>uint8
</literal>,
2843 <literal>uint16
</literal>,
<literal>uint32
</literal>,
<literal>ipv4address
</literal>, or
2844 <literal>string
</literal>. Special characters in the data string may be escaped using
2845 <ulink url=
"https://en.wikipedia.org/wiki/Escape_sequences_in_C#Table_of_escape_sequences">C-style
2846 escapes
</ulink>. This setting can be specified multiple times. If an empty string is specified,
2847 then all options specified earlier are cleared. Defaults to unset.
</para>
2851 <term><varname>BindToInterface=
</varname></term>
2853 <para>Takes a boolean value. When
<literal>yes
</literal>, DHCP server socket will be bound
2854 to its network interface and all socket communication will be restricted to this interface.
2855 Defaults to
<literal>yes
</literal>, except if
<varname>RelayTarget=
</varname> is used (see below),
2856 in which case it defaults to
<literal>no
</literal>.
</para>
2860 <term><varname>RelayTarget=
</varname></term>
2862 <para>Takes an IPv4 address, which must be in the format described in
2863 <citerefentry project='man-pages'
><refentrytitle>inet_pton
</refentrytitle><manvolnum>3</manvolnum></citerefentry>.
2864 Turns this DHCP server into a DHCP relay agent. See
<ulink url=
"https://tools.ietf.org/html/rfc1542">RFC
1542</ulink>.
2865 The address is the address of DHCP server or another relay agent to forward DHCP messages to and from.
</para>
2869 <term><varname>RelayAgentCircuitId=
</varname></term>
2871 <para>Specifies value for Agent Circuit ID suboption of Relay Agent Information option.
2872 Takes a string, which must be in the format
<literal>string:
<replaceable>value
</replaceable></literal>,
2873 where
<literal><replaceable>value
</replaceable></literal> should be replaced with the value of the suboption.
2874 Defaults to unset (means no Agent Circuit ID suboption is generated).
2875 Ignored if
<varname>RelayTarget=
</varname> is not specified.
</para>
2879 <term><varname>RelayAgentRemoteId=
</varname></term>
2881 <para>Specifies value for Agent Remote ID suboption of Relay Agent Information option.
2882 Takes a string, which must be in the format
<literal>string:
<replaceable>value
</replaceable></literal>,
2883 where
<literal><replaceable>value
</replaceable></literal> should be replaced with the value of the suboption.
2884 Defaults to unset (means no Agent Remote ID suboption is generated).
2885 Ignored if
<varname>RelayTarget=
</varname> is not specified.
</para>
2893 <title>[DHCPServerStaticLease] Section Options
</title>
2894 <para>The
<literal>[DHCPServerStaticLease]
</literal> section configures a static DHCP lease to assign a
2895 fixed IPv4 address to a specific device based on its MAC address. This section can be specified multiple
2898 <variablelist class='network-directives'
>
2900 <term><varname>MACAddress=
</varname></term>
2902 <listitem><para>The hardware address of a device to match. This key is mandatory.
</para></listitem>
2906 <term><varname>Address=
</varname></term>
2908 <listitem><para>The IPv4 address that should be assigned to the device that was matched with
2909 <varname>MACAddress=
</varname>. This key is mandatory.
</para></listitem>
2915 <title>[IPv6SendRA] Section Options
</title>
2916 <para>The [IPv6SendRA] section contains settings for sending IPv6 Router Advertisements and whether
2917 to act as a router, if enabled via the
<varname>IPv6SendRA=
</varname> option described above. IPv6
2918 network prefixes or routes are defined with one or more [IPv6Prefix] or [IPv6RoutePrefix] sections.
2921 <variablelist class='network-directives'
>
2924 <term><varname>Managed=
</varname></term>
2925 <term><varname>OtherInformation=
</varname></term>
2927 <listitem><para>Takes a boolean. Controls whether a DHCPv6 server is used to acquire IPv6
2928 addresses on the network link when
<varname>Managed=
</varname>
2929 is set to
<literal>true
</literal> or if only additional network
2930 information can be obtained via DHCPv6 for the network link when
2931 <varname>OtherInformation=
</varname> is set to
2932 <literal>true
</literal>. Both settings default to
2933 <literal>false
</literal>, which means that a DHCPv6 server is not being
2934 used.
</para></listitem>
2938 <term><varname>RouterLifetimeSec=
</varname></term>
2940 <listitem><para>Takes a timespan. Configures the IPv6 router lifetime in seconds. The value must be
0
2941 seconds, or between
4 seconds and
9000 seconds. When set to
0, the host is not acting as a router.
2942 Defaults to
1800 seconds (
30 minutes).
</para>
2947 <term><varname>RouterPreference=
</varname></term>
2949 <listitem><para>Configures IPv6 router preference if
2950 <varname>RouterLifetimeSec=
</varname> is non-zero. Valid values are
2951 <literal>high
</literal>,
<literal>medium
</literal> and
2952 <literal>low
</literal>, with
<literal>normal
</literal> and
2953 <literal>default
</literal> added as synonyms for
2954 <literal>medium
</literal> just to make configuration easier. See
2955 <ulink url=
"https://tools.ietf.org/html/rfc4191">RFC
4191</ulink>
2956 for details. Defaults to
<literal>medium
</literal>.
</para></listitem>
2960 <term><varname>UplinkInterface=
</varname></term>
2961 <listitem><para>Specifies the name or the index of the uplink interface, or one of the special
2962 values
<literal>:none
</literal> and
<literal>:auto
</literal>. When emitting DNS servers or
2963 search domains is enabled but no servers are specified, the servers configured in the uplink
2964 interface will be emitted. When
<literal>:auto
</literal>, the value specified to the same
2965 setting in the [DHCPPrefixDelegation] section will be used if
2966 <varname>DHCPPrefixDelegation=
</varname> is enabled, otherwise the link which has a default
2967 gateway with the highest priority will be automatically selected. When
<literal>:none
</literal>,
2968 no uplink interface will be selected. Defaults to
<literal>:auto
</literal>.
</para></listitem>
2972 <term><varname>EmitDNS=
</varname></term>
2973 <term><varname>DNS=
</varname></term>
2975 <listitem><para><varname>DNS=
</varname> specifies a list of recursive DNS server IPv6 addresses
2976 that are distributed via Router Advertisement messages when
<varname>EmitDNS=
</varname> is true.
2977 <varname>DNS=
</varname> also takes special value
<literal>_link_local
</literal>; in that case
2978 the IPv6 link-local address is distributed. If
<varname>DNS=
</varname> is empty, DNS servers are
2979 read from the [Network] section. If the [Network] section does not contain any DNS servers
2980 either, DNS servers from the uplink interface specified in
<varname>UplinkInterface=
</varname>
2981 will be used. When
<varname>EmitDNS=
</varname> is false, no DNS server information is sent in
2982 Router Advertisement messages.
<varname>EmitDNS=
</varname> defaults to true.
</para></listitem>
2986 <term><varname>EmitDomains=
</varname></term>
2987 <term><varname>Domains=
</varname></term>
2989 <listitem><para>A list of DNS search domains distributed via Router Advertisement messages when
2990 <varname>EmitDomains=
</varname> is true. If
<varname>Domains=
</varname> is empty, DNS search
2991 domains are read from the [Network] section. If the [Network] section does not contain any DNS
2992 search domains either, DNS search domains from the uplink interface specified in
2993 <varname>UplinkInterface=
</varname> will be used. When
<varname>EmitDomains=
</varname> is false,
2994 no DNS search domain information is sent in Router Advertisement messages.
2995 <varname>EmitDomains=
</varname> defaults to true.
</para></listitem>
2999 <term><varname>DNSLifetimeSec=
</varname></term>
3001 <listitem><para>Lifetime in seconds for the DNS server addresses listed in
3002 <varname>DNS=
</varname> and search domains listed in
<varname>Domains=
</varname>. Defaults to
3003 3600 seconds (one hour).
</para></listitem>
3010 <title>[IPv6Prefix] Section Options
</title>
3011 <para>One or more [IPv6Prefix] sections contain the IPv6 prefixes that are announced via Router
3012 Advertisements. See
<ulink url=
"https://tools.ietf.org/html/rfc4861">RFC
4861</ulink> for further
3015 <variablelist class='network-directives'
>
3018 <term><varname>AddressAutoconfiguration=
</varname></term>
3019 <term><varname>OnLink=
</varname></term>
3021 <listitem><para>Takes a boolean to specify whether IPv6 addresses can be
3022 autoconfigured with this prefix and whether the prefix can be used for
3023 onlink determination. Both settings default to
<literal>true
</literal>
3024 in order to ease configuration.
3029 <term><varname>Prefix=
</varname></term>
3031 <listitem><para>The IPv6 prefix that is to be distributed to hosts. Similarly to configuring static
3032 IPv6 addresses, the setting is configured as an IPv6 prefix and its prefix length, separated by a
3033 <literal>/
</literal> character. Use multiple [IPv6Prefix] sections to configure multiple IPv6
3034 prefixes since prefix lifetimes, address autoconfiguration and onlink status may differ from one
3035 prefix to another.
</para></listitem>
3039 <term><varname>PreferredLifetimeSec=
</varname></term>
3040 <term><varname>ValidLifetimeSec=
</varname></term>
3042 <listitem><para>Preferred and valid lifetimes for the prefix measured in seconds.
3043 <varname>PreferredLifetimeSec=
</varname> defaults to
1800 seconds (
30 minutes) and
3044 <varname>ValidLifetimeSec=
</varname> defaults to
3600 seconds (one hour).
</para></listitem>
3048 <term><varname>Assign=
</varname></term>
3049 <listitem><para>Takes a boolean. When true, adds an address from the prefix. Default to false.
3054 <term><varname>Token=
</varname></term>
3056 <para>Specifies an optional address generation mode for assigning an address in each
3057 prefix. This accepts the same syntax as
<varname>Token=
</varname> in the [IPv6AcceptRA]
3058 section. If
<varname>Assign=
</varname> is set to false, then this setting will be ignored.
3059 Defaults to unset, which means the EUI-
64 algorithm will be used.
</para>
3064 <term><varname>RouteMetric=
</varname></term>
3066 <para>The metric of the prefix route. Takes an unsigned integer in the range
0…
4294967295.
3067 When unset or set to
0, the kernel's default value is used. This setting is ignored when
3068 <varname>Assign=
</varname> is false.
</para>
3075 <title>[IPv6RoutePrefix] Section Options
</title>
3076 <para>One or more [IPv6RoutePrefix] sections contain the IPv6
3077 prefix routes that are announced via Router Advertisements. See
3078 <ulink url=
"https://tools.ietf.org/html/rfc4191">RFC
4191</ulink>
3079 for further details.
</para>
3081 <variablelist class='network-directives'
>
3084 <term><varname>Route=
</varname></term>
3086 <listitem><para>The IPv6 route that is to be distributed to hosts. Similarly to configuring static
3087 IPv6 routes, the setting is configured as an IPv6 prefix routes and its prefix route length,
3088 separated by a
<literal>/
</literal> character. Use multiple [IPv6RoutePrefix] sections to configure
3089 multiple IPv6 prefix routes.
</para></listitem>
3093 <term><varname>LifetimeSec=
</varname></term>
3095 <listitem><para>Lifetime for the route prefix measured in seconds.
3096 <varname>LifetimeSec=
</varname> defaults to
3600 seconds (one hour).
</para></listitem>
3103 <title>[Bridge] Section Options
</title>
3104 <para>The [Bridge] section accepts the following keys:
</para>
3105 <variablelist class='network-directives'
>
3107 <term><varname>UnicastFlood=
</varname></term>
3109 <para>Takes a boolean. Controls whether the bridge should flood
3110 traffic for which an FDB entry is missing and the destination
3111 is unknown through this port. When unset, the kernel's default will be used.
3116 <term><varname>MulticastFlood=
</varname></term>
3118 <para>Takes a boolean. Controls whether the bridge should flood
3119 traffic for which an MDB entry is missing and the destination
3120 is unknown through this port. When unset, the kernel's default will be used.
3125 <term><varname>MulticastToUnicast=
</varname></term>
3127 <para>Takes a boolean. Multicast to unicast works on top of the multicast snooping feature of
3128 the bridge. Which means unicast copies are only delivered to hosts which are interested in it.
3129 When unset, the kernel's default will be used.
3134 <term><varname>NeighborSuppression=
</varname></term>
3136 <para>Takes a boolean. Configures whether ARP and ND neighbor suppression is enabled for
3137 this port. When unset, the kernel's default will be used.
3142 <term><varname>Learning=
</varname></term>
3144 <para>Takes a boolean. Configures whether MAC address learning is enabled for
3145 this port. When unset, the kernel's default will be used.
3150 <term><varname>HairPin=
</varname></term>
3152 <para>Takes a boolean. Configures whether traffic may be sent back out of the port on which it
3153 was received. When this flag is false, then the bridge will not forward traffic back out of the
3154 receiving port. When unset, the kernel's default will be used.
</para>
3158 <term><varname>Isolated=
</varname></term>
3160 <para>Takes a boolean. Configures whether this port is isolated or not. Within a bridge,
3161 isolated ports can only communicate with non-isolated ports. When set to true, this port can only
3162 communicate with other ports whose Isolated setting is false. When set to false, this port
3163 can communicate with any other ports. When unset, the kernel's default will be used.
</para>
3167 <term><varname>UseBPDU=
</varname></term>
3169 <para>Takes a boolean. Configures whether STP Bridge Protocol Data Units will be
3170 processed by the bridge port. When unset, the kernel's default will be used.
</para>
3174 <term><varname>FastLeave=
</varname></term>
3176 <para>Takes a boolean. This flag allows the bridge to immediately stop multicast
3177 traffic on a port that receives an IGMP Leave message. It is only used with
3178 IGMP snooping if enabled on the bridge. When unset, the kernel's default will be used.
</para>
3182 <term><varname>AllowPortToBeRoot=
</varname></term>
3184 <para>Takes a boolean. Configures whether a given port is allowed to
3185 become a root port. Only used when STP is enabled on the bridge.
3186 When unset, the kernel's default will be used.
</para>
3190 <term><varname>ProxyARP=
</varname></term>
3192 <para>Takes a boolean. Configures whether proxy ARP to be enabled on this port.
3193 When unset, the kernel's default will be used.
</para>
3197 <term><varname>ProxyARPWiFi=
</varname></term>
3199 <para>Takes a boolean. Configures whether proxy ARP to be enabled on this port
3200 which meets extended requirements by IEEE
802.11 and Hotspot
2.0 specifications.
3201 When unset, the kernel's default will be used.
</para>
3205 <term><varname>MulticastRouter=
</varname></term>
3207 <para>Configures this port for having multicast routers attached. A port with a multicast
3208 router will receive all multicast traffic. Takes one of
<literal>no
</literal>
3209 to disable multicast routers on this port,
<literal>query
</literal> to let the system detect
3210 the presence of routers,
<literal>permanent
</literal> to permanently enable multicast traffic
3211 forwarding on this port, or
<literal>temporary
</literal> to enable multicast routers temporarily
3212 on this port, not depending on incoming queries. When unset, the kernel's default will be used.
</para>
3216 <term><varname>Cost=
</varname></term>
3218 <para>Sets the
"cost" of sending packets of this interface.
3219 Each port in a bridge may have a different speed and the cost
3220 is used to decide which link to use. Faster interfaces
3221 should have lower costs. It is an integer value between
1 and
3226 <term><varname>Priority=
</varname></term>
3228 <para>Sets the
"priority" of sending packets on this interface.
3229 Each port in a bridge may have a different priority which is used
3230 to decide which link to use. Lower value means higher priority.
3231 It is an integer value between
0 to
63. Networkd does not set any
3232 default, meaning the kernel default value of
32 is used.
</para>
3238 <title>[BridgeFDB] Section Options
</title>
3239 <para>The [BridgeFDB] section manages the forwarding database table of a port and accepts the following
3240 keys. Specify several [BridgeFDB] sections to configure several static MAC table entries.
</para>
3242 <variablelist class='network-directives'
>
3244 <term><varname>MACAddress=
</varname></term>
3246 <para>As in the [Network] section. This key is mandatory.
</para>
3250 <term><varname>Destination=
</varname></term>
3252 <para>Takes an IP address of the destination VXLAN tunnel endpoint.
</para>
3256 <term><varname>VLANId=
</varname></term>
3258 <para>The VLAN ID for the new static MAC table entry. If
3259 omitted, no VLAN ID information is appended to the new static MAC
3264 <term><varname>VNI=
</varname></term>
3266 <para>The VXLAN Network Identifier (or VXLAN Segment ID) to use to connect to
3267 the remote VXLAN tunnel endpoint. Takes a number in the range
1…
16777215.
3268 Defaults to unset.
</para>
3272 <term><varname>AssociatedWith=
</varname></term>
3274 <para>Specifies where the address is associated with. Takes one of
<literal>use
</literal>,
3275 <literal>self
</literal>,
<literal>master
</literal> or
<literal>router
</literal>.
3276 <literal>use
</literal> means the address is in use. User space can use this option to
3277 indicate to the kernel that the fdb entry is in use.
<literal>self
</literal> means
3278 the address is associated with the port drivers fdb. Usually hardware.
<literal>master
</literal>
3279 means the address is associated with master devices fdb.
<literal>router
</literal> means
3280 the destination address is associated with a router. Note that it's valid if the referenced
3281 device is a VXLAN type device and has route shortcircuit enabled. Defaults to
<literal>self
</literal>.
</para>
3285 <term><varname>OutgoingInterface=
</varname></term>
3287 <para>Specifies the name or index of the outgoing interface for the VXLAN device driver to
3288 reach the remote VXLAN tunnel endpoint. Defaults to unset.
</para>
3294 <title>[BridgeMDB] Section Options
</title>
3295 <para>The [BridgeMDB] section manages the multicast membership entries forwarding database table of a port and accepts the following
3296 keys. Specify several [BridgeMDB] sections to configure several permanent multicast membership entries.
</para>
3298 <variablelist class='network-directives'
>
3300 <term><varname>MulticastGroupAddress=
</varname></term>
3302 <para>Specifies the IPv4 or IPv6 multicast group address to add. This setting is mandatory.
</para>
3306 <term><varname>VLANId=
</varname></term>
3308 <para>The VLAN ID for the new entry. Valid ranges are
0 (no VLAN) to
4094. Optional, defaults to
0.
</para>
3315 <title>[LLDP] Section Options
</title>
3316 <para>The [LLDP] section manages the Link Layer Discovery Protocol (LLDP) and accepts the following
3318 <variablelist class='network-directives'
>
3320 <term><varname>MUDURL=
</varname></term>
3322 <para>When configured, the specified Manufacturer Usage Descriptions (MUD) URL will be sent in
3323 LLDP packets. The syntax and semantics are the same as for
<varname>MUDURL=
</varname> in the
3324 [DHCPv4] section described above.
</para>
3326 <para>The MUD URLs received via LLDP packets are saved and can be read using the
3327 <function>sd_lldp_neighbor_get_mud_url()
</function> function.
</para>
3334 <title>[CAN] Section Options
</title>
3335 <para>The [CAN] section manages the Controller Area Network (CAN bus) and accepts the
3336 following keys:
</para>
3337 <variablelist class='network-directives'
>
3339 <term><varname>BitRate=
</varname></term>
3341 <para>The bitrate of CAN device in bits per second. The usual SI prefixes (K, M) with the base of
1000 can
3342 be used here. Takes a number in the range
1…
4294967295.
</para>
3346 <term><varname>SamplePoint=
</varname></term>
3348 <para>Optional sample point in percent with one decimal (e.g.
<literal>75%
</literal>,
3349 <literal>87.5%
</literal>) or permille (e.g.
<literal>875‰
</literal>). This will be ignored when
3350 <varname>BitRate=
</varname> is unspecified.
</para>
3354 <term><varname>TimeQuantaNSec=
</varname></term>
3355 <term><varname>PropagationSegment=
</varname></term>
3356 <term><varname>PhaseBufferSegment1=
</varname></term>
3357 <term><varname>PhaseBufferSegment2=
</varname></term>
3358 <term><varname>SyncJumpWidth=
</varname></term>
3360 <para>Specifies the time quanta, propagation segment, phase buffer segment
1 and
2, and the
3361 synchronization jump width, which allow one to define the CAN bit-timing in a hardware
3362 independent format as proposed by the Bosch CAN
2.0 Specification.
3363 <varname>TimeQuantaNSec=
</varname> takes a timespan in nanoseconds.
3364 <varname>PropagationSegment=
</varname>,
<varname>PhaseBufferSegment1=
</varname>,
3365 <varname>PhaseBufferSegment2=
</varname>, and
<varname>SyncJumpWidth=
</varname> take number
3366 of time quantum specified in
<varname>TimeQuantaNSec=
</varname> and must be an unsigned
3367 integer in the range
0…
4294967295. These settings except for
3368 <varname>SyncJumpWidth=
</varname> will be ignored when
<varname>BitRate=
</varname> is
3373 <term><varname>DataBitRate=
</varname></term>
3374 <term><varname>DataSamplePoint=
</varname></term>
3376 <para>The bitrate and sample point for the data phase, if CAN-FD is used. These settings are
3377 analogous to the
<varname>BitRate=
</varname> and
<varname>SamplePoint=
</varname> keys.
</para>
3381 <term><varname>DataTimeQuantaNSec=
</varname></term>
3382 <term><varname>DataPropagationSegment=
</varname></term>
3383 <term><varname>DataPhaseBufferSegment1=
</varname></term>
3384 <term><varname>DataPhaseBufferSegment2=
</varname></term>
3385 <term><varname>DataSyncJumpWidth=
</varname></term>
3387 <para>Specifies the time quanta, propagation segment, phase buffer segment
1 and
2, and the
3388 synchronization jump width for the data phase, if CAN-FD is used. These settings are
3389 analogous to the
<varname>TimeQuantaNSec=
</varname> or related settings.
</para>
3393 <term><varname>FDMode=
</varname></term>
3395 <para>Takes a boolean. When
<literal>yes
</literal>, CAN-FD mode is enabled for the interface.
3396 Note, that a bitrate and optional sample point should also be set for the CAN-FD data phase using
3397 the
<varname>DataBitRate=
</varname> and
<varname>DataSamplePoint=
</varname> keys, or
3398 <varname>DataTimeQuanta=
</varname> and related settings.
</para>
3402 <term><varname>FDNonISO=
</varname></term>
3404 <para>Takes a boolean. When
<literal>yes
</literal>, non-ISO CAN-FD mode is enabled for the
3405 interface. When unset, the kernel's default will be used.
</para>
3409 <term><varname>RestartSec=
</varname></term>
3411 <para>Automatic restart delay time. If set to a non-zero value, a restart of the CAN controller will be
3412 triggered automatically in case of a bus-off condition after the specified delay time. Subsecond delays can
3413 be specified using decimals (e.g.
<literal>0.1s
</literal>) or a
<literal>ms
</literal> or
3414 <literal>us
</literal> postfix. Using
<literal>infinity
</literal> or
<literal>0</literal> will turn the
3415 automatic restart off. By default automatic restart is disabled.
</para>
3419 <term><varname>Termination=
</varname></term>
3421 <para>Takes a boolean or a termination resistor value in ohm in the range
0…
65535. When
3422 <literal>yes
</literal>, the termination resistor is set to
120 ohm. When
3423 <literal>no
</literal> or
<literal>0</literal> is set, the termination resistor is disabled.
3424 When unset, the kernel's default will be used.
</para>
3428 <term><varname>TripleSampling=
</varname></term>
3430 <para>Takes a boolean. When
<literal>yes
</literal>, three samples (instead of one) are used to determine
3431 the value of a received bit by majority rule. When unset, the kernel's default will be used.
</para>
3435 <term><varname>BusErrorReporting=
</varname></term>
3437 <para>Takes a boolean. When
<literal>yes
</literal>, reporting of CAN bus errors is activated
3438 (those include single bit, frame format, and bit stuffing errors, unable to send dominant bit,
3439 unable to send recessive bit, bus overload, active error announcement, error occurred on
3440 transmission). When unset, the kernel's default will be used. Note: in case of a CAN bus with a
3441 single CAN device, sending a CAN frame may result in a huge number of CAN bus errors.
</para>
3445 <term><varname>ListenOnly=
</varname></term>
3447 <para>Takes a boolean. When
<literal>yes
</literal>, listen-only mode is enabled. When the
3448 interface is in listen-only mode, the interface neither transmit CAN frames nor send ACK
3449 bit. Listen-only mode is important to debug CAN networks without interfering with the
3450 communication or acknowledge the CAN frame. When unset, the kernel's default will be used.
3455 <term><varname>Loopback=
</varname></term>
3457 <para>Takes a boolean. When
<literal>yes
</literal>, loopback mode is enabled. When the
3458 loopback mode is enabled, the interface treats messages transmitted by itself as received
3459 messages. The loopback mode is important to debug CAN networks. When unset, the kernel's
3460 default will be used.
</para>
3464 <term><varname>OneShot=
</varname></term>
3466 <para>Takes a boolean. When
<literal>yes
</literal>, one-shot mode is enabled. When unset,
3467 the kernel's default will be used.
</para>
3471 <term><varname>PresumeAck=
</varname></term>
3473 <para>Takes a boolean. When
<literal>yes
</literal>, the interface will ignore missing CAN
3474 ACKs. When unset, the kernel's default will be used.
</para>
3478 <term><varname>ClassicDataLengthCode=
</varname></term>
3480 <para>Takes a boolean. When
<literal>yes
</literal>, the interface will handle the
4bit data
3481 length code (DLC). When unset, the kernel's default will be used.
</para>
3488 <title>[IPoIB] Section Options
</title>
3489 <para>The [IPoIB] section manages the IP over Infiniband and accepts the following keys:
</para>
3490 <variablelist class='network-directives'
>
3491 <xi:include href=
"systemd.netdev.xml" xpointer=
"ipoib_mode" />
3492 <xi:include href=
"systemd.netdev.xml" xpointer=
"ipoib_umcast" />
3497 <title>[QDisc] Section Options
</title>
3498 <para>The [QDisc] section manages the traffic control queueing discipline (qdisc).
</para>
3500 <variablelist class='network-directives'
>
3502 <term><varname>Parent=
</varname></term>
3504 <para>Specifies the parent Queueing Discipline (qdisc). Takes one of
<literal>clsact
</literal>
3505 or
<literal>ingress
</literal>. This is mandatory.
</para>
3509 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3514 <title>[NetworkEmulator] Section Options
</title>
3515 <para>The [NetworkEmulator] section manages the queueing discipline (qdisc) of the network emulator. It
3516 can be used to configure the kernel packet scheduler and simulate packet delay and loss for UDP or TCP
3517 applications, or limit the bandwidth usage of a particular service to simulate internet connections.
3520 <variablelist class='network-directives'
>
3521 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3522 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3525 <term><varname>DelaySec=
</varname></term>
3527 <para>Specifies the fixed amount of delay to be added to all packets going out of the
3528 interface. Defaults to unset.
</para>
3533 <term><varname>DelayJitterSec=
</varname></term>
3535 <para>Specifies the chosen delay to be added to the packets outgoing to the network
3536 interface. Defaults to unset.
</para>
3541 <term><varname>PacketLimit=
</varname></term>
3543 <para>Specifies the maximum number of packets the qdisc may hold queued at a time.
3544 An unsigned integer in the range
0…
4294967294. Defaults to
1000.
</para>
3549 <term><varname>LossRate=
</varname></term>
3551 <para>Specifies an independent loss probability to be added to the packets outgoing from the
3552 network interface. Takes a percentage value, suffixed with
"%". Defaults to unset.
</para>
3557 <term><varname>DuplicateRate=
</varname></term>
3559 <para>Specifies that the chosen percent of packets is duplicated before queuing them.
3560 Takes a percentage value, suffixed with
"%". Defaults to unset.
</para>
3567 <title>[TokenBucketFilter] Section Options
</title>
3568 <para>The [TokenBucketFilter] section manages the queueing discipline (qdisc) of token bucket filter
3571 <variablelist class='network-directives'
>
3572 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3573 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3576 <term><varname>LatencySec=
</varname></term>
3578 <para>Specifies the latency parameter, which specifies the maximum amount of time a
3579 packet can sit in the Token Bucket Filter (TBF). Defaults to unset.
</para>
3584 <term><varname>LimitBytes=
</varname></term>
3586 <para>Takes the number of bytes that can be queued waiting for tokens to become available.
3587 When the size is suffixed with K, M, or G, it is parsed as Kilobytes, Megabytes, or Gigabytes,
3588 respectively, to the base of
1024. Defaults to unset.
</para>
3593 <term><varname>BurstBytes=
</varname></term>
3595 <para>Specifies the size of the bucket. This is the maximum amount of bytes that tokens
3596 can be available for instantaneous transfer. When the size is suffixed with K, M, or G, it is
3597 parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to
3603 <term><varname>Rate=
</varname></term>
3605 <para>Specifies the device specific bandwidth. When suffixed with K, M, or G, the specified
3606 bandwidth is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of
1000.
3607 Defaults to unset.
</para>
3612 <term><varname>MPUBytes=
</varname></term>
3614 <para>The Minimum Packet Unit (MPU) determines the minimal token usage (specified in bytes)
3615 for a packet. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,
3616 Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to zero.
</para>
3621 <term><varname>PeakRate=
</varname></term>
3623 <para>Takes the maximum depletion rate of the bucket. When suffixed with K, M, or G, the
3624 specified size is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of
3625 1000. Defaults to unset.
</para>
3630 <term><varname>MTUBytes=
</varname></term>
3632 <para>Specifies the size of the peakrate bucket. When suffixed with K, M, or G, the specified
3633 size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
1024.
3634 Defaults to unset.
</para>
3641 <title>[PIE] Section Options
</title>
3642 <para>The [PIE] section manages the queueing discipline (qdisc) of Proportional Integral
3643 controller-Enhanced (PIE).
</para>
3645 <variablelist class='network-directives'
>
3646 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3647 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3650 <term><varname>PacketLimit=
</varname></term>
3652 <para>Specifies the hard limit on the queue size in number of packets. When this limit is reached,
3653 incoming packets are dropped. An unsigned integer in the range
1…
4294967294. Defaults to unset and
3654 kernel's default is used.
</para>
3661 <title>[FlowQueuePIE] Section Options
</title>
3662 <para>The
<literal>[FlowQueuePIE]
</literal> section manages the queueing discipline
3663 (qdisc) of Flow Queue Proportional Integral controller-Enhanced (fq_pie).
</para>
3665 <variablelist class='network-directives'
>
3666 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3667 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3670 <term><varname>PacketLimit=
</varname></term>
3672 <para>Specifies the hard limit on the queue size in number of packets. When this limit is reached,
3673 incoming packets are dropped. An unsigned integer ranges
1 to
4294967294. Defaults to unset and
3674 kernel's default is used.
</para>
3681 <title>[StochasticFairBlue] Section Options
</title>
3682 <para>The [StochasticFairBlue] section manages the queueing discipline (qdisc) of stochastic fair blue
3685 <variablelist class='network-directives'
>
3686 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3687 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3690 <term><varname>PacketLimit=
</varname></term>
3692 <para>Specifies the hard limit on the queue size in number of packets. When this limit is reached,
3693 incoming packets are dropped. An unsigned integer in the range
0…
4294967294. Defaults to unset and
3694 kernel's default is used.
</para>
3701 <title>[StochasticFairnessQueueing] Section Options
</title>
3702 <para>The [StochasticFairnessQueueing] section manages the queueing discipline (qdisc) of stochastic
3703 fairness queueing (sfq).
</para>
3705 <variablelist class='network-directives'
>
3706 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3707 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3710 <term><varname>PerturbPeriodSec=
</varname></term>
3712 <para>Specifies the interval in seconds for queue algorithm perturbation. Defaults to unset.
</para>
3719 <title>[BFIFO] Section Options
</title>
3720 <para>The [BFIFO] section manages the queueing discipline (qdisc) of Byte limited Packet First In First
3723 <variablelist class='network-directives'
>
3724 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3725 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3728 <term><varname>LimitBytes=
</varname></term>
3730 <para>Specifies the hard limit in bytes on the FIFO buffer size. The size limit prevents overflow
3731 in case the kernel is unable to dequeue packets as quickly as it receives them. When this limit is
3732 reached, incoming packets are dropped. When suffixed with K, M, or G, the specified size is parsed
3733 as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to unset and
3734 kernel default is used.
</para>
3741 <title>[PFIFO] Section Options
</title>
3742 <para>The [PFIFO] section manages the queueing discipline (qdisc) of Packet First In First Out
3745 <variablelist class='network-directives'
>
3746 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3747 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3750 <term><varname>PacketLimit=
</varname></term>
3752 <para>Specifies the hard limit on the number of packets in the FIFO queue. The size limit prevents
3753 overflow in case the kernel is unable to dequeue packets as quickly as it receives them. When this
3754 limit is reached, incoming packets are dropped. An unsigned integer in the range
3755 0…
4294967294. Defaults to unset and kernel's default is used.
</para>
3762 <title>[PFIFOHeadDrop] Section Options
</title>
3763 <para>The [PFIFOHeadDrop] section manages the queueing discipline (qdisc) of Packet First In First Out
3764 Head Drop (pfifo_head_drop).
</para>
3766 <variablelist class='network-directives'
>
3767 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3768 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3771 <term><varname>PacketLimit=
</varname></term>
3773 <para>As in [PFIFO] section.
</para></listitem>
3779 <title>[PFIFOFast] Section Options
</title>
3780 <para>The [PFIFOFast] section manages the queueing discipline (qdisc) of Packet First In First Out Fast
3781 (pfifo_fast).
</para>
3783 <variablelist class='network-directives'
>
3784 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3785 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3790 <title>[CAKE] Section Options
</title>
3791 <para>The [CAKE] section manages the queueing discipline (qdisc) of Common Applications Kept Enhanced
3794 <variablelist class='network-directives'
>
3795 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
3796 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
3799 <term><varname>Bandwidth=
</varname></term>
3801 <para>Specifies the shaper bandwidth. When suffixed with K, M, or G, the specified size is
3802 parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of
1000. Defaults to
3803 unset and kernel's default is used.
</para>
3808 <term><varname>AutoRateIngress=
</varname></term>
3810 <para>Takes a boolean value. Enables automatic capacity estimation based on traffic arriving
3811 at this qdisc. This is most likely to be useful with cellular links, which tend to change
3812 quality randomly. If this setting is enabled, the
<varname>Bandwidth=
</varname> setting is
3813 used as an initial estimate. Defaults to unset, and the kernel's default is used.
</para>
3818 <term><varname>OverheadBytes=
</varname></term>
3820 <para>Specifies that bytes to be addeded to the size of each packet. Bytes may be negative.
3821 Takes an integer in the range -
64…
256. Defaults to unset and kernel's default is used.
3827 <term><varname>MPUBytes=
</varname></term>
3829 <para>Rounds each packet (including overhead) up to the specified bytes. Takes an integer in
3830 the range
1…
256. Defaults to unset and kernel's default is used.
</para>
3835 <term><varname>CompensationMode=
</varname></term>
3837 <para>Takes one of
<literal>none
</literal>,
<literal>atm
</literal>, or
<literal>ptm
</literal>.
3838 Specifies the compensation mode for overhead calculation. When
<literal>none
</literal>, no
3839 compensation is taken into account. When
<literal>atm
</literal>, enables the compensation for
3840 ATM cell framing, which is normally found on ADSL links. When
<literal>ptm
</literal>, enables
3841 the compensation for PTM encoding, which is normally found on VDSL2 links and uses a
64b/
65b
3842 encoding scheme. Defaults to unset and the kernel's default is used.
</para>
3847 <term><varname>UseRawPacketSize=
</varname></term>
3849 <para>Takes a boolean value. When true, the packet size reported by the Linux kernel will be
3850 used, instead of the underlying IP packet size. Defaults to unset, and the kernel's default
3856 <term><varname>FlowIsolationMode=
</varname></term>
3858 <para>CAKE places packets from different flows into different queues, then packets from each
3859 queue are delivered fairly. This specifies whether the fairness is based on source address,
3860 destination address, individual flows, or any combination of those. The available values are:
3865 <term><option>none
</option></term>
3867 The flow isolation is disabled, and all traffic passes through a single queue.
3871 <term><option>src-host
</option></term>
3873 Flows are defined only by source address. Equivalent to the
<literal>srchost
</literal>
3874 option for
<command>tc qdisc
</command> command. See also
3875 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3879 <term><option>dst-host
</option></term>
3881 Flows are defined only by destination address. Equivalent to the
3882 <literal>dsthost
</literal> option for
<command>tc qdisc
</command> command. See also
3883 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3887 <term><option>hosts
</option></term>
3889 Flows are defined by source-destination host pairs. Equivalent to the same option for
3890 <command>tc qdisc
</command> command. See also
3891 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3895 <term><option>flows
</option></term>
3897 Flows are defined by the entire
5-tuple of source address, destination address,
3898 transport protocol, source port and destination port. Equivalent to the same option for
3899 <command>tc qdisc
</command> command. See also
3900 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3904 <term><option>dual-src-host
</option></term>
3906 Flows are defined by the
5-tuple (see
<literal>flows
</literal> in the above), and
3907 fairness is applied first over source addresses, then over individual flows. Equivalent
3908 to the
<literal>dual-srchost
</literal> option for
<command>tc qdisc
</command> command.
3910 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3914 <term><option>dual-dst-host
</option></term>
3916 Flows are defined by the
5-tuple (see
<literal>flows
</literal> in the above), and
3917 fairness is applied first over destination addresses, then over individual flows.
3918 Equivalent to the
<literal>dual-dsthost
</literal> option for
3919 <command>tc qdisc
</command> command. See also
3920 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3924 <term><option>triple
</option></term>
3926 Flows are defined by the
5-tuple (see
<literal>flows
</literal>), and fairness is
3927 applied over source and destination addresses, and also over individual flows.
3928 Equivalent to the
<literal>triple-isolate
</literal> option for
3929 <command>tc qdisc
</command> command. See also
3930 <citerefentry project='man-pages'
><refentrytitle>tc-cake
</refentrytitle><manvolnum>8</manvolnum></citerefentry>.
3935 <para>Defaults to unset and the kernel's default is used.
</para>
3940 <term><varname>NAT=
</varname></term>
3942 <para>Takes a boolean value. When true, CAKE performs a NAT lookup before applying
3943 flow-isolation rules, to determine the true addresses and port numbers of the packet, to
3944 improve fairness between hosts inside the NAT. This has no practical effect when
3945 <varname>FlowIsolationMode=
</varname> is
<literal>none
</literal> or
<literal>flows
</literal>,
3946 or if NAT is performed on a different host. Defaults to unset, and the kernel's default is
3952 <term><varname>PriorityQueueingPreset=
</varname></term>
3954 <para>CAKE divides traffic into
<literal>tins
</literal>, and each tin has its own independent
3955 set of flow-isolation queues, bandwidth threshold, and priority. This specifies the preset of
3956 tin profiles. The available values are:
</para>
3960 <term><option>besteffort
</option></term>
3962 Disables priority queueing by placing all traffic in one tin.
3966 <term><option>precedence
</option></term>
3968 Enables priority queueing based on the legacy interpretation of TOS
3969 <literal>Precedence
</literal> field. Use of this preset on the modern Internet is
3974 <term><option>diffserv8
</option></term>
3976 Enables priority queueing based on the Differentiated Service
3977 (
<literal>DiffServ
</literal>) field with eight tins: Background Traffic, High
3978 Throughput, Best Effort, Video Streaming, Low Latency Transactions, Interactive Shell,
3979 Minimum Latency, and Network Control.
3983 <term><option>diffserv4
</option></term>
3985 Enables priority queueing based on the Differentiated Service
3986 (
<literal>DiffServ
</literal>) field with four tins: Background Traffic, Best Effort,
3987 Streaming Media, and Latency Sensitive.
3991 <term><option>diffserv3
</option></term>
3993 Enables priority queueing based on the Differentiated Service
3994 (
<literal>DiffServ
</literal>) field with three tins: Background Traffic, Best Effort,
3995 and Latency Sensitive.
4000 <para>Defaults to unset, and the kernel's default is used.
</para>
4005 <term><varname>FirewallMark=
</varname></term>
4007 <para>Takes an integer in the range
1…
4294967295. When specified, firewall-mark-based
4008 overriding of CAKE's tin selection is enabled. Defaults to unset, and the kernel's default is
4014 <term><varname>Wash=
</varname></term>
4016 <para>Takes a boolean value. When true, CAKE clears the DSCP fields, except for ECN bits, of
4017 any packet passing through CAKE. Defaults to unset, and the kernel's default is used.
</para>
4022 <term><varname>SplitGSO=
</varname></term>
4024 <para>Takes a boolean value. When true, CAKE will split General Segmentation Offload (GSO)
4025 super-packets into their on-the-wire components and dequeue them individually. Defaults to
4026 unset, and the kernel's default is used.
</para>
4034 <title>[ControlledDelay] Section Options
</title>
4035 <para>The [ControlledDelay] section manages the queueing discipline (qdisc) of
4036 controlled delay (CoDel).
</para>
4038 <variablelist class='network-directives'
>
4039 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4040 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4043 <term><varname>PacketLimit=
</varname></term>
4045 <para>Specifies the hard limit on the queue size in number of packets. When this limit is reached,
4046 incoming packets are dropped. An unsigned integer in the range
0…
4294967294. Defaults to unset and
4047 kernel's default is used.
</para>
4052 <term><varname>TargetSec=
</varname></term>
4054 <para>Takes a timespan. Specifies the acceptable minimum standing/persistent queue delay.
4055 Defaults to unset and kernel's default is used.
</para>
4060 <term><varname>IntervalSec=
</varname></term>
4062 <para>Takes a timespan. This is used to ensure that the measured minimum delay does not
4063 become too stale. Defaults to unset and kernel's default is used.
</para>
4068 <term><varname>ECN=
</varname></term>
4070 <para>Takes a boolean. This can be used to mark packets instead of dropping them. Defaults to
4071 unset and kernel's default is used.
</para>
4076 <term><varname>CEThresholdSec=
</varname></term>
4078 <para>Takes a timespan. This sets a threshold above which all packets are marked with ECN
4079 Congestion Experienced (CE). Defaults to unset and kernel's default is used.
</para>
4086 <title>[DeficitRoundRobinScheduler] Section Options
</title>
4087 <para>The [DeficitRoundRobinScheduler] section manages the queueing discipline (qdisc) of Deficit Round
4088 Robin Scheduler (DRR).
</para>
4090 <variablelist class='network-directives'
>
4091 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4092 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4097 <title>[DeficitRoundRobinSchedulerClass] Section Options
</title>
4098 <para>The [DeficitRoundRobinSchedulerClass] section manages the traffic control class of Deficit Round
4099 Robin Scheduler (DRR).
</para>
4101 <variablelist class='network-directives'
>
4102 <xi:include href=
"tc.xml" xpointer=
"tclass-parent" />
4103 <xi:include href=
"tc.xml" xpointer=
"tclass-classid" />
4106 <term><varname>QuantumBytes=
</varname></term>
4108 <para>Specifies the amount of bytes a flow is allowed to dequeue before the scheduler moves
4109 to the next class. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,
4110 Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to the MTU of the
4119 <title>[EnhancedTransmissionSelection] Section Options
</title>
4120 <para>The [EnhancedTransmissionSelection] section manages the queueing discipline (qdisc) of Enhanced
4121 Transmission Selection (ETS).
</para>
4123 <variablelist class='network-directives'
>
4124 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4125 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4128 <term><varname>Bands=
</varname></term>
4130 <para>Specifies the number of bands. An unsigned integer in the range
1…
16. This value has to be at
4131 least large enough to cover the strict bands specified through the
<varname>StrictBands=
</varname>
4132 and bandwidth-sharing bands specified in
<varname>QuantumBytes=
</varname>.
</para>
4137 <term><varname>StrictBands=
</varname></term>
4139 <para>Specifies the number of bands that should be created in strict mode. An unsigned integer in
4140 the range
1…
16.
</para>
4145 <term><varname>QuantumBytes=
</varname></term>
4147 <para>Specifies the white-space separated list of quantum used in band-sharing bands. When
4148 suffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or Gigabytes,
4149 respectively, to the base of
1024. This setting can be specified multiple times. If an empty
4150 string is assigned, then the all previous assignments are cleared.
</para>
4155 <term><varname>PriorityMap=
</varname></term>
4157 <para>The priority map maps the priority of a packet to a band. The argument is a whitespace
4158 separated list of numbers. The first number indicates which band the packets with priority
0 should
4159 be put to, the second is for priority
1, and so on. There can be up to
16 numbers in the list. If
4160 there are fewer, the default band that traffic with one of the unmentioned priorities goes to is
4161 the last one. Each band number must be in the range
0…
255. This setting can be specified multiple
4162 times. If an empty string is assigned, then the all previous assignments are cleared.
</para>
4169 <title>[GenericRandomEarlyDetection] Section Options
</title>
4170 <para>The [GenericRandomEarlyDetection] section manages the queueing discipline (qdisc) of Generic Random
4171 Early Detection (GRED).
</para>
4173 <variablelist class='network-directives'
>
4174 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4175 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4178 <term><varname>VirtualQueues=
</varname></term>
4180 <para>Specifies the number of virtual queues. Takes an integer in the range
1…
16. Defaults to unset
4181 and kernel's default is used.
</para>
4186 <term><varname>DefaultVirtualQueue=
</varname></term>
4188 <para>Specifies the number of default virtual queue. This must be less than
<varname>VirtualQueue=
</varname>.
4189 Defaults to unset and kernel's default is used.
</para>
4194 <term><varname>GenericRIO=
</varname></term>
4196 <para>Takes a boolean. It turns on the RIO-like buffering scheme. Defaults to
4197 unset and kernel's default is used.
</para>
4204 <title>[FairQueueingControlledDelay] Section Options
</title>
4205 <para>The [FairQueueingControlledDelay] section manages the queueing discipline (qdisc) of fair queuing
4206 controlled delay (FQ-CoDel).
</para>
4208 <variablelist class='network-directives'
>
4209 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4210 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4213 <term><varname>PacketLimit=
</varname></term>
4215 <para>Specifies the hard limit on the real queue size. When this limit is reached, incoming packets are
4216 dropped. Defaults to unset and kernel's default is used.
</para>
4221 <term><varname>MemoryLimitBytes=
</varname></term>
4223 <para>Specifies the limit on the total number of bytes that can be queued in this FQ-CoDel instance.
4224 When suffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or Gigabytes,
4225 respectively, to the base of
1024. Defaults to unset and kernel's default is used.
</para>
4230 <term><varname>Flows=
</varname></term>
4232 <para>Specifies the number of flows into which the incoming packets are classified.
4233 Defaults to unset and kernel's default is used.
</para>
4238 <term><varname>TargetSec=
</varname></term>
4240 <para>Takes a timespan. Specifies the acceptable minimum standing/persistent queue delay.
4241 Defaults to unset and kernel's default is used.
</para>
4246 <term><varname>IntervalSec=
</varname></term>
4248 <para>Takes a timespan. This is used to ensure that the measured minimum delay does not
4249 become too stale. Defaults to unset and kernel's default is used.
</para>
4254 <term><varname>QuantumBytes=
</varname></term>
4256 <para>Specifies the number of bytes used as the
"deficit" in the fair queuing algorithm timespan.
4257 When suffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or Gigabytes,
4258 respectively, to the base of
1024. Defaults to unset and kernel's default is used.
</para>
4263 <term><varname>ECN=
</varname></term>
4265 <para>Takes a boolean. This can be used to mark packets instead of dropping them. Defaults to
4266 unset and kernel's default is used.
</para>
4271 <term><varname>CEThresholdSec=
</varname></term>
4273 <para>Takes a timespan. This sets a threshold above which all packets are marked with ECN
4274 Congestion Experienced (CE). Defaults to unset and kernel's default is used.
</para>
4281 <title>[FairQueueing] Section Options
</title>
4282 <para>The [FairQueueing] section manages the queueing discipline (qdisc) of fair queue traffic policing
4285 <variablelist class='network-directives'
>
4286 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4287 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4290 <term><varname>PacketLimit=
</varname></term>
4292 <para>Specifies the hard limit on the real queue size. When this limit is reached, incoming packets are
4293 dropped. Defaults to unset and kernel's default is used.
</para>
4298 <term><varname>FlowLimit=
</varname></term>
4300 <para>Specifies the hard limit on the maximum number of packets queued per flow. Defaults to
4301 unset and kernel's default is used.
</para>
4306 <term><varname>QuantumBytes=
</varname></term>
4308 <para>Specifies the credit per dequeue RR round, i.e. the amount of bytes a flow is allowed
4309 to dequeue at once. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,
4310 Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to unset and kernel's
4311 default is used.
</para>
4316 <term><varname>InitialQuantumBytes=
</varname></term>
4318 <para>Specifies the initial sending rate credit, i.e. the amount of bytes a new flow is
4319 allowed to dequeue initially. When suffixed with K, M, or G, the specified size is parsed as
4320 Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
1024. Defaults to unset and
4321 kernel's default is used.
</para>
4326 <term><varname>MaximumRate=
</varname></term>
4328 <para>Specifies the maximum sending rate of a flow. When suffixed with K, M, or G, the
4329 specified size is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of
4330 1000. Defaults to unset and kernel's default is used.
</para>
4335 <term><varname>Buckets=
</varname></term>
4337 <para>Specifies the size of the hash table used for flow lookups. Defaults to unset and
4338 kernel's default is used.
</para>
4343 <term><varname>OrphanMask=
</varname></term>
4345 <para>Takes an unsigned integer. For packets not owned by a socket, fq is able to mask a part
4346 of hash and reduce number of buckets associated with the traffic. Defaults to unset and
4347 kernel's default is used.
</para>
4352 <term><varname>Pacing=
</varname></term>
4354 <para>Takes a boolean, and enables or disables flow pacing. Defaults to unset and kernel's
4355 default is used.
</para>
4360 <term><varname>CEThresholdSec=
</varname></term>
4362 <para>Takes a timespan. This sets a threshold above which all packets are marked with ECN
4363 Congestion Experienced (CE). Defaults to unset and kernel's default is used.
</para>
4370 <title>[TrivialLinkEqualizer] Section Options
</title>
4371 <para>The [TrivialLinkEqualizer] section manages the queueing discipline (qdisc) of trivial link
4372 equalizer (teql).
</para>
4374 <variablelist class='network-directives'
>
4375 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4376 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4379 <term><varname>Id=
</varname></term>
4381 <para>Specifies the interface ID
<literal>N
</literal> of teql. Defaults to
<literal>0</literal>.
4382 Note that when teql is used, currently, the module
<constant>sch_teql
</constant> with
4383 <constant>max_equalizers=N+
1</constant> option must be loaded before
4384 <command>systemd-networkd
</command> is started.
</para>
4391 <title>[HierarchyTokenBucket] Section Options
</title>
4392 <para>The [HierarchyTokenBucket] section manages the queueing discipline (qdisc) of hierarchy token
4393 bucket (htb).
</para>
4395 <variablelist class='network-directives'
>
4396 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4397 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4400 <term><varname>DefaultClass=
</varname></term>
4402 <para>Takes the minor id in hexadecimal of the default class. Unclassified traffic gets sent
4403 to the class. Defaults to unset.
</para>
4408 <term><varname>RateToQuantum=
</varname></term>
4410 <para>Takes an unsigned integer. The DRR quantums are calculated by dividing the value
4411 configured in
<varname>Rate=
</varname> by
<varname>RateToQuantum=
</varname>.
</para>
4418 <title>[HierarchyTokenBucketClass] Section Options
</title>
4419 <para>The [HierarchyTokenBucketClass] section manages the traffic control class of hierarchy token bucket
4422 <variablelist class='network-directives'
>
4423 <xi:include href=
"tc.xml" xpointer=
"tclass-parent" />
4424 <xi:include href=
"tc.xml" xpointer=
"tclass-classid" />
4427 <term><varname>Priority=
</varname></term>
4429 <para>Specifies the priority of the class. In the round-robin process, classes with the lowest
4430 priority field are tried for packets first.
</para>
4435 <term><varname>QuantumBytes=
</varname></term>
4437 <para>Specifies how many bytes to serve from leaf at once. When suffixed with K, M, or G, the
4438 specified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
4444 <term><varname>MTUBytes=
</varname></term>
4446 <para>Specifies the maximum packet size we create. When suffixed with K, M, or G, the specified
4447 size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
1024.
</para>
4452 <term><varname>OverheadBytes=
</varname></term>
4454 <para>Takes an unsigned integer which specifies per-packet size overhead used in rate
4455 computations. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,
4456 Megabytes, or Gigabytes, respectively, to the base of
1024.
</para>
4461 <term><varname>Rate=
</varname></term>
4463 <para>Specifies the maximum rate this class and all its children are guaranteed. When suffixed
4464 with K, M, or G, the specified size is parsed as Kilobits, Megabits, or Gigabits, respectively,
4465 to the base of
1000. This setting is mandatory.
</para>
4470 <term><varname>CeilRate=
</varname></term>
4472 <para>Specifies the maximum rate at which a class can send, if its parent has bandwidth to spare.
4473 When suffixed with K, M, or G, the specified size is parsed as Kilobits, Megabits, or Gigabits,
4474 respectively, to the base of
1000. When unset, the value specified with
<varname>Rate=
</varname>
4480 <term><varname>BufferBytes=
</varname></term>
4482 <para>Specifies the maximum bytes burst which can be accumulated during idle period. When suffixed
4483 with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively,
4484 to the base of
1024.
</para>
4489 <term><varname>CeilBufferBytes=
</varname></term>
4491 <para>Specifies the maximum bytes burst for ceil which can be accumulated during idle period.
4492 When suffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or Gigabytes,
4493 respectively, to the base of
1024.
</para>
4500 <title>[HeavyHitterFilter] Section Options
</title>
4501 <para>The [HeavyHitterFilter] section manages the queueing discipline (qdisc) of Heavy Hitter Filter
4504 <variablelist class='network-directives'
>
4505 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4506 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4509 <term><varname>PacketLimit=
</varname></term>
4511 <para>Specifies the hard limit on the queue size in number of packets. When this limit is reached,
4512 incoming packets are dropped. An unsigned integer in the range
0…
4294967294. Defaults to unset and
4513 kernel's default is used.
</para>
4520 <title>[QuickFairQueueing] Section Options
</title>
4521 <para>The [QuickFairQueueing] section manages the queueing discipline (qdisc) of Quick Fair Queueing
4524 <variablelist class='network-directives'
>
4525 <xi:include href=
"tc.xml" xpointer=
"qdisc-parent" />
4526 <xi:include href=
"tc.xml" xpointer=
"qdisc-handle" />
4531 <title>[QuickFairQueueingClass] Section Options
</title>
4532 <para>The [QuickFairQueueingClass] section manages the traffic control class of Quick Fair Queueing
4535 <variablelist class='network-directives'
>
4536 <xi:include href=
"tc.xml" xpointer=
"tclass-parent" />
4537 <xi:include href=
"tc.xml" xpointer=
"tclass-classid" />
4540 <term><varname>Weight=
</varname></term>
4542 <para>Specifies the weight of the class. Takes an integer in the range
1…
1023. Defaults to
4543 unset in which case the kernel default is used.
</para>
4548 <term><varname>MaxPacketBytes=
</varname></term>
4550 <para>Specifies the maximum packet size in bytes for the class. When suffixed with K, M, or G, the
4551 specified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of
4552 1024. When unset, the kernel default is used.
</para>
4559 <title>[BridgeVLAN] Section Options
</title>
4560 <para>The [BridgeVLAN] section manages the VLAN ID configuration of a bridge port and accepts the
4561 following keys. Specify several [BridgeVLAN] sections to configure several VLAN entries. The
4562 <varname>VLANFiltering=
</varname> option has to be enabled, see the [Bridge] section in
4563 <citerefentry><refentrytitle>systemd.netdev
</refentrytitle><manvolnum>5</manvolnum></citerefentry>.
</para>
4565 <variablelist class='network-directives'
>
4567 <term><varname>VLAN=
</varname></term>
4569 <para>The VLAN ID allowed on the port. This can be either a single ID or a range M-N. Takes
4570 an integer in the range
1…
4094.
</para>
4574 <term><varname>EgressUntagged=
</varname></term>
4576 <para>The VLAN ID specified here will be used to untag frames on egress. Configuring
4577 <varname>EgressUntagged=
</varname> implicates the use of
<varname>VLAN=
</varname> above and will enable the
4578 VLAN ID for ingress as well. This can be either a single ID or a range M-N.
</para>
4582 <term><varname>PVID=
</varname></term>
4584 <para>The Port VLAN ID specified here is assigned to all untagged frames at ingress.
4585 <varname>PVID=
</varname> can be used only once. Configuring
<varname>PVID=
</varname> implicates the use of
4586 <varname>VLAN=
</varname> above and will enable the VLAN ID for ingress as well.
</para>
4593 <title>Examples
</title>
4595 <title>Static network configuration
</title>
4597 <programlisting># /etc/systemd/network/
50-static.network
4602 Address=
192.168.0.15/
24
4603 Gateway=
192.168.0.1</programlisting>
4605 <para>This brings interface
<literal>enp2s0
</literal> up with a static address. The
4606 specified gateway will be used for a default route.
</para>
4610 <title>DHCP on ethernet links
</title>
4612 <programlisting># /etc/systemd/network/
80-dhcp.network
4617 DHCP=yes
</programlisting>
4619 <para>This will enable DHCPv4 and DHCPv6 on all interfaces with names starting with
4620 <literal>en
</literal> (i.e. ethernet interfaces).
</para>
4624 <title>IPv6 Prefix Delegation (DHCPv6 PD)
</title>
4626 <programlisting># /etc/systemd/network/
55-dhcpv6-pd-upstream.network
4633 # The below setting is optional, to also assign an address in the delegated prefix
4634 # to the upstream interface. If not necessary, then comment out the line below and
4635 # the [DHCPPrefixDelegation] section.
4636 DHCPPrefixDelegation=yes
4638 # If the upstream network provides Router Advertisement with Managed bit set,
4639 # then comment out the line below and WithoutRA= setting in the [DHCPv6] section.
4645 [DHCPPrefixDelegation]
4646 UplinkInterface=:self
4648 Announce=no
</programlisting>
4650 <programlisting># /etc/systemd/network/
55-dhcpv6-pd-downstream.network
4655 DHCPPrefixDelegation=yes
4658 # It is expected that the host is acting as a router. So, usually it is not
4659 # necessary to receive Router Advertisement from other hosts in the downstream network.
4662 [DHCPPrefixDelegation]
4663 UplinkInterface=enp1s0
4665 Announce=yes
</programlisting>
4667 <para>This will enable DHCPv6-PD on the interface enp1s0 as an upstream interface where the
4668 DHCPv6 client is running and enp2s0 as a downstream interface where the prefix is delegated to.
4669 The delegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.
4674 <title>IPv6 Prefix Delegation (DHCPv4
6RD)
</title>
4676 <programlisting># /etc/systemd/network/
55-dhcpv4-
6rd-upstream.network
4683 # When DHCPv4-
6RD is used, the upstream network does not support IPv6.
4684 # Hence, it is not necessary to wait for Router Advertisement, which is enabled by default.
4688 Use6RD=yes
</programlisting>
4690 <programlisting># /etc/systemd/network/
55-dhcpv4-
6rd-downstream.network
4695 DHCPPrefixDelegation=yes
4698 # It is expected that the host is acting as a router. So, usually it is not
4699 # necessary to receive Router Advertisement from other hosts in the downstream network.
4702 [DHCPPrefixDelegation]
4703 UplinkInterface=enp1s0
4705 Announce=yes
</programlisting>
4707 <para>This will enable DHCPv4-
6RD on the interface enp1s0 as an upstream interface where the
4708 DHCPv4 client is running and enp2s0 as a downstream interface where the prefix is delegated to.
4709 The delegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.
4714 <title>A bridge with two enslaved links
</title>
4716 <programlisting># /etc/systemd/network/
25-bridge-static.netdev
4719 Kind=bridge
</programlisting>
4721 <programlisting># /etc/systemd/network/
25-bridge-static.network
4726 Address=
192.168.0.15/
24
4728 DNS=
192.168.0.1</programlisting>
4730 <programlisting># /etc/systemd/network/
25-bridge-slave-interface-
1.network
4735 Bridge=bridge0
</programlisting>
4737 <programlisting># /etc/systemd/network/
25-bridge-slave-interface-
2.network
4742 Bridge=bridge0
</programlisting>
4744 <para>This creates a bridge and attaches devices
<literal>enp2s0
</literal> and
4745 <literal>wlp3s0
</literal> to it. The bridge will have the specified static address
4746 and network assigned, and a default route via the specified gateway will be
4747 added. The specified DNS server will be added to the global list of DNS resolvers.
4752 <title>Bridge port with VLAN forwarding
</title>
4755 # /etc/systemd/network/
25-bridge-slave-interface-
1.network
4771 EgressUntagged=
300-
400</programlisting>
4773 <para>This overrides the configuration specified in the previous example for the
4774 interface
<literal>enp2s0
</literal>, and enables VLAN on that bridge port. VLAN IDs
4775 1-
32,
42,
100-
400 will be allowed. Packets tagged with VLAN IDs
42,
300-
400 will be
4776 untagged when they leave on this interface. Untagged packets which arrive on this
4777 interface will be assigned VLAN ID
42.
</para>
4781 <title>Various tunnels
</title>
4783 <programlisting>/etc/systemd/network/
25-tunnels.network
4794 <programlisting>/etc/systemd/network/
25-tunnel-ipip.netdev
4800 <programlisting>/etc/systemd/network/
25-tunnel-sit.netdev
4806 <programlisting>/etc/systemd/network/
25-tunnel-gre.netdev
4812 <programlisting>/etc/systemd/network/
25-tunnel-vti.netdev
4818 <para>This will bring interface
<literal>ens1
</literal> up and create an IPIP tunnel,
4819 a SIT tunnel, a GRE tunnel, and a VTI tunnel using it.
</para>
4823 <title>A bond device
</title>
4825 <programlisting># /etc/systemd/network/
30-bond1.network
4833 <programlisting># /etc/systemd/network/
30-bond1.netdev
4839 <programlisting># /etc/systemd/network/
30-bond1-dev1.network
4841 MACAddress=
52:
54:
00:e9:
64:
41
4847 <programlisting># /etc/systemd/network/
30-bond1-dev2.network
4849 MACAddress=
52:
54:
00:e9:
64:
42
4855 <para>This will create a bond device
<literal>bond1
</literal> and enslave the two
4856 devices with MAC addresses
52:
54:
00:e9:
64:
41 and
52:
54:
00:e9:
64:
42 to it. IPv6 DHCP
4857 will be used to acquire an address.
</para>
4861 <title>Virtual Routing and Forwarding (VRF)
</title>
4862 <para>Add the
<literal>bond1
</literal> interface to the VRF master interface
4863 <literal>vrf1
</literal>. This will redirect routes generated on this interface to be
4864 within the routing table defined during VRF creation. For kernels before
4.8 traffic
4865 won't be redirected towards the VRFs routing table unless specific ip-rules are added.
4867 <programlisting># /etc/systemd/network/
25-vrf.network
4877 <title>MacVTap
</title>
4878 <para>This brings up a network interface
<literal>macvtap-test
</literal>
4879 and attaches it to
<literal>enp0s25
</literal>.
</para>
4880 <programlisting># /usr/lib/systemd/network/
25-macvtap.network
4885 MACVTAP=macvtap-test
4890 <title>A Xfrm interface with physical underlying device.
</title>
4892 <programlisting># /etc/systemd/network/
27-xfrm.netdev
4898 InterfaceId=
7</programlisting>
4900 <programlisting># /etc/systemd/network/
27-eth0.network
4905 Xfrm=xfrm0
</programlisting>
4907 <para>This creates a
<literal>xfrm0
</literal> interface and binds it to the
<literal>eth0
</literal> device.
4908 This allows hardware based ipsec offloading to the
<literal>eth0
</literal> nic.
4909 If offloading is not needed, xfrm interfaces can be assigned to the
<literal>lo
</literal> device.
4915 <title>See Also
</title>
4917 <citerefentry><refentrytitle>systemd
</refentrytitle><manvolnum>1</manvolnum></citerefentry>,
4918 <citerefentry><refentrytitle>systemd-networkd.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
4919 <citerefentry><refentrytitle>systemd.link
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
4920 <citerefentry><refentrytitle>systemd.netdev
</refentrytitle><manvolnum>5</manvolnum></citerefentry>,
4921 <citerefentry><refentrytitle>systemd-network-generator.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>,
4922 <citerefentry><refentrytitle>systemd-resolved.service
</refentrytitle><manvolnum>8</manvolnum></citerefentry>