{ "content": [ { "type": "text", "text": "# systemd.network (man)\n\n## NAME\n\nsystemd.network - Network configuration\n\n## SYNOPSIS\n\nnetwork.network\n\n## DESCRIPTION\n\nA plain ini-style text file that encodes network configuration for matching network\ninterfaces, used by systemd-networkd(8). See systemd.syntax(7) for a general description of\nthe syntax.\n\n## Sections\n\n- **NAME**\n- **SYNOPSIS**\n- **DESCRIPTION**\n- **EXAMPLES** (11 subsections)\n- **SEE ALSO**\n- **NOTES**\n\nUse structuredContent.sections for detailed options, examples, and full documentation.\n" } ], "structuredContent": { "command": "systemd.network", "section": "", "mode": "man", "summary": "systemd.network - Network configuration", "synopsis": "network.network", "tldr_summary": null, "tldr_examples": [], "tldr_source": null, "flags": [], "examples": [ "# /etc/systemd/network/50-static.network", "[Match]", "Name=enp2s0", "[Network]", "Address=192.168.0.15/24", "Gateway=192.168.0.1", "This brings interface \"enp2s0\" up with a static address. The specified gateway will be used", "for a default route.", "# /etc/systemd/network/80-dhcp.network", "[Match]", "Name=en*", "[Network]", "DHCP=yes", "This will enable DHCPv4 and DHCPv6 on all interfaces with names starting with \"en\" (i.e.", "ethernet interfaces).", "# /etc/systemd/network/55-dhcpv6-pd-upstream.network", "[Match]", "Name=enp1s0", "[Network]", "DHCP=ipv6", "# The below setting is optional, to also assign an address in the delegated prefix", "# to the upstream interface. If not necessary, then comment out the line below and", "# the [DHCPPrefixDelegation] section.", "DHCPPrefixDelegation=yes", "# If the upstream network provides Router Advertisement with Managed bit set,", "# then comment out the line below and WithoutRA= setting in the [DHCPv6] section.", "IPv6AcceptRA=no", "[DHCPv6]", "WithoutRA=solicit", "[DHCPPrefixDelegation]", "UplinkInterface=:self", "SubnetId=0", "Announce=no", "# /etc/systemd/network/55-dhcpv6-pd-downstream.network", "[Match]", "Name=enp2s0", "[Network]", "DHCPPrefixDelegation=yes", "IPv6SendRA=yes", "# It is expected that the host is acting as a router. So, usually it is not", "# necessary to receive Router Advertisement from other hosts in the downstream network.", "IPv6AcceptRA=no", "[DHCPPrefixDelegation]", "UplinkInterface=enp1s0", "SubnetId=1", "Announce=yes", "This will enable DHCPv6-PD on the interface enp1s0 as an upstream interface where the DHCPv6", "client is running and enp2s0 as a downstream interface where the prefix is delegated to. The", "delegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.", "# /etc/systemd/network/55-dhcpv4-6rd-upstream.network", "[Match]", "Name=enp1s0", "[Network]", "DHCP=ipv4", "# When DHCPv4-6RD is used, the upstream network does not support IPv6.", "# Hence, it is not necessary to wait for Router Advertisement, which is enabled by default.", "IPv6AcceptRA=no", "[DHCPv4]", "Use6RD=yes", "# /etc/systemd/network/55-dhcpv4-6rd-downstream.network", "[Match]", "Name=enp2s0", "[Network]", "DHCPPrefixDelegation=yes", "IPv6SendRA=yes", "# It is expected that the host is acting as a router. So, usually it is not", "# necessary to receive Router Advertisement from other hosts in the downstream network.", "IPv6AcceptRA=no", "[DHCPPrefixDelegation]", "UplinkInterface=enp1s0", "SubnetId=1", "Announce=yes", "This will enable DHCPv4-6RD on the interface enp1s0 as an upstream interface where the DHCPv4", "client is running and enp2s0 as a downstream interface where the prefix is delegated to. The", "delegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.", "# /etc/systemd/network/25-bridge-static.netdev", "[NetDev]", "Name=bridge0", "Kind=bridge", "# /etc/systemd/network/25-bridge-static.network", "[Match]", "Name=bridge0", "[Network]", "Address=192.168.0.15/24", "Gateway=192.168.0.1", "DNS=192.168.0.1", "# /etc/systemd/network/25-bridge-slave-interface-1.network", "[Match]", "Name=enp2s0", "[Network]", "Bridge=bridge0", "# /etc/systemd/network/25-bridge-slave-interface-2.network", "[Match]", "Name=wlp3s0", "[Network]", "Bridge=bridge0", "This creates a bridge and attaches devices \"enp2s0\" and \"wlp3s0\" to it. The bridge will have", "the specified static address and network assigned, and a default route via the specified", "gateway will be added. The specified DNS server will be added to the global list of DNS", "resolvers.", "# /etc/systemd/network/25-bridge-slave-interface-1.network", "[Match]", "Name=enp2s0", "[Network]", "Bridge=bridge0", "[BridgeVLAN]", "VLAN=1-32", "PVID=42", "EgressUntagged=42", "[BridgeVLAN]", "VLAN=100-200", "[BridgeVLAN]", "EgressUntagged=300-400", "This overrides the configuration specified in the previous example for the interface", "\"enp2s0\", and enables VLAN on that bridge port. VLAN IDs 1-32, 42, 100-400 will be allowed.", "Packets tagged with VLAN IDs 42, 300-400 will be untagged when they leave on this interface.", "Untagged packets which arrive on this interface will be assigned VLAN ID 42.", "/etc/systemd/network/25-tunnels.network", "[Match]", "Name=ens1", "[Network]", "Tunnel=ipip-tun", "Tunnel=sit-tun", "Tunnel=gre-tun", "Tunnel=vti-tun", "/etc/systemd/network/25-tunnel-ipip.netdev", "[NetDev]", "Name=ipip-tun", "Kind=ipip", "/etc/systemd/network/25-tunnel-sit.netdev", "[NetDev]", "Name=sit-tun", "Kind=sit", "/etc/systemd/network/25-tunnel-gre.netdev", "[NetDev]", "Name=gre-tun", "Kind=gre", "/etc/systemd/network/25-tunnel-vti.netdev", "[NetDev]", "Name=vti-tun", "Kind=vti", "This will bring interface \"ens1\" up and create an IPIP tunnel, a SIT tunnel, a GRE tunnel,", "and a VTI tunnel using it.", "# /etc/systemd/network/30-bond1.network", "[Match]", "Name=bond1", "[Network]", "DHCP=ipv6", "# /etc/systemd/network/30-bond1.netdev", "[NetDev]", "Name=bond1", "Kind=bond", "# /etc/systemd/network/30-bond1-dev1.network", "[Match]", "MACAddress=52:54:00:e9:64:41", "[Network]", "Bond=bond1", "# /etc/systemd/network/30-bond1-dev2.network", "[Match]", "MACAddress=52:54:00:e9:64:42", "[Network]", "Bond=bond1", "This will create a bond device \"bond1\" and enslave the two devices with MAC addresses", "52:54:00:e9:64:41 and 52:54:00:e9:64:42 to it. IPv6 DHCP will be used to acquire an address.", "Add the \"bond1\" interface to the VRF master interface \"vrf1\". This will redirect routes", "generated on this interface to be within the routing table defined during VRF creation. For", "kernels before 4.8 traffic won't be redirected towards the VRFs routing table unless specific", "ip-rules are added.", "# /etc/systemd/network/25-vrf.network", "[Match]", "Name=bond1", "[Network]", "VRF=vrf1", "This brings up a network interface \"macvtap-test\" and attaches it to \"enp0s25\".", "# /usr/lib/systemd/network/25-macvtap.network", "[Match]", "Name=enp0s25", "[Network]", "MACVTAP=macvtap-test", "# /etc/systemd/network/27-xfrm.netdev", "[NetDev]", "Name=xfrm0", "Kind=xfrm", "[Xfrm]", "InterfaceId=7", "# /etc/systemd/network/27-eth0.network", "[Match]", "Name=eth0", "[Network]", "Xfrm=xfrm0", "This creates a \"xfrm0\" interface and binds it to the \"eth0\" device. This allows hardware", "based ipsec offloading to the \"eth0\" nic. If offloading is not needed, xfrm interfaces can be", "assigned to the \"lo\" device." ], "see_also": [ { "name": "systemd", "section": "1", "url": "https://www.chedong.com/phpMan.php/man/systemd/1/json" }, { "name": "systemd-networkd.service", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/systemd-networkd.service/8/json" }, { "name": "systemd.link", "section": "5", "url": "https://www.chedong.com/phpMan.php/man/systemd.link/5/json" }, { "name": "systemd.netdev", "section": "5", "url": "https://www.chedong.com/phpMan.php/man/systemd.netdev/5/json" }, { "name": "generator.service", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/generator.service/8/json" }, { "name": "systemd-resolved.service", "section": "8", "url": "https://www.chedong.com/phpMan.php/man/systemd-resolved.service/8/json" } ], "section_outline": [ { "name": "NAME", "lines": 2, "subsections": [] }, { "name": "SYNOPSIS", "lines": 2, "subsections": [] }, { "name": "DESCRIPTION", "lines": 4086, "subsections": [] }, { "name": "EXAMPLES", "lines": 1, "subsections": [ { "name": "Example 1. Static network configuration", "lines": 11 }, { "name": "Example 2. DHCP on ethernet links", "lines": 10 }, { "name": "Example 3. IPv6 Prefix Delegation (DHCPv6 PD)", "lines": 45 }, { "name": "Example 4. IPv6 Prefix Delegation (DHCPv4 6RD)", "lines": 35 }, { "name": "Example 5. A bridge with two enslaved links", "lines": 33 }, { "name": "Example 6. Bridge port with VLAN forwarding", "lines": 23 }, { "name": "Example 7. Various tunnels", "lines": 38 }, { "name": "Example 8. A bond device", "lines": 29 }, { "name": "Example 9. Virtual Routing and Forwarding (VRF)", "lines": 12 }, { "name": "Example 10. MacVTap", "lines": 9 }, { "name": "Example 11. A Xfrm interface with physical underlying device.", "lines": 19 } ] }, { "name": "SEE ALSO", "lines": 3, "subsections": [] }, { "name": "NOTES", "lines": 115, "subsections": [] } ], "sections": { "NAME": { "content": "systemd.network - Network configuration\n", "subsections": [] }, "SYNOPSIS": { "content": "network.network\n", "subsections": [] }, "DESCRIPTION": { "content": "A plain ini-style text file that encodes network configuration for matching network\ninterfaces, used by systemd-networkd(8). See systemd.syntax(7) for a general description of\nthe syntax.\n\nThe main network file must have the extension .network; other extensions are ignored.\nNetworks are applied to links whenever the links appear.\n\nThe .network files are read from the files located in the system network directories\n/usr/lib/systemd/network and /usr/local/lib/systemd/network, the volatile runtime network\ndirectory /run/systemd/network and the local administration network directory\n/etc/systemd/network. All configuration files are collectively sorted and processed in\nalphanumeric order, regardless of the directories in which they live. However, files with\nidentical filenames replace each other. It is recommended that each filename is prefixed with\na number smaller than \"70\" (e.g. 10-eth0.network). Otherwise, the default .network files or\nthose generated by systemd-network-generator.service(8) may take precedence over user\nconfigured files. Files in /etc/ have the highest priority, files in /run/ take precedence\nover files with the same name under /usr/. This can be used to override a system-supplied\nconfiguration file with a local file if needed. As a special case, an empty file (file size\n0) or symlink with the same name pointing to /dev/null disables the configuration file\nentirely (it is \"masked\").\n\nAlong with the network file foo.network, a \"drop-in\" directory foo.network.d/ may exist. All\nfiles with the suffix \".conf\" from this directory will be merged in the alphanumeric order\nand parsed after the main file itself has been parsed. This is useful to alter or add\nconfiguration settings, without having to modify the main configuration file. Each drop-in\nfile must have appropriate section headers.\n\nIn addition to /etc/systemd/network, drop-in \".d\" directories can be placed in\n/usr/lib/systemd/network or /run/systemd/network directories. Drop-in files in /etc/ take\nprecedence over those in /run/ which in turn take precedence over those in /usr/lib/. Drop-in\nfiles under any of these directories take precedence over the main network file wherever\nlocated.\n\n[MATCH] SECTION OPTIONS\nThe network file contains a [Match] section, which determines if a given network file may be\napplied to a given interface; and a [Network] section specifying how the interface should be\nconfigured. The first (in alphanumeric order) of the network files that matches a given\ninterface is applied, all later files are ignored, even if they match as well.\n\nNote that any network interfaces that have the IDNETMANAGEDBY= udev property set will\nnever be matched by any .network files – unless the property's value is the string\n\"io.systemd.Network\" – even if the [Match] section would otherwise match. This may be used to\nexclude specific network interfaces from systemd-networkd's management, while keeping the\n[Match] section generic. The IDNETMANAGEDBY= property thus declares intended ownership of\nthe device, and permits ensuring that concurrent network management implementations do not\ncompete for management of specific devices.\n\nA network file is said to match a network interface if all matches specified by the [Match]\nsection are satisfied. When a network file does not contain valid settings in [Match]\nsection, then the file will match all interfaces and systemd-networkd warns about that. Hint:\nto avoid the warning and to make it clear that all interfaces shall be matched, add the\nfollowing:\n\nName=*\n\nThe following keys are accepted:\n\nMACAddress=\nA whitespace-separated list of hardware addresses. The acceptable formats are:\n\ncolon-delimited hexadecimal\nEach field must be one byte. E.g. \"12:34:56:78:90:ab\" or \"AA:BB:CC:DD:EE:FF\".\n\nAdded in version 250.\n\nhyphen-delimited hexadecimal\nEach field must be one byte. E.g. \"12-34-56-78-90-ab\" or \"AA-BB-CC-DD-EE-FF\".\n\nAdded in version 250.\n\ndot-delimited hexadecimal\nEach field must be two bytes. E.g. \"1234.5678.90ab\" or \"AABB.CCDD.EEFF\".\n\nAdded in version 250.\n\nIPv4 address format\nE.g. \"127.0.0.1\" or \"192.168.0.1\".\n\nAdded in version 250.\n\nIPv6 address format\nE.g. \"2001:0db8:85a3::8a2e:0370:7334\" or \"::1\".\n\nAdded in version 250.\n\nThe total length of each MAC address must be 4 (for IPv4 tunnel), 6 (for Ethernet), 16\n(for IPv6 tunnel), or 20 (for InfiniBand). This option may appear more than once, in\nwhich case the lists are merged. If the empty string is assigned to this option, the list\nof hardware addresses defined prior to this is reset. Defaults to unset.\n\nAdded in version 211.\n\nPermanentMACAddress=\nA whitespace-separated list of hardware's permanent addresses. While MACAddress= matches\nthe device's current MAC address, this matches the device's permanent MAC address, which\nmay be different from the current one. Use full colon-, hyphen- or dot-delimited\nhexadecimal, or IPv4 or IPv6 address format. This option may appear more than once, in\nwhich case the lists are merged. If the empty string is assigned to this option, the list\nof hardware addresses defined prior to this is reset. Defaults to unset.\n\nAdded in version 245.\n\nPath=\nA whitespace-separated list of shell-style globs matching the persistent path, as exposed\nby the udev property IDPATH.\n\nAdded in version 211.\n\nDriver=\nA whitespace-separated list of shell-style globs matching the driver currently bound to\nthe device, as exposed by the udev property IDNETDRIVER of its parent device, or if\nthat is not set, the driver as exposed by ethtool -i of the device itself. If the list is\nprefixed with a \"!\", the test is inverted.\n\nAdded in version 211.\n\nType=\nA whitespace-separated list of shell-style globs matching the device type, as exposed by\nnetworkctl list. If the list is prefixed with a \"!\", the test is inverted. Some valid\nvalues are \"ether\", \"loopback\", \"wlan\", \"wwan\". Valid types are named either from the\nudev \"DEVTYPE\" attribute, or \"ARPHRD\" macros in linux/ifarp.h, so this is not\ncomprehensive.\n\nAdded in version 211.\n\nKind=\nA whitespace-separated list of shell-style globs matching the device kind, as exposed by\nnetworkctl status INTERFACE or ip -d link show INTERFACE. If the list is prefixed with a\n\"!\", the test is inverted. Some valid values are \"bond\", \"bridge\", \"gre\", \"tun\", \"veth\".\nValid kinds are given by netlink's \"IFLAINFOKIND\" attribute, so this is not\ncomprehensive.\n\nAdded in version 251.\n\nProperty=\nA whitespace-separated list of udev property names with their values after equals sign\n(\"=\"). If multiple properties are specified, the test results are ANDed. If the list is\nprefixed with a \"!\", the test is inverted. If a value contains white spaces, then please\nquote whole key and value pair. If a value contains quotation, then please escape the\nquotation with \"\\\".\n\nExample: if a .link file has the following:\n\nProperty=IDMODELID=9999 \"IDVENDORFROMDATABASE=vendor name\" \"KEY=with \\\"quotation\\\"\"\n\nthen, the .link file matches only when an interface has all the above three properties.\n\nAdded in version 243.\n\nName=\nA whitespace-separated list of shell-style globs matching the device name, as exposed by\nthe udev property \"INTERFACE\", or device's alternative names. If the list is prefixed\nwith a \"!\", the test is inverted.\n\nAdded in version 211.\n\nWLANInterfaceType=\nA whitespace-separated list of wireless network type. Supported values are \"ad-hoc\",\n\"station\", \"ap\", \"ap-vlan\", \"wds\", \"monitor\", \"mesh-point\", \"p2p-client\", \"p2p-go\",\n\"p2p-device\", \"ocb\", and \"nan\". If the list is prefixed with a \"!\", the test is inverted.\n\nAdded in version 244.\n\nSSID=\nA whitespace-separated list of shell-style globs matching the SSID of the currently\nconnected wireless LAN. If the list is prefixed with a \"!\", the test is inverted.\n\nAdded in version 244.\n\nBSSID=\nA whitespace-separated list of hardware address of the currently connected wireless LAN.\nUse full colon-, hyphen- or dot-delimited hexadecimal. See the example in MACAddress=.\nThis option may appear more than once, in which case the lists are merged. If the empty\nstring is assigned to this option, the list is reset.\n\nAdded in version 244.\n\nHost=\nMatches against the hostname or machine ID of the host. See ConditionHost= in\nsystemd.unit(5) for details. When prefixed with an exclamation mark (\"!\"), the result is\nnegated. If an empty string is assigned, the previously assigned value is cleared.\n\nAdded in version 211.\n\nVirtualization=\nChecks whether the system is executed in a virtualized environment and optionally test\nwhether it is a specific implementation. See ConditionVirtualization= in systemd.unit(5)\nfor details. When prefixed with an exclamation mark (\"!\"), the result is negated. If an\nempty string is assigned, the previously assigned value is cleared.\n\nAdded in version 211.\n\nKernelCommandLine=\nChecks whether a specific kernel command line option is set. See\nConditionKernelCommandLine= in systemd.unit(5) for details. When prefixed with an\nexclamation mark (\"!\"), the result is negated. If an empty string is assigned, the\npreviously assigned value is cleared.\n\nAdded in version 211.\n\nKernelVersion=\nChecks whether the kernel version (as reported by uname -r) matches a certain expression.\nSee ConditionKernelVersion= in systemd.unit(5) for details. When prefixed with an\nexclamation mark (\"!\"), the result is negated. If an empty string is assigned, the\npreviously assigned value is cleared.\n\nAdded in version 237.\n\nCredential=\nChecks whether the specified credential was passed to the systemd-udevd.service service.\nSee \u001b[34mSystem and Service Credentials[1] for details. When prefixed with an exclamation mark\n(\"!\"), the result is negated. If an empty string is assigned, the previously assigned\nvalue is cleared.\n\nAdded in version 252.\n\nArchitecture=\nChecks whether the system is running on a specific architecture. See\nConditionArchitecture= in systemd.unit(5) for details. When prefixed with an exclamation\nmark (\"!\"), the result is negated. If an empty string is assigned, the previously\nassigned value is cleared.\n\nAdded in version 211.\n\nFirmware=\nChecks whether the system is running on a machine with the specified firmware. See\nConditionFirmware= in systemd.unit(5) for details. When prefixed with an exclamation mark\n(\"!\"), the result is negated. If an empty string is assigned, the previously assigned\nvalue is cleared.\n\nAdded in version 249.\n\n[LINK] SECTION OPTIONS\nThe [Link] section accepts the following keys:\n\nMACAddress=\nThe hardware address to set for the device.\n\nAdded in version 218.\n\nMTUBytes=\nThe maximum transmission unit in bytes to set for the device. The usual suffixes K, M, G,\nare supported and are understood to the base of 1024.\n\nNote that if IPv6 is enabled on the interface, and the MTU is chosen below 1280 (the\nminimum MTU for IPv6) it will automatically be increased to this value.\n\nAdded in version 218.\n\nARP=\nTakes a boolean. If set to true, the ARP (low-level Address Resolution Protocol) for this\ninterface is enabled. When unset, the kernel's default will be used.\n\nFor example, disabling ARP is useful when creating multiple MACVLAN or VLAN virtual\ninterfaces atop a single lower-level physical interface, which will then only serve as a\nlink/\"bridge\" device aggregating traffic to the same physical link and not participate in\nthe network otherwise. Defaults to unset.\n\nAdded in version 232.\n\nMulticast=\nTakes a boolean. If set to true, the multicast flag on the device is enabled. Defaults to\nunset.\n\nAdded in version 239.\n\nAllMulticast=\nTakes a boolean. If set to true, the driver retrieves all multicast packets from the\nnetwork. This happens when multicast routing is enabled. Defaults to unset.\n\nAdded in version 239.\n\nPromiscuous=\nTakes a boolean. If set to true, promiscuous mode of the interface is enabled. Defaults\nto unset.\n\nIf this is set to false for the underlying link of a \"passthru\" mode MACVLAN/MACVTAP, the\nvirtual interface will be created with the \"nopromisc\" flag set.\n\nAdded in version 248.\n\nUnmanaged=\nTakes a boolean. When \"yes\", no attempts are made to bring up or configure matching\nlinks, equivalent to when there are no matching network files. Defaults to \"no\".\n\nThis is useful for preventing later matching network files from interfering with certain\ninterfaces that are fully controlled by other applications.\n\nAdded in version 233.\n\nGroup=\nLink groups are similar to port ranges found in managed switches. When network interfaces\nare added to a numbered group, operations on all the interfaces from that group can be\nperformed at once. Takes an unsigned integer in the range 0...2147483647. Defaults to\nunset.\n\nAdded in version 246.\n\nRequiredForOnline=\nTakes a boolean or a minimum operational state and an optional maximum operational state.\nPlease see networkctl(1) for possible operational states. When \"yes\", the network is\ndeemed required when determining whether the system is online (including when running\nsystemd-networkd-wait-online). When \"no\", the network is ignored when determining the\nonline state. When a minimum operational state and an optional maximum operational state\nare set, \"yes\" is implied, and this controls the minimum and maximum operational state\nrequired for the network interface to be considered online.\n\nDefaults to \"yes\" when ActivationPolicy= is not set, or set to \"up\", \"always-up\", or\n\"bound\". Defaults to \"no\" when ActivationPolicy= is set to \"manual\" or \"down\". This is\nforced to \"no\" when ActivationPolicy= is set to \"always-down\".\n\nThe network will be brought up normally (as configured by ActivationPolicy=), but in the\nevent that there is no address being assigned by DHCP or the cable is not plugged in, the\nlink will simply remain offline and be skipped automatically by\nsystemd-networkd-wait-online if \"RequiredForOnline=no\".\n\nAdded in version 236.\n\nRequiredFamilyForOnline=\nTakes an address family. When specified, an IP address in the given family is deemed\nrequired when determining whether the link is online (including when running\nsystemd-networkd-wait-online). Takes one of \"ipv4\", \"ipv6\", \"both\", or \"any\". Defaults to\n\"no\". Note that this option has no effect if \"RequiredForOnline=no\".\n\nAdded in version 249.\n\nActivationPolicy=\nSpecifies the policy for systemd-networkd managing the link administrative state.\nSpecifically, this controls how systemd-networkd changes the network device's \"IFFUP\"\nflag, which is sometimes controlled by system administrators by running e.g., ip link set\ndev eth0 up or ip link set dev eth0 down, and can also be changed with networkctl up eth0\nor networkctl down eth0.\n\nTakes one of \"up\", \"always-up\", \"manual\", \"always-down\", \"down\", or \"bound\". When\n\"manual\", systemd-networkd will not change the link's admin state automatically; the\nsystem administrator must bring the interface up or down manually, as desired. When \"up\"\n(the default) or \"always-up\", or \"down\" or \"always-down\", systemd-networkd will set the\nlink up or down, respectively, when the interface is (re)configured. When \"always-up\" or\n\"always-down\", systemd-networkd will set the link up or down, respectively, any time\nsystemd-networkd detects a change in the administrative state. When BindCarrier= is also\nset, this is automatically set to \"bound\" and any other value is ignored.\n\nWhen the policy is set to \"down\" or \"manual\", the default value of RequiredForOnline= is\n\"no\". When the policy is set to \"always-down\", the value of RequiredForOnline= forced to\n\"no\".\n\nThe administrative state is not the same as the carrier state, so using \"always-up\" does\nnot mean the link will never lose carrier. The link carrier depends on both the\nadministrative state as well as the network device's physical connection. However, to\navoid reconfiguration failures, when using \"always-up\", IgnoreCarrierLoss= is forced to\ntrue.\n\nAdded in version 248.\n\n[SR-IOV] SECTION OPTIONS\nThe [SR-IOV] section accepts the following keys. Specify several [SR-IOV] sections to\nconfigure several SR-IOVs. SR-IOV provides the ability to partition a single physical PCI\nresource into virtual PCI functions which can then be injected into a VM. In the case of\nnetwork VFs, SR-IOV improves north-south network performance (that is, traffic with endpoints\noutside the host machine) by allowing traffic to bypass the host machine’s network stack.\n\nVirtualFunction=\nSpecifies a Virtual Function (VF), lightweight PCIe function designed solely to move data\nin and out. Takes an integer in the range 0...2147483646. This option is compulsory.\n\nAdded in version 251.\n\nVLANId=\nSpecifies VLAN ID of the virtual function. Takes an integer in the range 1...4095.\n\nAdded in version 251.\n\nQualityOfService=\nSpecifies quality of service of the virtual function. Takes an integer in the range\n1...4294967294.\n\nAdded in version 251.\n\nVLANProtocol=\nSpecifies VLAN protocol of the virtual function. Takes \"802.1Q\" or \"802.1ad\".\n\nAdded in version 251.\n\nMACSpoofCheck=\nTakes a boolean. Controls the MAC spoof checking. When unset, the kernel's default will\nbe used.\n\nAdded in version 251.\n\nQueryReceiveSideScaling=\nTakes a boolean. Toggle the ability of querying the receive side scaling (RSS)\nconfiguration of the virtual function (VF). The VF RSS information like RSS hash key may\nbe considered sensitive on some devices where this information is shared between VF and\nthe physical function (PF). When unset, the kernel's default will be used.\n\nAdded in version 251.\n\nTrust=\nTakes a boolean. Allows one to set trust mode of the virtual function (VF). When set, VF\nusers can set a specific feature which may impact security and/or performance. When\nunset, the kernel's default will be used.\n\nAdded in version 251.\n\nLinkState=\nAllows one to set the link state of the virtual function (VF). Takes a boolean or a\nspecial value \"auto\". Setting to \"auto\" means a reflection of the physical function (PF)\nlink state, \"yes\" lets the VF to communicate with other VFs on this host even if the PF\nlink state is down, \"no\" causes the hardware to drop any packets sent by the VF. When\nunset, the kernel's default will be used.\n\nAdded in version 251.\n\nMACAddress=\nSpecifies the MAC address for the virtual function.\n\nAdded in version 251.\n\n[NETWORK] SECTION OPTIONS\nThe [Network] section accepts the following keys:\n\nDescription=\nA description of the device. This is only used for presentation purposes.\n\nAdded in version 211.\n\nDHCP=\nEnables DHCPv4 and/or DHCPv6 client support. Accepts \"yes\", \"no\", \"ipv4\", or \"ipv6\".\nDefaults to \"no\".\n\nNote that DHCPv6 will by default be triggered by Router Advertisements, if reception is\nenabled, regardless of this parameter. By explicitly enabling DHCPv6 support here, the\nDHCPv6 client will be started in the mode specified by the WithoutRA= setting in the\n[DHCPv6] section, regardless of the presence of routers on the link, or what flags the\nrouters pass. See IPv6AcceptRA=.\n\nFurthermore, note that by default the domain name specified through DHCP, on Ubuntu, are\nused for name resolution. See option UseDomains= below.\n\nSee the [DHCPv4] or [DHCPv6] sections below for further configuration options for the\nDHCP client support.\n\nAdded in version 211.\n\nDHCPServer=\nTakes a boolean. If set to \"yes\", DHCPv4 server will be started. Defaults to \"no\".\nFurther settings for the DHCP server may be set in the [DHCPServer] section described\nbelow.\n\nAdded in version 215.\n\nLinkLocalAddressing=\nEnables link-local address autoconfiguration. Accepts yes, no, ipv4, and ipv6. An IPv6\nlink-local address is configured when yes or ipv6. An IPv4 link-local address is\nconfigured when yes or ipv4 and when DHCPv4 autoconfiguration has been unsuccessful for\nsome time. (IPv4 link-local address autoconfiguration will usually happen in parallel\nwith repeated attempts to acquire a DHCPv4 lease).\n\nDefaults to no when KeepMaster= or Bridge= is set or when the specified MACVLAN=/MACVTAP=\nhas Mode=passthru, or ipv6 otherwise.\n\nAdded in version 219.\n\nIPv6LinkLocalAddressGenerationMode=\nSpecifies how IPv6 link-local address is generated. Takes one of \"eui64\", \"none\",\n\"stable-privacy\" and \"random\". When unset, \"stable-privacy\" is used if\nIPv6StableSecretAddress= is specified, and if not, \"eui64\" is used. Note that if\nLinkLocalAddressing= is \"no\" or \"ipv4\", then IPv6LinkLocalAddressGenerationMode= will be\nignored. Also, even if LinkLocalAddressing= is \"yes\" or \"ipv6\", setting\nIPv6LinkLocalAddressGenerationMode=none disables to configure an IPv6 link-local address.\n\nAdded in version 246.\n\nIPv6StableSecretAddress=\nTakes an IPv6 address. The specified address will be used as a stable secret for\ngenerating IPv6 link-local address. If this setting is specified, and\nIPv6LinkLocalAddressGenerationMode= is unset, then\nIPv6LinkLocalAddressGenerationMode=stable-privacy is implied. If this setting is not\nspecified, and \"stable-privacy\" is set to IPv6LinkLocalAddressGenerationMode=, then a\nstable secret address will be generated from the local machine ID and the interface name.\n\nAdded in version 249.\n\nIPv4LLStartAddress=\nSpecifies the first IPv4 link-local address to try. Takes an IPv4 address for example\n169.254.1.2, from the link-local address range: 169.254.0.0/16 except for 169.254.0.0/24\nand 169.254.255.0/24. This setting may be useful if the device should always have the\nsame address as long as there is no address conflict. When unset, a random address will\nbe automatically selected. Defaults to unset.\n\nAdded in version 252.\n\nIPv4LLRoute=\nTakes a boolean. If set to true, sets up the route needed for non-IPv4LL hosts to\ncommunicate with IPv4LL-only hosts. Defaults to false.\n\nAdded in version 216.\n\nDefaultRouteOnDevice=\nTakes a boolean. If set to true, sets up the IPv4 default route bound to the interface.\nDefaults to false. This is useful when creating routes on point-to-point interfaces. This\nis equivalent to e.g. the following,\n\nip route add default dev veth99\n\nor,\n\n[Route]\nGateway=0.0.0.0\n\nCurrently, there are no way to specify e.g., the table for the route configured by this\nsetting. To configure the default route with such an additional property, please use the\nfollowing instead:\n\n[Route]\nGateway=0.0.0.0\nTable=1234\n\nIf you'd like to create an IPv6 default route bound to the interface, please use the\nfollowing:\n\n[Route]\nGateway=::\nTable=1234\n\nAdded in version 243.\n\nLLMNR=\nTakes a boolean or \"resolve\". When true, enables \u001b[34mLink-Local Multicast Name Resolution[2]\non the link. When set to \"resolve\", only resolution is enabled, but not host registration\nand announcement. Defaults to true. This setting is read by systemd-resolved.service(8).\n\nAdded in version 216.\n\nMulticastDNS=\nTakes a boolean or \"resolve\". When true, enables \u001b[34mMulticast DNS[3] support on the link.\nWhen set to \"resolve\", only resolution is enabled, but not host or service registration\nand announcement. Defaults to false. This setting is read by systemd-resolved.service(8).\n\nAdded in version 229.\n\nDNSOverTLS=\nTakes a boolean or \"opportunistic\". When true, enables \u001b[34mDNS-over-TLS[4] support on the\nlink. When set to \"opportunistic\", compatibility with non-DNS-over-TLS servers is\nincreased, by automatically turning off DNS-over-TLS servers in this case. This option\ndefines a per-interface setting for resolved.conf(5)'s global DNSOverTLS= option.\nDefaults to unset, and the global setting will be used. This setting is read by systemd-\nresolved.service(8).\n\nAdded in version 239.\n\nDNSSEC=\nTakes a boolean or \"allow-downgrade\". When true, enables \u001b[34mDNSSEC[5] DNS validation support\non the link. When set to \"allow-downgrade\", compatibility with non-DNSSEC capable\nnetworks is increased, by automatically turning off DNSSEC in this case. This option\ndefines a per-interface setting for resolved.conf(5)'s global DNSSEC= option. Defaults to\nunset, and the global setting will be used. This setting is read by systemd-\nresolved.service(8).\n\nAdded in version 229.\n\nDNSSECNegativeTrustAnchors=\nA space-separated list of DNSSEC negative trust anchor domains. If specified and DNSSEC\nis enabled, look-ups done via the interface's DNS server will be subject to the list of\nnegative trust anchors, and not require authentication for the specified domains, or\nanything below it. Use this to disable DNSSEC authentication for specific private\ndomains, that cannot be proven valid using the Internet DNS hierarchy. Defaults to the\nempty list. This setting is read by systemd-resolved.service(8).\n\nAdded in version 229.\n\nLLDP=\nControls support for Ethernet LLDP packet reception. LLDP is a link-layer protocol\ncommonly implemented on professional routers and bridges which announces which physical\nport a system is connected to, as well as other related data. Accepts a boolean or the\nspecial value \"routers-only\". When true, incoming LLDP packets are accepted and a\ndatabase of all LLDP neighbors maintained. If \"routers-only\" is set only LLDP data of\nvarious types of routers is collected and LLDP data about other types of devices ignored\n(such as stations, telephones and others). If false, LLDP reception is disabled. Defaults\nto \"routers-only\". Use networkctl(1) to query the collected neighbor data. LLDP is only\navailable on Ethernet links. See EmitLLDP= below for enabling LLDP packet emission from\nthe local system.\n\nAdded in version 219.\n\nEmitLLDP=\nControls support for Ethernet LLDP packet emission. Accepts a boolean parameter or the\nspecial values \"nearest-bridge\", \"non-tpmr-bridge\" and \"customer-bridge\". Defaults to\nfalse, which turns off LLDP packet emission. If not false, a short LLDP packet with\ninformation about the local system is sent out in regular intervals on the link. The LLDP\npacket will contain information about the local hostname, the local machine ID (as stored\nin machine-id(5)) and the local interface name, as well as the pretty hostname of the\nsystem (as set in machine-info(5)). LLDP emission is only available on Ethernet links.\nNote that this setting passes data suitable for identification of host to the network and\nshould thus not be enabled on untrusted networks, where such identification data should\nnot be made available. Use this option to permit other systems to identify on which\ninterfaces they are connected to this system. The three special values control\npropagation of the LLDP packets. The \"nearest-bridge\" setting permits propagation only to\nthe nearest connected bridge, \"non-tpmr-bridge\" permits propagation across Two-Port MAC\nRelays, but not any other bridges, and \"customer-bridge\" permits propagation until a\ncustomer bridge is reached. For details about these concepts, see \u001b[34mIEEE 802.1AB-2016[6].\nNote that configuring this setting to true is equivalent to \"nearest-bridge\", the\nrecommended and most restricted level of propagation. See LLDP= above for an option to\nenable LLDP reception.\n\nAdded in version 230.\n\nBindCarrier=\nA link name or a list of link names. When set, controls the behavior of the current link.\nWhen all links in the list are in an operational down state, the current link is brought\ndown. When at least one link has carrier, the current interface is brought up.\n\nThis forces ActivationPolicy= to be set to \"bound\".\n\nAdded in version 220.\n\nAddress=\nA static IPv4 or IPv6 address and its prefix length, separated by a \"/\" character.\nSpecify this key more than once to configure several addresses. The format of the address\nmust be as described in inetpton(3). This is a short-hand for an [Address] section only\ncontaining an Address key (see below). This option may be specified more than once.\n\nIf the specified address is \"0.0.0.0\" (for IPv4) or \"::\" (for IPv6), a new address range\nof the requested size is automatically allocated from a system-wide pool of unused\nranges. Note that the prefix length must be equal or larger than 8 for IPv4, and 64 for\nIPv6. The allocated range is checked against all current network interfaces and all known\nnetwork configuration files to avoid address range conflicts. The default system-wide\npool consists of 192.168.0.0/16, 172.16.0.0/12 and 10.0.0.0/8 for IPv4, and fd00::/8 for\nIPv6. This functionality is useful to manage a large number of dynamically created\nnetwork interfaces with the same network configuration and automatic address range\nassignment.\n\nAdded in version 211.\n\nGateway=\nThe gateway address, which must be in the format described in inetpton(3). This is a\nshort-hand for a [Route] section only containing a Gateway= key. This option may be\nspecified more than once.\n\nAdded in version 211.\n\nDNS=\nA DNS server address, which must be in the format described in inetpton(3). This option\nmay be specified more than once. Each address can optionally take a port number separated\nwith \":\", a network interface name or index separated with \"%\", and a Server Name\nIndication (SNI) separated with \"#\". When IPv6 address is specified with a port number,\nthen the address must be in the square brackets. That is, the acceptable full formats are\n\"111.222.333.444:9953%ifname#example.com\" for IPv4 and\n\"[1111:2222::3333]:9953%ifname#example.com\" for IPv6. If an empty string is assigned,\nthen the all previous assignments are cleared. This setting is read by systemd-\nresolved.service(8).\n\nAdded in version 211.\n\nDomains=\nA whitespace-separated list of domains which should be resolved using the DNS servers on\nthis link. Each item in the list should be a domain name, optionally prefixed with a\ntilde (\"~\"). The domains with the prefix are called \"routing-only domains\". The domains\nwithout the prefix are called \"search domains\" and are first used as search suffixes for\nextending single-label hostnames (hostnames containing no dots) to become fully qualified\ndomain names (FQDNs). If a single-label hostname is resolved on this interface, each of\nthe specified search domains are appended to it in turn, converting it into a fully\nqualified domain name, until one of them may be successfully resolved.\n\nBoth \"search\" and \"routing-only\" domains are used for routing of DNS queries: look-ups\nfor hostnames ending in those domains (hence also single label names, if any \"search\ndomains\" are listed), are routed to the DNS servers configured for this interface. The\ndomain routing logic is particularly useful on multi-homed hosts with DNS servers serving\nparticular private DNS zones on each interface.\n\nThe \"routing-only\" domain \"~.\" (the tilde indicating definition of a routing domain, the\ndot referring to the DNS root domain which is the implied suffix of all valid DNS names)\nhas special effect. It causes all DNS traffic which does not match another configured\ndomain routing entry to be routed to DNS servers specified for this interface. This\nsetting is useful to prefer a certain set of DNS servers if a link on which they are\nconnected is available.\n\nThis setting is read by systemd-resolved.service(8). \"Search domains\" correspond to the\ndomain and search entries in resolv.conf(5). Domain name routing has no equivalent in the\ntraditional glibc API, which has no concept of domain name servers limited to a specific\nlink.\n\nAdded in version 216.\n\nDNSDefaultRoute=\nTakes a boolean argument. If true, this link's configured DNS servers are used for\nresolving domain names that do not match any link's configured Domains= setting. If\nfalse, this link's configured DNS servers are never used for such domains, and are\nexclusively used for resolving names that match at least one of the domains configured on\nthis link. If not specified defaults to an automatic mode: queries not matching any\nlink's configured domains will be routed to this link if it has no routing-only domains\nconfigured.\n\nAdded in version 240.\n\nNTP=\nAn NTP server address (either an IP address, or a hostname). This option may be specified\nmore than once. This setting is read by systemd-timesyncd.service(8).\n\nAdded in version 216.\n\nIPForward=\nConfigures IP packet forwarding for the system. If enabled, incoming packets on any\nnetwork interface will be forwarded to any other interfaces according to the routing\ntable. Takes a boolean, or the values \"ipv4\" or \"ipv6\", which only enable IP packet\nforwarding for the specified address family. This controls the net.ipv4.ipforward and\nnet.ipv6.conf.all.forwarding sysctl options of the network interface (see \u001b[34mIP Sysctl[7]\nfor details about sysctl options). Defaults to \"no\".\n\nNote: this setting controls a global kernel option, and does so one way only: if a\nnetwork that has this setting enabled is set up the global setting is turned on. However,\nit is never turned off again, even after all networks with this setting enabled are shut\ndown again.\n\nTo allow IP packet forwarding only between specific network interfaces use a firewall.\n\nAdded in version 219.\n\nIPMasquerade=\nConfigures IP masquerading for the network interface. If enabled, packets forwarded from\nthe network interface will be appear as coming from the local host. Takes one of \"ipv4\",\n\"ipv6\", \"both\", or \"no\". Defaults to \"no\". If enabled, this automatically sets IPForward=\nto one of \"ipv4\", \"ipv6\" or \"yes\".\n\nNote. Any positive boolean values such as \"yes\" or \"true\" are now deprecated. Please use\none of the values in the above.\n\nAdded in version 219.\n\nIPv6PrivacyExtensions=\nConfigures use of stateless temporary addresses that change over time (see \u001b[34mRFC 4941[8],\nPrivacy Extensions for Stateless Address Autoconfiguration in IPv6). Takes a boolean or\nthe special values \"prefer-public\" and \"kernel\". When true, enables the privacy\nextensions and prefers temporary addresses over public addresses. When \"prefer-public\",\nenables the privacy extensions, but prefers public addresses over temporary addresses.\nWhen false, the privacy extensions remain disabled. When \"kernel\", the kernel's default\nsetting will be left in place. When unspecified, the value specified in the same setting\nin networkd.conf(5), which defaults to \"no\", will be used.\n\nAdded in version 222.\n\nIPv6AcceptRA=\nTakes a boolean. Controls IPv6 Router Advertisement (RA) reception support for the\ninterface. If true, RAs are accepted; if false, RAs are ignored. When RAs are accepted,\nthey may trigger the start of the DHCPv6 client if the relevant flags are set in the RA\ndata, or if no routers are found on the link. The default is to disable RA reception for\nbridge devices or when IP forwarding is enabled, and to enable it otherwise. Cannot be\nenabled on devices aggregated in a bond device or when link-local addressing is disabled.\n\nFurther settings for the IPv6 RA support may be configured in the [IPv6AcceptRA] section,\nsee below.\n\nAlso see \u001b[34mIP Sysctl[7] in the kernel documentation regarding \"acceptra\", but note that\nsystemd's setting of 1 (i.e. true) corresponds to kernel's setting of 2.\n\nNote that kernel's implementation of the IPv6 RA protocol is always disabled, regardless\nof this setting. If this option is enabled, a userspace implementation of the IPv6 RA\nprotocol is used, and the kernel's own implementation remains disabled, since\nsystemd-networkd needs to know all details supplied in the advertisements, and these are\nnot available from the kernel if the kernel's own implementation is used.\n\nAdded in version 231.\n\nIPv6DuplicateAddressDetection=\nConfigures the amount of IPv6 Duplicate Address Detection (DAD) probes to send. When\nunset, the kernel's default will be used.\n\nAdded in version 228.\n\nIPv6HopLimit=\nConfigures IPv6 Hop Limit. Takes an integer in the range 1...255. For each router that\nforwards the packet, the hop limit is decremented by 1. When the hop limit field reaches\nzero, the packet is discarded. When unset, the kernel's default will be used.\n\nAdded in version 228.\n\nIPv4ReversePathFilter=\nConfigure IPv4 Reverse Path Filtering. If enabled, when an IPv4 packet is received, the\nmachine will first check whether the source of the packet would be routed through the\ninterface it came in. If there is no route to the source on that interface, the machine\nwill drop the packet. Takes one of \"no\", \"strict\", or \"loose\". When \"no\", no source\nvalidation will be done. When \"strict\", mode each incoming packet is tested against the\nFIB and if the incoming interface is not the best reverse path, the packet check will\nfail. By default failed packets are discarded. When \"loose\", mode each incoming packet's\nsource address is tested against the FIB. The packet is dropped only if the source\naddress is not reachable via any interface on that router. See \u001b[34mRFC 3704[9]. When unset,\nthe kernel's default will be used.\n\nAdded in version 255.\n\nIPv4AcceptLocal=\nTakes a boolean. Accept packets with local source addresses. In combination with suitable\nrouting, this can be used to direct packets between two local interfaces over the wire\nand have them accepted properly. When unset, the kernel's default will be used.\n\nAdded in version 246.\n\nIPv4RouteLocalnet=\nTakes a boolean. When true, the kernel does not consider loopback addresses as martian\nsource or destination while routing. This enables the use of 127.0.0.0/8 for local\nrouting purposes. When unset, the kernel's default will be used.\n\nAdded in version 248.\n\nIPv4ProxyARP=\nTakes a boolean. Configures proxy ARP for IPv4. Proxy ARP is the technique in which one\nhost, usually a router, answers ARP requests intended for another machine. By \"faking\"\nits identity, the router accepts responsibility for routing packets to the \"real\"\ndestination. See \u001b[34mRFC 1027[9]. When unset, the kernel's default will be used.\n\nAdded in version 233.\n\nIPv6ProxyNDP=\nTakes a boolean. Configures proxy NDP for IPv6. Proxy NDP (Neighbor Discovery Protocol)\nis a technique for IPv6 to allow routing of addresses to a different destination when\npeers expect them to be present on a certain physical link. In this case a router answers\nNeighbour Advertisement messages intended for another machine by offering its own MAC\naddress as destination. Unlike proxy ARP for IPv4, it is not enabled globally, but will\nonly send Neighbour Advertisement messages for addresses in the IPv6 neighbor proxy\ntable, which can also be shown by ip -6 neighbour show proxy. systemd-networkd will\ncontrol the per-interface `proxyndp` switch for each configured interface depending on\nthis option. When unset, the kernel's default will be used.\n\nAdded in version 234.\n\nIPv6ProxyNDPAddress=\nAn IPv6 address, for which Neighbour Advertisement messages will be proxied. This option\nmay be specified more than once. systemd-networkd will add the IPv6ProxyNDPAddress=\nentries to the kernel's IPv6 neighbor proxy table. This setting implies IPv6ProxyNDP=yes\nbut has no effect if IPv6ProxyNDP= has been set to false. When unset, the kernel's\ndefault will be used.\n\nAdded in version 233.\n\nIPv6SendRA=\nWhether to enable or disable Router Advertisement sending on a link. Takes a boolean\nvalue. When enabled, prefixes configured in [IPv6Prefix] sections and routes configured\nin the [IPv6RoutePrefix] sections are distributed as defined in the [IPv6SendRA] section.\nIf DHCPPrefixDelegation= is enabled, then the delegated prefixes are also distributed.\nSee DHCPPrefixDelegation= setting and the [IPv6SendRA], [IPv6Prefix], [IPv6RoutePrefix],\nand [DHCPPrefixDelegation] sections for more configuration options.\n\nAdded in version 247.\n\nDHCPPrefixDelegation=\nTakes a boolean value. When enabled, requests subnet prefixes on another link via the\nDHCPv6 protocol or via the 6RD option in the DHCPv4 protocol. An address within each\ndelegated prefix will be assigned, and the prefixes will be announced through IPv6 Router\nAdvertisement if IPv6SendRA= is enabled. This behaviour can be configured in the\n[DHCPPrefixDelegation] section. Defaults to disabled.\n\nAdded in version 250.\n\nIPv6MTUBytes=\nConfigures IPv6 maximum transmission unit (MTU). An integer greater than or equal to 1280\nbytes. When unset, the kernel's default will be used.\n\nAdded in version 239.\n\nKeepMaster=\nTakes a boolean value. When enabled, the current master interface index will not be\nchanged, and BatmanAdvanced=, Bond=, Bridge=, and VRF= settings are ignored. This may be\nuseful when a netdev with a master interface is created by another program, e.g.\nsystemd-nspawn(1). Defaults to false.\n\nAdded in version 250.\n\nBatmanAdvanced=, Bond=, Bridge=, VRF=\nThe name of the B.A.T.M.A.N. Advanced, bond, bridge, or VRF interface to add the link to.\nSee systemd.netdev(5).\n\nAdded in version 211.\n\nIPoIB=, IPVLAN=, IPVTAP=, MACsec=, MACVLAN=, MACVTAP=, Tunnel=, VLAN=, VXLAN=, Xfrm=\nThe name of an IPoIB, IPVLAN, IPVTAP, MACsec, MACVLAN, MACVTAP, tunnel, VLAN, VXLAN, or\nXfrm to be created on the link. See systemd.netdev(5). This option may be specified more\nthan once.\n\nAdded in version 211.\n\nActiveSlave=\nTakes a boolean. Specifies the new active slave. The \"ActiveSlave=\" option is only valid\nfor following modes: \"active-backup\", \"balance-alb\", and \"balance-tlb\". Defaults to\nfalse.\n\nAdded in version 235.\n\nPrimarySlave=\nTakes a boolean. Specifies which slave is the primary device. The specified device will\nalways be the active slave while it is available. Only when the primary is off-line will\nalternate devices be used. This is useful when one slave is preferred over another, e.g.\nwhen one slave has higher throughput than another. The \"PrimarySlave=\" option is only\nvalid for following modes: \"active-backup\", \"balance-alb\", and \"balance-tlb\". Defaults to\nfalse.\n\nAdded in version 235.\n\nConfigureWithoutCarrier=\nTakes a boolean. Allows networkd to configure a specific link even if it has no carrier.\nDefaults to false. If enabled, and the IgnoreCarrierLoss= setting is not explicitly set,\nthen it is enabled as well.\n\nAdded in version 235.\n\nIgnoreCarrierLoss=\nTakes a boolean or a timespan. When true, systemd-networkd retains both the static and\ndynamic configuration of the interface even if its carrier is lost. When false,\nsystemd-networkd drops both the static and dynamic configuration of the interface. When a\ntimespan is specified, systemd-networkd waits for the specified timespan, and ignores the\ncarrier loss if the link regain its carrier within the timespan. Setting 0 seconds is\nequivalent to \"no\", and \"infinite\" is equivalent to \"yes\".\n\nSetting a finite timespan may be useful when e.g. in the following cases:\n\n• A wireless interface connecting to a network which has multiple access points with\nthe same SSID.\n\n• Enslaving a wireless interface to a bond interface, which may disconnect from the\nconnected access point and causes its carrier to be lost.\n\n• The driver of the interface resets when the MTU is changed.\n\nWhen Bond= is specified to a wireless interface, defaults to 3 seconds. When the DHCPv4\nclient is enabled and UseMTU= in the [DHCPv4] section enabled, defaults to 5 seconds.\nOtherwise, defaults to the value specified with ConfigureWithoutCarrier=. When\nActivationPolicy= is set to \"always-up\", this is forced to \"yes\", and ignored any user\nspecified values.\n\nAdded in version 242.\n\nKeepConfiguration=\nTakes a boolean or one of \"static\", \"dhcp-on-stop\", \"dhcp\". When \"static\",\nsystemd-networkd will not drop static addresses and routes on starting up process. When\nset to \"dhcp-on-stop\", systemd-networkd will not drop addresses and routes on stopping\nthe daemon. When \"dhcp\", the addresses and routes provided by a DHCP server will never be\ndropped even if the DHCP lease expires. This is contrary to the DHCP specification, but\nmay be the best choice if, e.g., the root filesystem relies on this connection. The\nsetting \"dhcp\" implies \"dhcp-on-stop\", and \"yes\" implies \"dhcp\" and \"static\". Defaults to\n\"dhcp-on-stop\" when systemd-networkd is running in initrd, \"yes\" when the root filesystem\nis a network filesystem, and \"no\" otherwise.\n\nAdded in version 243.\n\n[ADDRESS] SECTION OPTIONS\nAn [Address] section accepts the following keys. Specify several [Address] sections to\nconfigure several addresses.\n\nAddress=\nAs in the [Network] section. This setting is mandatory. Each [Address] section can\ncontain one Address= setting.\n\nAdded in version 211.\n\nPeer=\nThe peer address in a point-to-point connection. Accepts the same format as the Address=\nsetting.\n\nAdded in version 216.\n\nBroadcast=\nTakes an IPv4 address or boolean value. The address must be in the format described in\ninetpton(3). If set to true, then the IPv4 broadcast address will be derived from the\nAddress= setting. If set to false, then the broadcast address will not be set. Defaults\nto true, except for wireguard interfaces, where it default to false.\n\nAdded in version 211.\n\nLabel=\nSpecifies the label for the IPv4 address. The label must be a 7-bit ASCII string with a\nlength of 1...15 characters. Defaults to unset.\n\nAdded in version 211.\n\nPreferredLifetime=\nAllows the default \"preferred lifetime\" of the address to be overridden. Only three\nsettings are accepted: \"forever\", \"infinity\", which is the default and means that the\naddress never expires, and \"0\", which means that the address is considered immediately\n\"expired\" and will not be used, unless explicitly requested. A setting of\nPreferredLifetime=0 is useful for addresses which are added to be used only by a specific\napplication, which is then configured to use them explicitly.\n\nAdded in version 230.\n\nScope=\nThe scope of the address, which can be \"global\" (valid everywhere on the network, even\nthrough a gateway), \"link\" (only valid on this device, will not traverse a gateway) or\n\"host\" (only valid within the device itself, e.g. 127.0.0.1) or an integer in the range\n0...255. Defaults to \"global\". IPv4 only - IPv6 scope is automatically assigned by the\nkernel and cannot be set manually.\n\nAdded in version 235.\n\nRouteMetric=\nThe metric of the prefix route, which is pointing to the subnet of the configured IP\naddress, taking the configured prefix length into account. Takes an unsigned integer in\nthe range 0...4294967295. When unset or set to 0, the kernel's default value is used.\nThis setting will be ignored when AddPrefixRoute= is false.\n\nAdded in version 246.\n\nHomeAddress=\nTakes a boolean. Designates this address the \"home address\" as defined in \u001b[34mRFC 6275[10].\nSupported only on IPv6. Defaults to false.\n\nAdded in version 232.\n\nDuplicateAddressDetection=\nTakes one of \"ipv4\", \"ipv6\", \"both\", or \"none\". When \"ipv4\", performs IPv4 Address\nConflict Detection. See \u001b[34mRFC 5227[11]. When \"ipv6\", performs IPv6 Duplicate Address\nDetection. See \u001b[34mRFC 4862[12]. Defaults to \"ipv4\" for IPv4 link-local addresses, \"ipv6\" for\nIPv6 addresses, and \"none\" otherwise.\n\nAdded in version 232.\n\nManageTemporaryAddress=\nTakes a boolean. If true the kernel manage temporary addresses created from this one as\ntemplate on behalf of Privacy Extensions \u001b[34mRFC 3041[13]. For this to become active, the\nusetempaddr sysctl setting has to be set to a value greater than zero. The given address\nneeds to have a prefix length of 64. This flag allows using privacy extensions in a\nmanually configured network, just like if stateless auto-configuration was active.\nDefaults to false.\n\nAdded in version 232.\n\nAddPrefixRoute=\nTakes a boolean. When true, the prefix route for the address is automatically added.\nDefaults to true.\n\nAdded in version 245.\n\nAutoJoin=\nTakes a boolean. Joining multicast group on ethernet level via ip maddr command would not\nwork if we have an Ethernet switch that does IGMP snooping since the switch would not\nreplicate multicast packets on ports that did not have IGMP reports for the multicast\naddresses. Linux vxlan interfaces created via ip link add vxlan or networkd's netdev kind\nvxlan have the group option that enables them to do the required join. By extending ip\naddress command with option \"autojoin\" we can get similar functionality for openvswitch\n(OVS) vxlan interfaces as well as other tunneling mechanisms that need to receive\nmulticast traffic. Defaults to \"no\".\n\nAdded in version 232.\n\nNetLabel=label\nThis setting provides a method for integrating static and dynamic network configuration\ninto Linux \u001b[34mNetLabel[14] subsystem rules, used by \u001b[34mLinux Security Modules (LSMs)[15] for\nnetwork access control. The label, with suitable LSM rules, can be used to control\nconnectivity of (for example) a service with peers in the local network. At least with\nSELinux, only the ingress can be controlled but not egress. The benefit of using this\nsetting is that it may be possible to apply interface independent part of NetLabel\nconfiguration at very early stage of system boot sequence, at the time when the network\ninterfaces are not available yet, with netlabelctl(8), and the per-interface\nconfiguration with systemd-networkd once the interfaces appear later. Currently this\nfeature is only implemented for SELinux.\n\nThe option expects a single NetLabel label. The label must conform to lexical\nrestrictions of LSM labels. When an interface is configured with IP addresses, the\naddresses and subnetwork masks will be appended to the \u001b[34mNetLabel Fallback Peer\n\u001b[34mLabeling[16] rules. They will be removed when the interface is deconfigured. Failures to\nmanage the labels will be ignored.\n\nWarning: Once labeling is enabled for network traffic, a lot of LSM access control points\nin Linux networking stack go from dormant to active. Care should be taken to avoid\ngetting into a situation where for example remote connectivity is broken, when the\nsecurity policy hasn't been updated to consider LSM per-packet access controls and no\nrules would allow any network traffic. Also note that additional configuration with\nnetlabelctl(8) is needed.\n\nExample:\n\n[Address]\nNetLabel=systemu:objectr:localnetpeert:s0\n\nWith the example rules applying for interface \"eth0\", when the interface is configured\nwith an IPv4 address of 10.0.0.123/8, systemd-networkd performs the equivalent of\nnetlabelctl operation\n\nnetlabelctl unlbl add interface eth0 address:10.0.0.0/8 label:systemu:objectr:localnetpeert:s0\n\nand the reverse operation when the IPv4 address is deconfigured. The configuration can be\nused with LSM rules; in case of SELinux to allow a SELinux domain to receive data from\nobjects of SELinux \"peer\" class. For example:\n\ntype localnetpeert;\nallow myservert localnetpeert:peer recv;\n\nThe effect of the above configuration and rules (in absence of other rules as may be the\ncase) is to only allow \"myservert\" (and nothing else) to receive data from local subnet\n10.0.0.0/8 of interface \"eth0\".\n\nAdded in version 252.\n\nNFTSet=source:family:table:set\nThis setting provides a method for integrating network configuration into firewall rules\nwith \u001b[34mNFT[17] sets. The benefit of using the setting is that static network configuration\n(or dynamically obtained network addresses, see similar directives in other sections) can\nbe used in firewall rules with the indirection of NFT set types. For example, access\ncould be granted for hosts in the local subnetwork only. Firewall rules using IP address\nof an interface are also instantly updated when the network configuration changes, for\nexample via DHCP.\n\nThis option expects a whitespace separated list of NFT set definitions. Each definition\nconsists of a colon-separated tuple of source type (one of \"address\", \"prefix\" or\n\"ifindex\"), NFT address family (one of \"arp\", \"bridge\", \"inet\", \"ip\", \"ip6\", or\n\"netdev\"), table name and set name. The names of tables and sets must conform to lexical\nrestrictions of NFT table names. The type of the element used in the NFT filter must\nmatch the type implied by the directive (\"address\", \"prefix\" or \"ifindex\") and address\ntype (IPv4 or IPv6) as shown in the table below.\n\nTable 1. Defined source type values\n┌─────────────┬─────────────────┬──────────────────────────┐\n│ Source type │ Description │ Corresponding NFT type │\n│ │ │ name │\n├─────────────┼─────────────────┼──────────────────────────┤\n│ \"address\" │ host IP address │ \"ipv4addr\" or │\n│ │ │ \"ipv6addr\" │\n├─────────────┼─────────────────┼──────────────────────────┤\n│ \"prefix\" │ network prefix │ \"ipv4addr\" or │\n│ │ │ \"ipv6addr\", with \"flags │\n│ │ │ interval\" │\n├─────────────┼─────────────────┼──────────────────────────┤\n│ \"ifindex\" │ interface index │ \"ifaceindex\" │\n└─────────────┴─────────────────┴──────────────────────────┘\n\nWhen an interface is configured with IP addresses, the addresses, subnetwork masks or\ninterface index will be appended to the NFT sets. The information will be removed when\nthe interface is deconfigured. systemd-networkd only inserts elements to (or removes\nfrom) the sets, so the related NFT rules, tables and sets must be prepared elsewhere in\nadvance. Failures to manage the sets will be ignored.\n\nExample:\n\n[Address]\nNFTSet=prefix:netdev:filter:ethipv4prefix\n\nCorresponding NFT rules:\n\ntable netdev filter {\nset ethipv4prefix {\ntype ipv4addr\nflags interval\n}\nchain ethingress {\ntype filter hook ingress device \"eth0\" priority filter; policy drop;\nip daddr != @ethipv4prefix drop\naccept\n}\n}\n\nAdded in version 255.\n\n[NEIGHBOR] SECTION OPTIONS\nA [Neighbor] section accepts the following keys. The neighbor section adds a permanent,\nstatic entry to the neighbor table (IPv6) or ARP table (IPv4) for the given hardware address\non the links matched for the network. Specify several [Neighbor] sections to configure\nseveral static neighbors.\n\nAddress=\nThe IP address of the neighbor.\n\nAdded in version 240.\n\nLinkLayerAddress=\nThe link layer address (MAC address or IP address) of the neighbor.\n\nAdded in version 243.\n\n[IPV6ADDRESSLABEL] SECTION OPTIONS\nAn [IPv6AddressLabel] section accepts the following keys. Specify several [IPv6AddressLabel]\nsections to configure several address labels. IPv6 address labels are used for address\nselection. See \u001b[34mRFC 3484[18]. Precedence is managed by userspace, and only the label itself is\nstored in the kernel.\n\nLabel=\nThe label for the prefix, an unsigned integer in the range 0...4294967294. 0xffffffff is\nreserved. This setting is mandatory.\n\nAdded in version 234.\n\nPrefix=\nIPv6 prefix is an address with a prefix length, separated by a slash \"/\" character. This\nsetting is mandatory.\n\nAdded in version 234.\n\n[ROUTINGPOLICYRULE] SECTION OPTIONS\nAn [RoutingPolicyRule] section accepts the following settings. Specify several\n[RoutingPolicyRule] sections to configure several rules.\n\nTypeOfService=\nThis specifies the Type of Service (ToS) field of packets to match; it takes an unsigned\ninteger in the range 0...255. The field can be used to specify precedence (the first 3\nbits) and ToS (the next 3 bits). The field can be also used to specify Differentiated\nServices Code Point (DSCP) (the first 6 bits) and Explicit Congestion Notification (ECN)\n(the last 2 bits). See \u001b[34mType of Service[19] and \u001b[34mDifferentiated services[20] for more\ndetails.\n\nAdded in version 235.\n\nFrom=\nSpecifies the source address prefix to match. Possibly followed by a slash and the prefix\nlength.\n\nAdded in version 235.\n\nTo=\nSpecifies the destination address prefix to match. Possibly followed by a slash and the\nprefix length.\n\nAdded in version 235.\n\nFirewallMark=\nSpecifies the iptables firewall mark value to match (a number in the range\n1...4294967295). Optionally, the firewall mask (also a number between 1...4294967295) can\nbe suffixed with a slash (\"/\"), e.g., \"7/255\".\n\nAdded in version 235.\n\nTable=\nSpecifies the routing table identifier to look up if the rule selector matches. Takes one\nof predefined names \"default\", \"main\", and \"local\", and names defined in RouteTable= in\nnetworkd.conf(5), or a number between 1 and 4294967295. Defaults to \"main\".\n\nAdded in version 235.\n\nPriority=\nSpecifies the priority of this rule. Priority= is an integer in the range\n0...4294967295. Higher number means lower priority, and rules get processed in order of\nincreasing number. Defaults to unset, and the kernel will pick a value dynamically.\n\nAdded in version 235.\n\nIncomingInterface=\nSpecifies incoming device to match. If the interface is loopback, the rule only matches\npackets originating from this host.\n\nAdded in version 236.\n\nOutgoingInterface=\nSpecifies the outgoing device to match. The outgoing interface is only available for\npackets originating from local sockets that are bound to a device.\n\nAdded in version 236.\n\nSourcePort=\nSpecifies the source IP port or IP port range match in forwarding information base (FIB)\nrules. A port range is specified by the lower and upper port separated by a dash.\nDefaults to unset.\n\nAdded in version 240.\n\nDestinationPort=\nSpecifies the destination IP port or IP port range match in forwarding information base\n(FIB) rules. A port range is specified by the lower and upper port separated by a dash.\nDefaults to unset.\n\nAdded in version 240.\n\nIPProtocol=\nSpecifies the IP protocol to match in forwarding information base (FIB) rules. Takes IP\nprotocol name such as \"tcp\", \"udp\" or \"sctp\", or IP protocol number such as \"6\" for \"tcp\"\nor \"17\" for \"udp\". Defaults to unset.\n\nAdded in version 240.\n\nInvertRule=\nA boolean. Specifies whether the rule is to be inverted. Defaults to false.\n\nAdded in version 240.\n\nFamily=\nTakes a special value \"ipv4\", \"ipv6\", or \"both\". By default, the address family is\ndetermined by the address specified in To= or From=. If neither To= nor From= are\nspecified, then defaults to \"ipv4\".\n\nAdded in version 243.\n\nUser=\nTakes a username, a user ID, or a range of user IDs separated by a dash. Defaults to\nunset.\n\nAdded in version 245.\n\nSuppressPrefixLength=\nTakes a number N in the range 0...128 and rejects routing decisions that have a prefix\nlength of N or less. Defaults to unset.\n\nAdded in version 245.\n\nSuppressInterfaceGroup=\nTakes an integer in the range 0...2147483647 and rejects routing decisions that have an\ninterface with the same group id. It has the same meaning as suppressifgroup in ip rule.\nDefaults to unset.\n\nAdded in version 250.\n\nType=\nSpecifies Routing Policy Database (RPDB) rule type. Takes one of \"blackhole\",\n\"unreachable\" or \"prohibit\".\n\nAdded in version 248.\n\n[NEXTHOP] SECTION OPTIONS\nThe [NextHop] section is used to manipulate entries in the kernel's \"nexthop\" tables. The\n[NextHop] section accepts the following settings. Specify several [NextHop] sections to\nconfigure several hops.\n\nId=\nThe id of the next hop. Takes an integer in the range 1...4294967295. If unspecified,\nthen automatically chosen by kernel.\n\nAdded in version 244.\n\nGateway=\nAs in the [Network] section.\n\nAdded in version 244.\n\nFamily=\nTakes one of the special values \"ipv4\" or \"ipv6\". By default, the family is determined by\nthe address specified in Gateway=. If Gateway= is not specified, then defaults to \"ipv4\".\n\nAdded in version 248.\n\nOnLink=\nTakes a boolean. If set to true, the kernel does not have to check if the gateway is\nreachable directly by the current machine (i.e., attached to the local network), so that\nwe can insert the nexthop in the kernel table without it being complained about. Defaults\nto \"no\".\n\nAdded in version 248.\n\nBlackhole=\nTakes a boolean. If enabled, packets to the corresponding routes are discarded silently,\nand Gateway= cannot be specified. Defaults to \"no\".\n\nAdded in version 248.\n\nGroup=\nTakes a whitespace separated list of nexthop IDs. Each ID must be in the range\n1...4294967295. Optionally, each nexthop ID can take a weight after a colon\n(\"id[:weight]\"). The weight must be in the range 1...255. If the weight is not specified,\nthen it is assumed that the weight is 1. This setting cannot be specified with Gateway=,\nFamily=, Blackhole=. This setting can be specified multiple times. If an empty string is\nassigned, then the all previous assignments are cleared. Defaults to unset.\n\nAdded in version 249.\n\n[ROUTE] SECTION OPTIONS\nThe [Route] section accepts the following settings. Specify several [Route] sections to\nconfigure several routes.\n\nGateway=\nTakes the gateway address or the special values \"dhcp4\" and \"ipv6ra\". If \"dhcp4\" or\n\"ipv6ra\" is set, then the gateway address provided by DHCPv4 or IPv6 RA is used.\n\nAdded in version 211.\n\nGatewayOnLink=\nTakes a boolean. If set to true, the kernel does not have to check if the gateway is\nreachable directly by the current machine (i.e., attached to the local network), so that\nwe can insert the route in the kernel table without it being complained about. Defaults\nto \"no\".\n\nAdded in version 234.\n\nDestination=\nThe destination prefix of the route. Possibly followed by a slash and the prefix length.\nIf omitted, a full-length host route is assumed.\n\nAdded in version 211.\n\nSource=\nThe source prefix of the route. Possibly followed by a slash and the prefix length. If\nomitted, a full-length host route is assumed.\n\nAdded in version 218.\n\nMetric=\nThe metric of the route. Takes an unsigned integer in the range 0...4294967295. Defaults\nto unset, and the kernel's default will be used.\n\nAdded in version 216.\n\nIPv6Preference=\nSpecifies the route preference as defined in \u001b[34mRFC 4191[21] for Router Discovery messages.\nWhich can be one of \"low\" the route has a lowest priority, \"medium\" the route has a\ndefault priority or \"high\" the route has a highest priority.\n\nAdded in version 234.\n\nScope=\nThe scope of the IPv4 route, which can be \"global\", \"site\", \"link\", \"host\", or \"nowhere\":\n\n• \"global\" means the route can reach hosts more than one hop away.\n\n• \"site\" means an interior route in the local autonomous system.\n\n• \"link\" means the route can only reach hosts on the local network (one hop away).\n\n• \"host\" means the route will not leave the local machine (used for internal addresses\nlike 127.0.0.1).\n\n• \"nowhere\" means the destination doesn't exist.\n\nFor IPv4 route, defaults to \"host\" if Type= is \"local\" or \"nat\", and \"link\" if Type= is\n\"broadcast\", \"multicast\", \"anycast\", or \"unicast\". In other cases, defaults to \"global\".\nThe value is not used for IPv6.\n\nAdded in version 219.\n\nPreferredSource=\nThe preferred source address of the route. The address must be in the format described in\ninetpton(3).\n\nAdded in version 227.\n\nTable=\nThe table identifier for the route. Takes one of predefined names \"default\", \"main\", and\n\"local\", and names defined in RouteTable= in networkd.conf(5), or a number between 1 and\n4294967295. The table can be retrieved using ip route show table num. If unset and Type=\nis \"local\", \"broadcast\", \"anycast\", or \"nat\", then \"local\" is used. In other cases,\ndefaults to \"main\".\n\nAdded in version 230.\n\nHopLimit=\nConfigures per route hop limit. Takes an integer in the range 1...255. See also\nIPv6HopLimit=.\n\nAdded in version 255.\n\nProtocol=\nThe protocol identifier for the route. Takes a number between 0 and 255 or the special\nvalues \"kernel\", \"boot\", \"static\", \"ra\" and \"dhcp\". Defaults to \"static\".\n\nAdded in version 234.\n\nType=\nSpecifies the type for the route. Takes one of \"unicast\", \"local\", \"broadcast\",\n\"anycast\", \"multicast\", \"blackhole\", \"unreachable\", \"prohibit\", \"throw\", \"nat\", and\n\"xresolve\". If \"unicast\", a regular route is defined, i.e. a route indicating the path to\ntake to a destination network address. If \"blackhole\", packets to the defined route are\ndiscarded silently. If \"unreachable\", packets to the defined route are discarded and the\nICMP message \"Host Unreachable\" is generated. If \"prohibit\", packets to the defined route\nare discarded and the ICMP message \"Communication Administratively Prohibited\" is\ngenerated. If \"throw\", route lookup in the current routing table will fail and the route\nselection process will return to Routing Policy Database (RPDB). Defaults to \"unicast\".\n\nAdded in version 235.\n\nInitialCongestionWindow=\nThe TCP initial congestion window is used during the start of a TCP connection. During\nthe start of a TCP session, when a client requests a resource, the server's initial\ncongestion window determines how many packets will be sent during the initial burst of\ndata without waiting for acknowledgement. Takes a number between 1 and 1023. Note that\n100 is considered an extremely large value for this option. When unset, the kernel's\ndefault (typically 10) will be used.\n\nAdded in version 237.\n\nInitialAdvertisedReceiveWindow=\nThe TCP initial advertised receive window is the amount of receive data (in bytes) that\ncan initially be buffered at one time on a connection. The sending host can send only\nthat amount of data before waiting for an acknowledgment and window update from the\nreceiving host. Takes a number between 1 and 1023. Note that 100 is considered an\nextremely large value for this option. When unset, the kernel's default will be used.\n\nAdded in version 237.\n\nQuickAck=\nTakes a boolean. When true, the TCP quick ACK mode for the route is enabled. When unset,\nthe kernel's default will be used.\n\nAdded in version 237.\n\nFastOpenNoCookie=\nTakes a boolean. When true enables TCP fastopen without a cookie on a per-route basis.\nWhen unset, the kernel's default will be used.\n\nAdded in version 243.\n\nTTLPropagate=\nTakes a boolean. When true enables TTL propagation at Label Switched Path (LSP) egress.\nWhen unset, the kernel's default will be used.\n\nAdded in version 243.\n\nMTUBytes=\nThe maximum transmission unit in bytes to set for the route. The usual suffixes K, M, G,\nare supported and are understood to the base of 1024.\n\nAdded in version 239.\n\nTCPAdvertisedMaximumSegmentSize=\nSpecifies the Path MSS (in bytes) hints given on TCP layer. The usual suffixes K, M, G,\nare supported and are understood to the base of 1024. An unsigned integer in the range\n1...4294967294. When unset, the kernel's default will be used.\n\nAdded in version 248.\n\nTCPCongestionControlAlgorithm=\nSpecifies the TCP congestion control algorithm for the route. Takes a name of the\nalgorithm, e.g. \"bbr\", \"dctcp\", or \"vegas\". When unset, the kernel's default will be\nused.\n\nAdded in version 252.\n\nTCPRetransmissionTimeoutSec=\nSpecifies the TCP Retransmission Timeout (RTO) for the route. Takes time values in\nseconds. This value specifies the timeout of an alive TCP connection, when\nretransmissions remain unacknowledged. When unset, the kernel's default will be used.\n\nAdded in version 255.\n\nMultiPathRoute=address[@name] [weight]\nConfigures multipath route. Multipath routing is the technique of using multiple\nalternative paths through a network. Takes gateway address. Optionally, takes a network\ninterface name or index separated with \"@\", and a weight in 1..256 for this multipath\nroute separated with whitespace. This setting can be specified multiple times. If an\nempty string is assigned, then the all previous assignments are cleared.\n\nAdded in version 245.\n\nNextHop=\nSpecifies the nexthop id. Takes an unsigned integer in the range 1...4294967295. If set,\nthe corresponding [NextHop] section must be configured. Defaults to unset.\n\nAdded in version 248.\n\n[DHCPV4] SECTION OPTIONS\nThe [DHCPv4] section configures the DHCPv4 client, if it is enabled with the DHCP= setting\ndescribed above:\n\nRequestAddress=\nTakes an IPv4 address. When specified, the Requested IP Address option (option code 50)\nis added with it to the initial DHCPDISCOVER message sent by the DHCP client. Defaults to\nunset, and an already assigned dynamic address to the interface is automatically picked.\n\nAdded in version 255.\n\nSendHostname=\nWhen true (the default), the machine's hostname (or the value specified with Hostname=,\ndescribed below) will be sent to the DHCP server. Note that the hostname must consist\nonly of 7-bit ASCII lower-case characters and no spaces or dots, and be formatted as a\nvalid DNS domain name. Otherwise, the hostname is not sent even if this option is true.\n\nAdded in version 215.\n\nHostname=\nUse this value for the hostname which is sent to the DHCP server, instead of machine's\nhostname. Note that the specified hostname must consist only of 7-bit ASCII lower-case\ncharacters and no spaces or dots, and be formatted as a valid DNS domain name.\n\nAdded in version 223.\n\nMUDURL=\nWhen configured, the specified Manufacturer Usage Description (MUD) URL will be sent to\nthe DHCPv4 server. Takes a URL of length up to 255 characters. A superficial verification\nthat the string is a valid URL will be performed. DHCPv4 clients are intended to have at\nmost one MUD URL associated with them. See \u001b[34mRFC 8520[22].\n\nMUD is an embedded software standard defined by the IETF that allows IoT device makers to\nadvertise device specifications, including the intended communication patterns for their\ndevice when it connects to the network. The network can then use this to author a\ncontext-specific access policy, so the device functions only within those parameters.\n\nAdded in version 246.\n\nClientIdentifier=\nThe DHCPv4 client identifier to use. Takes one of mac or duid. If set to mac, the MAC\naddress of the link is used. If set to duid, an RFC4361-compliant Client ID, which is the\ncombination of IAID and DUID, is used. IAID can be configured by IAID=. DUID can be\nconfigured by DUIDType= and DUIDRawData=. Defaults to duid.\n\nAdded in version 220.\n\nVendorClassIdentifier=\nThe vendor class identifier used to identify vendor type and configuration.\n\nAdded in version 216.\n\nUserClass=\nA DHCPv4 client can use UserClass option to identify the type or category of user or\napplications it represents. The information contained in this option is a string that\nrepresents the user class of which the client is a member. Each class sets an identifying\nstring of information to be used by the DHCP service to classify clients. Takes a\nwhitespace-separated list of strings.\n\nAdded in version 239.\n\nDUIDType=\nOverride the global DUIDType= setting for this network. See networkd.conf(5) for a\ndescription of possible values.\n\nAdded in version 230.\n\nDUIDRawData=\nOverride the global DUIDRawData= setting for this network. See networkd.conf(5) for a\ndescription of possible values.\n\nAdded in version 230.\n\nIAID=\nThe DHCP Identity Association Identifier (IAID) for the interface, a 32-bit unsigned\ninteger.\n\nAdded in version 230.\n\nRapidCommit=\nTakes a boolean. The DHCPv4 client can obtain configuration parameters from a DHCPv4\nserver through a rapid two-message exchange (discover and ack). When the rapid commit\noption is set by both the DHCPv4 client and the DHCPv4 server, the two-message exchange\nis used. Otherwise, the four-message exchange (discover, offer, request, and ack) is\nused. The two-message exchange provides faster client configuration. See \u001b[34mRFC 4039[23] for\ndetails. Defaults to true when Anonymize=no and neither AllowList= nor DenyList= is\nspecified, and false otherwise.\n\nAdded in version 255.\n\nAnonymize=\nTakes a boolean. When true, the options sent to the DHCP server will follow the \u001b[34mRFC\n\u001b[34m7844[24] (Anonymity Profiles for DHCP Clients) to minimize disclosure of identifying\ninformation. Defaults to false.\n\nThis option should only be set to true when MACAddressPolicy= is set to random (see\nsystemd.link(5)).\n\nWhen true, ClientIdentifier=mac, RapidCommit=no, SendHostname=no, Use6RD=no,\nUseCaptivePortal=no, UseMTU=no, UseNTP=no, UseSIP=no, and UseTimezone=no are implied and\nthese settings in the .network file are silently ignored. Also, Hostname=, MUDURL=,\nRequestAddress, RequestOptions=, SendOption=, SendVendorOption=, UserClass=, and\nVendorClassIdentifier= are silently ignored.\n\nWith this option enabled DHCP requests will mimic those generated by Microsoft Windows,\nin order to reduce the ability to fingerprint and recognize installations. This means\nDHCP request sizes will grow and lease data will be more comprehensive than normally,\nthough most of the requested data is not actually used.\n\nAdded in version 235.\n\nRequestOptions=\nSets request options to be sent to the server in the DHCPv4 request options list. A\nwhitespace-separated list of integers in the range 1...254. Defaults to unset.\n\nAdded in version 244.\n\nSendOption=\nSend an arbitrary raw option in the DHCPv4 request. Takes a DHCP option number, data type\nand data separated with a colon (\"option:type:value\"). The option number must be an\ninteger in the range 1...254. The type takes one of \"uint8\", \"uint16\", \"uint32\",\n\"ipv4address\", or \"string\". Special characters in the data string may be escaped using\n\u001b[34mC-style escapes[25]. This setting can be specified multiple times. If an empty string is\nspecified, then all options specified earlier are cleared. Defaults to unset.\n\nAdded in version 244.\n\nSendVendorOption=\nSend an arbitrary vendor option in the DHCPv4 request. Takes a DHCP option number, data\ntype and data separated with a colon (\"option:type:value\"). The option number must be an\ninteger in the range 1...254. The type takes one of \"uint8\", \"uint16\", \"uint32\",\n\"ipv4address\", or \"string\". Special characters in the data string may be escaped using\n\u001b[34mC-style escapes[25]. This setting can be specified multiple times. If an empty string is\nspecified, then all options specified earlier are cleared. Defaults to unset.\n\nAdded in version 246.\n\nIPServiceType=\nTakes one of the special values \"none\", \"CS6\", or \"CS4\". When \"none\" no IP service type\nis set to the packet sent from the DHCPv4 client. When \"CS6\" (network control) or \"CS4\"\n(realtime), the corresponding service type will be set. Defaults to \"CS6\".\n\nAdded in version 244.\n\nSocketPriority=\nThe Linux socket option SOPRIORITY applied to the raw IP socket used for initial DHCPv4\nmessages. Unset by default. Usual values range from 0 to 6. More details about\nSOPRIORITY socket option in socket(7). Can be used in conjunction with [VLAN] section\nEgressQOSMaps= setting of .netdev file to set the 802.1Q VLAN ethernet tagged header\npriority, see systemd.netdev(5).\n\nAdded in version 253.\n\nLabel=\nSpecifies the label for the IPv4 address received from the DHCP server. The label must be\na 7-bit ASCII string with a length of 1...15 characters. Defaults to unset.\n\nAdded in version 250.\n\nUseDNS=\nWhen true (the default), the DNS servers received from the DHCP server will be used.\n\nThis corresponds to the nameserver option in resolv.conf(5).\n\nAdded in version 211.\n\nRoutesToDNS=\nWhen true, the routes to the DNS servers received from the DHCP server will be\nconfigured. When UseDNS= is disabled, this setting is ignored. Defaults to true.\n\nAdded in version 243.\n\nUseNTP=\nWhen true (the default), the NTP servers received from the DHCP server will be used by\nsystemd-timesyncd.service.\n\nAdded in version 220.\n\nRoutesToNTP=\nWhen true, the routes to the NTP servers received from the DHCP server will be\nconfigured. When UseNTP= is disabled, this setting is ignored. Defaults to true.\n\nAdded in version 249.\n\nUseSIP=\nWhen true (the default), the SIP servers received from the DHCP server will be collected\nand made available to client programs.\n\nAdded in version 244.\n\nUseCaptivePortal=\nWhen true (the default), the captive portal advertised by the DHCP server will be\nrecorded and made available to client programs and displayed in the networkctl(1) status\noutput per-link.\n\nAdded in version 254.\n\nUseMTU=\nWhen true, the interface maximum transmission unit from the DHCP server will be used on\nthe current link. If MTUBytes= is set, then this setting is ignored. Defaults to false.\n\nNote, some drivers will reset the interfaces if the MTU is changed. For such interfaces,\nplease try to use IgnoreCarrierLoss= with a short timespan, e.g. \"3 seconds\".\n\nAdded in version 211.\n\nUseHostname=\nWhen true (the default), the hostname received from the DHCP server will be set as the\ntransient hostname of the system.\n\nAdded in version 211.\n\nUseDomains=\nTakes a boolean, or the special value route. When true, the domain name received from the\nDHCP server will be used as DNS search domain over this link, similarly to the effect of\nthe Domains= setting. If set to route, the domain name received from the DHCP server will\nbe used for routing DNS queries only, but not for searching, similarly to the effect of\nthe Domains= setting when the argument is prefixed with \"~\". Defaults to true on Ubuntu.\n\nIt is recommended to enable this option only on trusted networks, as setting this affects\nresolution of all hostnames, in particular of single-label names. It is generally safer\nto use the supplied domain only as routing domain, rather than as search domain, in order\nto not have it affect local resolution of single-label names.\n\nWhen set to true, this setting corresponds to the domain option in resolv.conf(5).\n\nAdded in version 216.\n\nUseRoutes=\nWhen true (the default), the static routes will be requested from the DHCP server and\nadded to the routing table with a metric of 1024, and a scope of global, link or host,\ndepending on the route's destination and gateway. If the destination is on the local\nhost, e.g., 127.x.x.x, or the same as the link's own address, the scope will be set to\nhost. Otherwise if the gateway is null (a direct route), a link scope will be used. For\nanything else, scope defaults to global.\n\nAdded in version 215.\n\nRouteMetric=\nSet the routing metric for routes specified by the DHCP server (including the prefix\nroute added for the specified prefix). Takes an unsigned integer in the range\n0...4294967295. Defaults to 1024.\n\nAdded in version 217.\n\nRouteTable=num\nThe table identifier for DHCP routes. Takes one of predefined names \"default\", \"main\",\nand \"local\", and names defined in RouteTable= in networkd.conf(5), or a number between\n1...4294967295.\n\nWhen used in combination with VRF=, the VRF's routing table is used when this parameter\nis not specified.\n\nAdded in version 232.\n\nRouteMTUBytes=\nSpecifies the MTU for the DHCP routes. Please see the [Route] section for further\ndetails.\n\nAdded in version 245.\n\nQuickAck=\nTakes a boolean. When true, the TCP quick ACK mode is enabled for the routes configured\nby the acquired DHCPv4 lease. When unset, the kernel's default will be used.\n\nAdded in version 253.\n\nInitialCongestionWindow=\nAs in the [Route] section.\n\nAdded in version 255.\n\nInitialAdvertisedReceiveWindow=\nAs in the [Route] section.\n\nAdded in version 255.\n\nUseGateway=\nWhen true, and the DHCP server provides a Router option, the default gateway based on the\nrouter address will be configured. Defaults to unset, and the value specified with\nUseRoutes= will be used.\n\nNote, when the server provides both the Router and Classless Static Routes option, and\nUseRoutes= is enabled, the Router option is always ignored regardless of this setting.\nSee \u001b[34mRFC 3442[26].\n\nAdded in version 246.\n\nUseTimezone=\nWhen true, the timezone received from the DHCP server will be set as timezone of the\nlocal system. Defaults to false.\n\nAdded in version 226.\n\nUse6RD=\nWhen true, subnets of the received IPv6 prefix are assigned to downstream interfaces\nwhich enables DHCPPrefixDelegation=. See also DHCPPrefixDelegation= in the [Network]\nsection, the [DHCPPrefixDelegation] section, and \u001b[34mRFC 5969[27]. Defaults to false.\n\nAdded in version 250.\n\nIPv6OnlyMode=\nWhen true, the DHCPv4 configuration will be delayed by the timespan provided by the DHCP\nserver and skip to configure dynamic IPv4 network connectivity if IPv6 connectivity is\nprovided within the timespan. See \u001b[34mRFC 8925[28]. Defaults to false.\n\nAdded in version 255.\n\nFallbackLeaseLifetimeSec=\nAllows one to set DHCPv4 lease lifetime when DHCPv4 server does not send the lease\nlifetime. Takes one of \"forever\" or \"infinity\". If specified, the acquired address never\nexpires. Defaults to unset.\n\nAdded in version 246.\n\nRequestBroadcast=\nRequest the server to use broadcast messages before the IP address has been configured.\nThis is necessary for devices that cannot receive RAW packets, or that cannot receive\npackets at all before an IP address has been configured. On the other hand, this must not\nbe enabled on networks where broadcasts are filtered out.\n\nAdded in version 216.\n\nMaxAttempts=\nSpecifies how many times the DHCPv4 client configuration should be attempted. Takes a\nnumber or \"infinity\". Defaults to \"infinity\". Note that the time between retries is\nincreased exponentially, up to approximately one per minute, so the network will not be\noverloaded even if this number is high. The default is suitable in most circumstances.\n\nAdded in version 243.\n\nListenPort=\nSet the port from which the DHCP client packets originate.\n\nAdded in version 233.\n\nDenyList=\nA whitespace-separated list of IPv4 addresses. Each address can optionally take a prefix\nlength after \"/\". DHCP offers from servers in the list are rejected. Note that if\nAllowList= is configured then DenyList= is ignored.\n\nNote that this filters only DHCP offers, so the filtering may not work when RapidCommit=\nis enabled. See also RapidCommit= in the above.\n\nAdded in version 246.\n\nAllowList=\nA whitespace-separated list of IPv4 addresses. Each address can optionally take a prefix\nlength after \"/\". DHCP offers from servers in the list are accepted.\n\nNote that this filters only DHCP offers, so the filtering may not work when RapidCommit=\nis enabled. See also RapidCommit= in the above.\n\nAdded in version 246.\n\nSendRelease=\nWhen true, the DHCPv4 client sends a DHCP release packet when it stops. Defaults to true.\n\nAdded in version 243.\n\nSendDecline=\nA boolean. When true, systemd-networkd performs IPv4 Duplicate Address Detection to the\nacquired address by the DHCPv4 client. If duplicate is detected, the DHCPv4 client\nrejects the address by sending a DHCPDECLINE packet to the DHCP server, and tries to\nobtain an IP address again. See \u001b[34mRFC 5227[11]. Defaults to false.\n\nAdded in version 245.\n\nNetLabel=\nThis applies the NetLabel for the addresses received with DHCP, like NetLabel= in\n[Address] section applies it to statically configured addresses. See NetLabel= in\n[Address] section for more details.\n\nAdded in version 252.\n\nNFTSet=\nThis applies the NFT set for the network configuration received with DHCP, like NFTSet=\nin [Address] section applies it to static configuration. See NFTSet= in [Address] section\nfor more details. For \"address\" or \"prefix\" source types, the type of the element used in\nthe NFT filter must be \"ipv4addr\".\n\nAdded in version 255.\n\n[DHCPV6] SECTION OPTIONS\nThe [DHCPv6] section configures the DHCPv6 client, if it is enabled with the DHCP= setting\ndescribed above, or invoked by the IPv6 Router Advertisement:\n\nMUDURL=, IAID=, DUIDType=, DUIDRawData=, RequestOptions=\nAs in the [DHCPv4] section.\n\nAdded in version 246.\n\nSendOption=\nAs in the [DHCPv4] section, however because DHCPv6 uses 16-bit fields to store option\nnumbers, the option number is an integer in the range 1...65536.\n\nAdded in version 246.\n\nSendVendorOption=\nSend an arbitrary vendor option in the DHCPv6 request. Takes an enterprise identifier,\nDHCP option number, data type, and data separated with a colon (\"enterprise\nidentifier:option:type:value\"). Enterprise identifier is an unsigned integer in the range\n1...4294967294. The option number must be an integer in the range 1...254. Data type\ntakes one of \"uint8\", \"uint16\", \"uint32\", \"ipv4address\", \"ipv6address\", or \"string\".\nSpecial characters in the data string may be escaped using \u001b[34mC-style escapes[25]. This\nsetting can be specified multiple times. If an empty string is specified, then all\noptions specified earlier are cleared. Defaults to unset.\n\nAdded in version 246.\n\nUserClass=\nA DHCPv6 client can use User Class option to identify the type or category of user or\napplications it represents. The information contained in this option is a string that\nrepresents the user class of which the client is a member. Each class sets an identifying\nstring of information to be used by the DHCP service to classify clients. Special\ncharacters in the data string may be escaped using \u001b[34mC-style escapes[25]. This setting can\nbe specified multiple times. If an empty string is specified, then all options specified\nearlier are cleared. Takes a whitespace-separated list of strings. Note that currently\nNUL bytes are not allowed.\n\nAdded in version 246.\n\nVendorClass=\nA DHCPv6 client can use VendorClass option to identify the vendor that manufactured the\nhardware on which the client is running. The information contained in the data area of\nthis option is contained in one or more opaque fields that identify details of the\nhardware configuration. Takes a whitespace-separated list of strings.\n\nAdded in version 246.\n\nPrefixDelegationHint=\nTakes an IPv6 address with prefix length in the same format as the Address= in the\n[Network] section. The DHCPv6 client will include a prefix hint in the DHCPv6\nsolicitation sent to the server. The prefix length must be in the range 1...128. Defaults\nto unset.\n\nAdded in version 244.\n\nRapidCommit=\nTakes a boolean. The DHCPv6 client can obtain configuration parameters from a DHCPv6\nserver through a rapid two-message exchange (solicit and reply). When the rapid commit\noption is set by both the DHCPv6 client and the DHCPv6 server, the two-message exchange\nis used. Otherwise, the four-message exchange (solicit, advertise, request, and reply) is\nused. The two-message exchange provides faster client configuration. See \u001b[34mRFC 3315[29] for\ndetails. Defaults to true, and the two-message exchange will be used if the server\nsupport it.\n\nAdded in version 252.\n\nSendHostname=\nWhen true (the default), the machine's hostname (or the value specified with Hostname=,\ndescribed below) will be sent to the DHCPv6 server. Note that the hostname must consist\nonly of 7-bit ASCII lower-case characters and no spaces or dots, and be formatted as a\nvalid DNS domain name. Otherwise, the hostname is not sent even if this option is true.\n\nAdded in version 255.\n\nHostname=\nUse this value for the hostname which is sent to the DHCPv6 server, instead of machine's\nhostname. Note that the specified hostname must consist only of 7-bit ASCII lower-case\ncharacters and no spaces or dots, and be formatted as a valid DNS domain name.\n\nAdded in version 255.\n\nUseAddress=\nWhen true (the default), the IP addresses provided by the DHCPv6 server will be assigned.\n\nAdded in version 248.\n\nUseCaptivePortal=\nWhen true (the default), the captive portal advertised by the DHCPv6 server will be\nrecorded and made available to client programs and displayed in the networkctl(1) status\noutput per-link.\n\nAdded in version 254.\n\nUseDelegatedPrefix=\nWhen true (the default), the client will request the DHCPv6 server to delegate prefixes.\nIf the server provides prefixes to be delegated, then subnets of the prefixes are\nassigned to the interfaces that have DHCPPrefixDelegation=yes. See also the\nDHCPPrefixDelegation= setting in the [Network] section, settings in the\n[DHCPPrefixDelegation] section, and \u001b[34mRFC 8415[30].\n\nAdded in version 250.\n\nUseDNS=, UseNTP=, UseHostname=, UseDomains=, NetLabel=, SendRelease=\nAs in the [DHCPv4] section.\n\nAdded in version 243.\n\nNFTSet=\nThis applies the NFT set for the network configuration received with DHCP, like NFTSet=\nin [Address] section applies it to static configuration. See NFTSet= in [Address] section\nfor more details. For \"address\" or \"prefix\" source types, the type of the element used in\nthe NFT filter must be \"ipv6addr\".\n\nAdded in version 255.\n\nWithoutRA=\nAllows DHCPv6 client to start without router advertisements's \"managed\" or \"other\nconfiguration\" flag. Takes one of \"no\", \"solicit\", or \"information-request\". If this is\nnot specified, \"solicit\" is used when DHCPPrefixDelegation= is enabled and\nUplinkInterface=:self is specified in the [DHCPPrefixDelegation] section. Otherwise,\ndefaults to \"no\", and the DHCPv6 client will be started when an RA is received. See also\nthe DHCPv6Client= setting in the [IPv6AcceptRA] section.\n\nAdded in version 246.\n\n[DHCPPREFIXDELEGATION] SECTION OPTIONS\nThe [DHCPPrefixDelegation] section configures subnet prefixes of the delegated prefixes\nacquired by a DHCPv6 client or by a DHCPv4 client through the 6RD option on another\ninterface. The settings in this section are used only when the DHCPPrefixDelegation= setting\nin the [Network] section is enabled.\n\nUplinkInterface=\nSpecifies the name or the index of the uplink interface, or one of the special values\n\":self\" and \":auto\". When \":self\", the interface itself is considered the uplink\ninterface, and WithoutRA=solicit is implied if the setting is not explicitly specified.\nWhen \":auto\", the first link which acquired prefixes to be delegated from the DHCPv6 or\nDHCPv4 server is selected. Defaults to \":auto\".\n\nAdded in version 250.\n\nSubnetId=\nConfigure a specific subnet ID on the interface from a (previously) received prefix\ndelegation. You can either set \"auto\" (the default) or a specific subnet ID (as defined\nin \u001b[34mRFC 4291[31], section 2.5.4), in which case the allowed value is hexadecimal, from 0\nto 0x7fffffffffffffff inclusive.\n\nAdded in version 246.\n\nAnnounce=\nTakes a boolean. When enabled, and IPv6SendRA= in [Network] section is enabled, the\ndelegated prefixes are distributed through the IPv6 Router Advertisement. This setting\nwill be ignored when the DHCPPrefixDelegation= setting is enabled on the upstream\ninterface. Defaults to yes.\n\nAdded in version 247.\n\nAssign=\nTakes a boolean. Specifies whether to add an address from the delegated prefixes which\nare received from the WAN interface by the DHCPv6 Prefix Delegation. When true (on LAN\ninterface), the EUI-64 algorithm will be used by default to form an interface identifier\nfrom the delegated prefixes. See also Token= setting below. Defaults to yes.\n\nAdded in version 246.\n\nToken=\nSpecifies an optional address generation mode for assigning an address in each delegated\nprefix. This accepts the same syntax as Token= in the [IPv6AcceptRA] section. If Assign=\nis set to false, then this setting will be ignored. Defaults to unset, which means the\nEUI-64 algorithm will be used.\n\nAdded in version 246.\n\nManageTemporaryAddress=\nAs in the [Address] section, but defaults to true.\n\nAdded in version 248.\n\nRouteMetric=\nThe metric of the route to the delegated prefix subnet. Takes an unsigned integer in the\nrange 0...4294967295. When set to 0, the kernel's default value is used. Defaults to 256.\n\nAdded in version 249.\n\nNetLabel=\nThis applies the NetLabel for the addresses received with DHCP, like NetLabel= in\n[Address] section applies it to statically configured addresses. See NetLabel= in\n[Address] section for more details.\n\nAdded in version 252.\n\nNFTSet=\nThis applies the NFT set for the network configuration received with DHCP, like NFTSet=\nin [Address] section applies it to static configuration. See NFTSet= in [Address] section\nfor more details. For \"address\" or \"prefix\" source types, the type of the element used in\nthe NFT filter must be \"ipv6addr\".\n\nAdded in version 255.\n\n[IPV6ACCEPTRA] SECTION OPTIONS\nThe [IPv6AcceptRA] section configures the IPv6 Router Advertisement (RA) client, if it is\nenabled with the IPv6AcceptRA= setting described above:\n\nToken=\nSpecifies an optional address generation mode for the Stateless Address Autoconfiguration\n(SLAAC). The following values are supported:\n\neui64\nThe EUI-64 algorithm will be used to generate an address for that prefix. Only\nsupported by Ethernet or InfiniBand interfaces.\n\nAdded in version 250.\n\nstatic:ADDRESS\nAn IPv6 address must be specified after a colon (\":\"), and the lower bits of the\nsupplied address are combined with the upper bits of a prefix received in a Router\nAdvertisement (RA) message to form a complete address. Note that if multiple prefixes\nare received in an RA message, or in multiple RA messages, addresses will be formed\nfrom each of them using the supplied address. This mode implements SLAAC but uses a\nstatic interface identifier instead of an identifier generated by using the EUI-64\nalgorithm. Because the interface identifier is static, if Duplicate Address Detection\ndetects that the computed address is a duplicate (in use by another node on the\nlink), then this mode will fail to provide an address for that prefix. If an IPv6\naddress without mode is specified, then \"static\" mode is assumed.\n\nAdded in version 250.\n\nprefixstable[:ADDRESS][,UUID]\nThe algorithm specified in \u001b[34mRFC 7217[32] will be used to generate interface\nidentifiers. This mode can optionally take an IPv6 address separated with a colon\n(\":\"). If an IPv6 address is specified, then an interface identifier is generated\nonly when a prefix received in an RA message matches the supplied address.\n\nThis mode can also optionally take a non-null UUID in the format which\nsdid128fromstring() accepts, e.g. \"86b123b969ba4b7eb8b3d8605123525a\" or\n\"86b123b9-69ba-4b7e-b8b3-d8605123525a\". If a UUID is specified, the value is used as\nthe secret key to generate interface identifiers. If not specified, then an\napplication specific ID generated with the system's machine-ID will be used as the\nsecret key. See sd-id128(3), sdid128fromstring(3), and sdid128getmachine(3).\n\nNote that the \"prefixstable\" algorithm uses both the interface name and MAC address\nas input to the hash to compute the interface identifier, so if either of those are\nchanged the resulting interface identifier (and address) will be changed, even if the\nprefix received in the RA message has not been changed.\n\nAdded in version 250.\n\nIf no address generation mode is specified (which is the default), or a received prefix\ndoes not match any of the addresses provided in \"prefixstable\" mode, then the EUI-64\nalgorithm will be used for Ethernet or InfiniBand interfaces, otherwise \"prefixstable\"\nwill be used to form an interface identifier for that prefix.\n\nThis setting can be specified multiple times. If an empty string is assigned, then the\nall previous assignments are cleared.\n\nExamples:\n\nToken=eui64\nToken=::1a:2b:3c:4d\nToken=static:::1a:2b:3c:4d\nToken=prefixstable\nToken=prefixstable:2002:da8:1::\n\nAdded in version 250.\n\nUseDNS=\nWhen true (the default), the DNS servers received in the Router Advertisement will be\nused.\n\nThis corresponds to the nameserver option in resolv.conf(5).\n\nAdded in version 231.\n\nUseDomains=\nTakes a boolean, or the special value \"route\". When true, the domain name received via\nIPv6 Router Advertisement (RA) will be used as DNS search domain over this link,\nsimilarly to the effect of the Domains= setting. If set to \"route\", the domain name\nreceived via IPv6 RA will be used for routing DNS queries only, but not for searching,\nsimilarly to the effect of the Domains= setting when the argument is prefixed with \"~\".\nDefaults to true on Ubuntu.\n\nIt is recommended to enable this option only on trusted networks, as setting this affects\nresolution of all hostnames, in particular of single-label names. It is generally safer\nto use the supplied domain only as routing domain, rather than as search domain, in order\nto not have it affect local resolution of single-label names.\n\nWhen set to true, this setting corresponds to the domain option in resolv.conf(5).\n\nAdded in version 231.\n\nRouteTable=num\nThe table identifier for the routes received in the Router Advertisement. Takes one of\npredefined names \"default\", \"main\", and \"local\", and names defined in RouteTable= in\nnetworkd.conf(5), or a number between 1...4294967295.\n\nWhen used in combination with VRF=, the VRF's routing table is used when this parameter\nis not specified.\n\nAdded in version 232.\n\nRouteMetric=\nSet the routing metric for the routes received in the Router Advertisement. Takes an\nunsigned integer in the range 0...4294967295, or three unsigned integer separated with\n\":\", in that case the first one is used when the router preference is high, the second is\nfor medium preference, and the last is for low preference (\"high:medium:low\"). Defaults\nto \"512:1024:2048\".\n\nAdded in version 249.\n\nQuickAck=\nTakes a boolean. When true, the TCP quick ACK mode is enabled for the routes configured\nby the received RAs. When unset, the kernel's default will be used.\n\nAdded in version 253.\n\nUseMTU=\nTakes a boolean. When true, the MTU received in the Router Advertisement will be used.\nDefaults to true.\n\nAdded in version 250.\n\nUseHopLimit=\nTakes a boolean. When true, the hop limit received in the Router Advertisement will be\nset to routes configured based on the advertisement. See also IPv6HopLimit=. Defaults to\ntrue.\n\nAdded in version 255.\n\nUseICMP6RateLimit=\nTakes a boolean. When true, the ICMP6 rate limit received in the Router Advertisement\nwill be set to ICMP6 rate limit based on the advertisement. Defaults to true.\n\nAdded in version 255.\n\nUseGateway=\nWhen true (the default), the router address will be configured as the default gateway.\n\nAdded in version 250.\n\nUseRoutePrefix=\nWhen true (the default), the routes corresponding to the route prefixes received in the\nRouter Advertisement will be configured.\n\nAdded in version 250.\n\nUseCaptivePortal=\nWhen true (the default), the captive portal received in the Router Advertisement will be\nrecorded and made available to client programs and displayed in the networkctl(1) status\noutput per-link.\n\nAdded in version 254.\n\nUsePREF64=\nWhen true, the IPv6 PREF64 (or NAT64) prefixes received in the Router Advertisement will\nbe recorded and made available to client programs and displayed in the networkctl(1)\nstatus output per-link. See \u001b[34mRFC 8781[33]. Defaults to false.\n\nAdded in version 255.\n\nUseAutonomousPrefix=\nWhen true (the default), the autonomous prefix received in the Router Advertisement will\nbe used and take precedence over any statically configured ones.\n\nAdded in version 242.\n\nUseOnLinkPrefix=\nWhen true (the default), the onlink prefix received in the Router Advertisement will be\nused and takes precedence over any statically configured ones.\n\nAdded in version 242.\n\nRouterDenyList=\nA whitespace-separated list of IPv6 router addresses. Each address can optionally take a\nprefix length after \"/\". Any information advertised by the listed router is ignored.\n\nAdded in version 248.\n\nRouterAllowList=\nA whitespace-separated list of IPv6 router addresses. Each address can optionally take a\nprefix length after \"/\". Only information advertised by the listed router is accepted.\nNote that if RouterAllowList= is configured then RouterDenyList= is ignored.\n\nAdded in version 248.\n\nPrefixDenyList=\nA whitespace-separated list of IPv6 prefixes. Each prefix can optionally take its prefix\nlength after \"/\". IPv6 prefixes supplied via router advertisements in the list are\nignored.\n\nAdded in version 248.\n\nPrefixAllowList=\nA whitespace-separated list of IPv6 prefixes. Each prefix can optionally take its prefix\nlength after \"/\". IPv6 prefixes supplied via router advertisements in the list are\nallowed. Note that if PrefixAllowList= is configured then PrefixDenyList= is ignored.\n\nAdded in version 248.\n\nRouteDenyList=\nA whitespace-separated list of IPv6 route prefixes. Each prefix can optionally take its\nprefix length after \"/\". IPv6 route prefixes supplied via router advertisements in the\nlist are ignored.\n\nAdded in version 248.\n\nRouteAllowList=\nA whitespace-separated list of IPv6 route prefixes. Each prefix can optionally take its\nprefix length after \"/\". IPv6 route prefixes supplied via router advertisements in the\nlist are allowed. Note that if RouteAllowList= is configured then RouteDenyList= is\nignored.\n\nAdded in version 248.\n\nDHCPv6Client=\nTakes a boolean, or the special value \"always\". When true, the DHCPv6 client will be\nstarted in \"solicit\" mode if the RA has the \"managed\" flag or \"information-request\" mode\nif the RA lacks the \"managed\" flag but has the \"other configuration\" flag. If set to\n\"always\", the DHCPv6 client will be started in \"solicit\" mode when an RA is received,\neven if neither the \"managed\" nor the \"other configuration\" flag is set in the RA. This\nwill be ignored when WithoutRA= in the [DHCPv6] section is enabled, or\nUplinkInterface=:self in the [DHCPPrefixDelegation] section is specified. Defaults to\ntrue.\n\nAdded in version 246.\n\nNetLabel=\nThis applies the NetLabel for the addresses received with RA, like NetLabel= in [Address]\nsection applies it to statically configured addresses. See NetLabel= in [Address] section\nfor more details.\n\nAdded in version 252.\n\nNFTSet=\nThis applies the NFT set for the network configuration received with RA, like NFTSet= in\n[Address] section applies it to static configuration. See NFTSet= in [Address] section\nfor more details. For \"address\" or \"prefix\" source types, the type of the element used in\nthe NFT filter must be \"ipv6addr\".\n\nAdded in version 255.\n\n[DHCPSERVER] SECTION OPTIONS\nThe [DHCPServer] section contains settings for the DHCP server, if enabled via the\nDHCPServer= option described above:\n\nServerAddress=\nSpecifies the server address for the DHCP server. Takes an IPv4 address with prefix\nlength separated with a slash, e.g. \"192.168.0.1/24\". Defaults to unset, and one of\nstatic IPv4 addresses configured in [Network] or [Address] section will be automatically\nselected. This setting may be useful when the interface on which the DHCP server is\nrunning has multiple static IPv4 addresses.\n\nThis implies Address= in [Network] or [Address] section with the same address and prefix\nlength. That is,\n\n[Network]\nDHCPServer=yes\nAddress=192.168.0.1/24\nAddress=192.168.0.2/24\n[DHCPServer]\nServerAddress=192.168.0.1/24\n\nor\n\n[Network]\nDHCPServer=yes\n[Address]\nAddress=192.168.0.1/24\n[Address]\nAddress=192.168.0.2/24\n[DHCPServer]\nServerAddress=192.168.0.1/24\n\nare equivalent to the following.\n\n[Network]\nDHCPServer=yes\nAddress=192.168.0.2/24\n[DHCPServer]\nServerAddress=192.168.0.1/24\n\nSince version 255, like the Address= setting in [Network] or [Address] section, this also\nsupports a null address, e.g. \"0.0.0.0/24\", and an unused address will be automatically\nselected. For more details about the automatic address selection, see Address= setting in\n[Network] section in the above.\n\nAdded in version 249.\n\nPoolOffset=, PoolSize=\nConfigures the pool of addresses to hand out. The pool is a contiguous sequence of IP\naddresses in the subnet configured for the server address, which does not include the\nsubnet nor the broadcast address. PoolOffset= takes the offset of the pool from the\nstart of subnet, or zero to use the default value. PoolSize= takes the number of IP\naddresses in the pool or zero to use the default value. By default, the pool starts at\nthe first address after the subnet address and takes up the rest of the subnet, excluding\nthe broadcast address. If the pool includes the server address (the default), this is\nreserved and not handed out to clients.\n\nAdded in version 226.\n\nDefaultLeaseTimeSec=, MaxLeaseTimeSec=\nControl the default and maximum DHCP lease time to pass to clients. These settings take\ntime values in seconds or another common time unit, depending on the suffix. The default\nlease time is used for clients that did not ask for a specific lease time. If a client\nasks for a lease time longer than the maximum lease time, it is automatically shortened\nto the specified time. The default lease time defaults to 1h, the maximum lease time to\n12h. Shorter lease times are beneficial if the configuration data in DHCP leases changes\nfrequently and clients shall learn the new settings with shorter latencies. Longer lease\ntimes reduce the generated DHCP network traffic.\n\nAdded in version 226.\n\nUplinkInterface=\nSpecifies the name or the index of the uplink interface, or one of the special values\n\":none\" and \":auto\". When emitting DNS, NTP, or SIP servers is enabled but no servers are\nspecified, the servers configured in the uplink interface will be emitted. When \":auto\",\nthe link which has a default gateway with the highest priority will be automatically\nselected. When \":none\", no uplink interface will be selected. Defaults to \":auto\".\n\nAdded in version 249.\n\nEmitDNS=, DNS=\nEmitDNS= takes a boolean. Configures whether the DHCP leases handed out to clients shall\ncontain DNS server information. Defaults to \"yes\". The DNS servers to pass to clients may\nbe configured with the DNS= option, which takes a list of IPv4 addresses, or special\nvalue \"serveraddress\" which will be converted to the address used by the DHCP server.\n\nIf the EmitDNS= option is enabled but no servers configured, the servers are\nautomatically propagated from an \"uplink\" interface that has appropriate servers set. The\n\"uplink\" interface is determined by the default route of the system with the highest\npriority. Note that this information is acquired at the time the lease is handed out, and\ndoes not take uplink interfaces into account that acquire DNS server information at a\nlater point. If no suitable uplink interface is found the DNS server data from\n/etc/resolv.conf is used. Also, note that the leases are not refreshed if the uplink\nnetwork configuration changes. To ensure clients regularly acquire the most current\nuplink DNS server information, it is thus advisable to shorten the DHCP lease time via\nMaxLeaseTimeSec= described above.\n\nThis setting can be specified multiple times. If an empty string is specified, then all\nDNS servers specified earlier are cleared.\n\nAdded in version 226.\n\nEmitNTP=, NTP=, EmitSIP=, SIP=, EmitPOP3=, POP3=, EmitSMTP=, SMTP=, EmitLPR=, LPR=\nSimilar to the EmitDNS= and DNS= settings described above, these settings configure\nwhether and what server information for the indicate protocol shall be emitted as part of\nthe DHCP lease. The same syntax, propagation semantics and defaults apply as for EmitDNS=\nand DNS=.\n\nAdded in version 226.\n\nEmitRouter=, Router=\nThe EmitRouter= setting takes a boolean value, and configures whether the DHCP lease\nshould contain the router option. The Router= setting takes an IPv4 address, and\nconfigures the router address to be emitted. When the Router= setting is not specified,\nthen the server address will be used for the router option. When the EmitRouter= setting\nis disabled, the Router= setting will be ignored. The EmitRouter= setting defaults to\ntrue, and the Router= setting defaults to unset.\n\nAdded in version 230.\n\nEmitTimezone=, Timezone=\nTakes a boolean. Configures whether the DHCP leases handed out to clients shall contain\ntimezone information. Defaults to \"yes\". The Timezone= setting takes a timezone string\n(such as \"Europe/Berlin\" or \"UTC\") to pass to clients. If no explicit timezone is set,\nthe system timezone of the local host is propagated, as determined by the /etc/localtime\nsymlink.\n\nAdded in version 226.\n\nBootServerAddress=\nTakes an IPv4 address of the boot server used by e.g. PXE boot systems. When specified,\nthis address is sent in the siaddr field of the DHCP message header. See \u001b[34mRFC 2131[34] for\nmore details. Defaults to unset.\n\nAdded in version 251.\n\nBootServerName=\nTakes a name of the boot server used by e.g. PXE boot systems. When specified, this name\nis sent in the DHCP option 66 (\"TFTP server name\"). See \u001b[34mRFC 2132[35] for more details.\nDefaults to unset.\n\nNote that typically setting one of BootServerName= or BootServerAddress= is sufficient,\nbut both can be set too, if desired.\n\nAdded in version 251.\n\nBootFilename=\nTakes a path or URL to a file loaded by e.g. a PXE boot loader. When specified, this path\nis sent in the DHCP option 67 (\"Bootfile name\"). See \u001b[34mRFC 2132[35] for more details.\nDefaults to unset.\n\nAdded in version 251.\n\nIPv6OnlyPreferredSec=\nTakes a timespan. Controls the \u001b[34mRFC 8925[28] IPv6-Only Preferred option. Specifies the\nDHCPv4 option to indicate that a host supports an IPv6-only mode and is willing to forgo\nobtaining an IPv4 address if the network provides IPv6 connectivity. Defaults to unset,\nand not send the option. The minimum allowed value is 300 seconds.\n\nAdded in version 255.\n\nSendOption=\nSend a raw option with value via DHCPv4 server. Takes a DHCP option number, data type and\ndata (\"option:type:value\"). The option number is an integer in the range 1...254. The\ntype takes one of \"uint8\", \"uint16\", \"uint32\", \"ipv4address\", \"ipv6address\", or \"string\".\nSpecial characters in the data string may be escaped using \u001b[34mC-style escapes[25]. This\nsetting can be specified multiple times. If an empty string is specified, then all\noptions specified earlier are cleared. Defaults to unset.\n\nAdded in version 244.\n\nSendVendorOption=\nSend a vendor option with value via DHCPv4 server. Takes a DHCP option number, data type\nand data (\"option:type:value\"). The option number is an integer in the range 1...254. The\ntype takes one of \"uint8\", \"uint16\", \"uint32\", \"ipv4address\", or \"string\". Special\ncharacters in the data string may be escaped using \u001b[34mC-style escapes[25]. This setting can\nbe specified multiple times. If an empty string is specified, then all options specified\nearlier are cleared. Defaults to unset.\n\nAdded in version 246.\n\nBindToInterface=\nTakes a boolean value. When \"yes\", DHCP server socket will be bound to its network\ninterface and all socket communication will be restricted to this interface. Defaults to\n\"yes\", except if RelayTarget= is used (see below), in which case it defaults to \"no\".\n\nAdded in version 249.\n\nRelayTarget=\nTakes an IPv4 address, which must be in the format described in inetpton(3). Turns this\nDHCP server into a DHCP relay agent. See \u001b[34mRFC 1542[36]. The address is the address of DHCP\nserver or another relay agent to forward DHCP messages to and from.\n\nAdded in version 249.\n\nRelayAgentCircuitId=\nSpecifies value for Agent Circuit ID suboption of Relay Agent Information option. Takes a\nstring, which must be in the format \"string:value\", where \"value\" should be replaced with\nthe value of the suboption. Defaults to unset (means no Agent Circuit ID suboption is\ngenerated). Ignored if RelayTarget= is not specified.\n\nAdded in version 249.\n\nRelayAgentRemoteId=\nSpecifies value for Agent Remote ID suboption of Relay Agent Information option. Takes a\nstring, which must be in the format \"string:value\", where \"value\" should be replaced with\nthe value of the suboption. Defaults to unset (means no Agent Remote ID suboption is\ngenerated). Ignored if RelayTarget= is not specified.\n\nAdded in version 249.\n\nRapidCommit=\nTakes a boolean. When true, the server supports \u001b[34mRFC 4039[37]. When a client sends a\nDHCPDISCOVER message with the Rapid Commit option to the server, then the server will\nreply with a DHCPACK message to the client, instead of DHCPOFFER. Defaults to true.\n\nAdded in version 255.\n\n[DHCPSERVERSTATICLEASE] SECTION OPTIONS\nThe \"[DHCPServerStaticLease]\" section configures a static DHCP lease to assign a fixed IPv4\naddress to a specific device based on its MAC address. This section can be specified multiple\ntimes.\n\nMACAddress=\nThe hardware address of a device to match. This key is mandatory.\n\nAdded in version 249.\n\nAddress=\nThe IPv4 address that should be assigned to the device that was matched with MACAddress=.\nThis key is mandatory.\n\nAdded in version 249.\n\n[IPV6SENDRA] SECTION OPTIONS\nThe [IPv6SendRA] section contains settings for sending IPv6 Router Advertisements and whether\nto act as a router, if enabled via the IPv6SendRA= option described above. IPv6 network\nprefixes or routes are defined with one or more [IPv6Prefix] or [IPv6RoutePrefix] sections.\n\nManaged=, OtherInformation=\nTakes a boolean. Controls whether a DHCPv6 server is used to acquire IPv6 addresses on\nthe network link when Managed= is set to \"true\" or if only additional network information\ncan be obtained via DHCPv6 for the network link when OtherInformation= is set to \"true\".\nBoth settings default to \"false\", which means that a DHCPv6 server is not being used.\n\nAdded in version 235.\n\nRouterLifetimeSec=\nTakes a timespan. Configures the IPv6 router lifetime in seconds. The value must be 0\nseconds, or between 4 seconds and 9000 seconds. When set to 0, the host is not acting as\na router. Defaults to 1800 seconds (30 minutes).\n\nAdded in version 235.\n\nRetransmitSec=\nTakes a timespan. Configures the retransmit time, used by clients to retransmit Neighbor\nSolicitation messages on address resolution and the Neighbor Unreachability Detection\nalgorithm. An integer the default unit of seconds, in the range 0...4294967295 msec.\nDefaults to 0.\n\nAdded in version 255.\n\nRouterPreference=\nConfigures IPv6 router preference if RouterLifetimeSec= is non-zero. Valid values are\n\"high\", \"medium\" and \"low\", with \"normal\" and \"default\" added as synonyms for \"medium\"\njust to make configuration easier. See \u001b[34mRFC 4191[21] for details. Defaults to \"medium\".\n\nAdded in version 235.\n\nHopLimit=\nConfigures hop limit. Takes an integer in the range 0...255. See also IPv6HopLimit=.\n\nAdded in version 255.\n\nUplinkInterface=\nSpecifies the name or the index of the uplink interface, or one of the special values\n\":none\" and \":auto\". When emitting DNS servers or search domains is enabled but no\nservers are specified, the servers configured in the uplink interface will be emitted.\nWhen \":auto\", the value specified to the same setting in the [DHCPPrefixDelegation]\nsection will be used if DHCPPrefixDelegation= is enabled, otherwise the link which has a\ndefault gateway with the highest priority will be automatically selected. When \":none\",\nno uplink interface will be selected. Defaults to \":auto\".\n\nAdded in version 250.\n\nEmitDNS=, DNS=\nDNS= specifies a list of recursive DNS server IPv6 addresses that are distributed via\nRouter Advertisement messages when EmitDNS= is true. DNS= also takes special value\n\"linklocal\"; in that case the IPv6 link-local address is distributed. If DNS= is empty,\nDNS servers are read from the [Network] section. If the [Network] section does not\ncontain any DNS servers either, DNS servers from the uplink interface specified in\nUplinkInterface= will be used. When EmitDNS= is false, no DNS server information is sent\nin Router Advertisement messages. EmitDNS= defaults to true.\n\nAdded in version 235.\n\nEmitDomains=, Domains=\nA list of DNS search domains distributed via Router Advertisement messages when\nEmitDomains= is true. If Domains= is empty, DNS search domains are read from the\n[Network] section. If the [Network] section does not contain any DNS search domains\neither, DNS search domains from the uplink interface specified in UplinkInterface= will\nbe used. When EmitDomains= is false, no DNS search domain information is sent in Router\nAdvertisement messages. EmitDomains= defaults to true.\n\nAdded in version 235.\n\nDNSLifetimeSec=\nLifetime in seconds for the DNS server addresses listed in DNS= and search domains listed\nin Domains=. Defaults to 3600 seconds (one hour).\n\nAdded in version 235.\n\nHomeAgent=\nTakes a boolean. Specifies that IPv6 router advertisements which indicates to hosts that\nthe router acts as a Home Agent and includes a Home Agent Option. Defaults to false. See\n\u001b[34mRFC 6275[10] for further details.\n\nAdded in version 255.\n\nHomeAgentLifetimeSec=\nTakes a timespan. Specifies the lifetime of the Home Agent. An integer the default unit\nof seconds, in the range 1...65535. Defaults to the value set to RouterLifetimeSec=.\n\nAdded in version 255.\n\nHomeAgentPreference=\nConfigures IPv6 Home Agent preference. Takes an integer in the range 0...65535. Defaults\nto 0.\n\nAdded in version 255.\n\n[IPV6PREFIX] SECTION OPTIONS\nOne or more [IPv6Prefix] sections contain the IPv6 prefixes that are announced via Router\nAdvertisements. See \u001b[34mRFC 4861[38] for further details.\n\nAddressAutoconfiguration=, OnLink=\nTakes a boolean to specify whether IPv6 addresses can be autoconfigured with this prefix\nand whether the prefix can be used for onlink determination. Both settings default to\n\"true\" in order to ease configuration.\n\nAdded in version 235.\n\nPrefix=\nThe IPv6 prefix that is to be distributed to hosts. Similarly to configuring static IPv6\naddresses, the setting is configured as an IPv6 prefix and its prefix length, separated\nby a \"/\" character. Use multiple [IPv6Prefix] sections to configure multiple IPv6\nprefixes since prefix lifetimes, address autoconfiguration and onlink status may differ\nfrom one prefix to another.\n\nAdded in version 235.\n\nPreferredLifetimeSec=, ValidLifetimeSec=\nPreferred and valid lifetimes for the prefix measured in seconds. PreferredLifetimeSec=\ndefaults to 1800 seconds (30 minutes) and ValidLifetimeSec= defaults to 3600 seconds (one\nhour).\n\nAdded in version 235.\n\nAssign=\nTakes a boolean. When true, adds an address from the prefix. Default to false.\n\nAdded in version 246.\n\nToken=\nSpecifies an optional address generation mode for assigning an address in each prefix.\nThis accepts the same syntax as Token= in the [IPv6AcceptRA] section. If Assign= is set\nto false, then this setting will be ignored. Defaults to unset, which means the EUI-64\nalgorithm will be used.\n\nAdded in version 250.\n\nRouteMetric=\nThe metric of the prefix route. Takes an unsigned integer in the range 0...4294967295.\nWhen unset or set to 0, the kernel's default value is used. This setting is ignored when\nAssign= is false.\n\nAdded in version 249.\n\n[IPV6ROUTEPREFIX] SECTION OPTIONS\nOne or more [IPv6RoutePrefix] sections contain the IPv6 prefix routes that are announced via\nRouter Advertisements. See \u001b[34mRFC 4191[21] for further details.\n\nRoute=\nThe IPv6 route that is to be distributed to hosts. Similarly to configuring static IPv6\nroutes, the setting is configured as an IPv6 prefix routes and its prefix route length,\nseparated by a \"/\" character. Use multiple [IPv6RoutePrefix] sections to configure\nmultiple IPv6 prefix routes.\n\nAdded in version 244.\n\nLifetimeSec=\nLifetime for the route prefix measured in seconds. LifetimeSec= defaults to 3600 seconds\n(one hour).\n\nAdded in version 244.\n\n[IPV6PREF64PREFIX] SECTION OPTIONS\nOne or more [IPv6PREF64Prefix] sections contain the IPv6 PREF64 (or NAT64) prefixes that are\nannounced via Router Advertisements. See \u001b[34mRFC 8781[33] for further details.\n\nPrefix=\nThe IPv6 PREF64 (or NAT64) prefix that is to be distributed to hosts. The setting holds\nan IPv6 prefix that should be set up for NAT64 translation (PLAT) to allow 464XLAT on the\nnetwork segment. Use multiple [IPv6PREF64Prefix] sections to configure multiple IPv6\nprefixes since prefix lifetime may differ from one prefix to another. The prefix is an\naddress with a prefix length, separated by a slash \"/\" character. Valid NAT64 prefix\nlength are 96, 64, 56, 48, 40, and 32 bits.\n\nAdded in version 255.\n\nLifetimeSec=\nLifetime for the prefix measured in seconds. Should be greater than or equal to\nRouterLifetimeSec=. LifetimeSec= defaults to 1800 seconds.\n\nAdded in version 255.\n\n[BRIDGE] SECTION OPTIONS\nThe [Bridge] section accepts the following keys:\n\nUnicastFlood=\nTakes a boolean. Controls whether the bridge should flood traffic for which an FDB entry\nis missing and the destination is unknown through this port. When unset, the kernel's\ndefault will be used.\n\nAdded in version 223.\n\nMulticastFlood=\nTakes a boolean. Controls whether the bridge should flood traffic for which an MDB entry\nis missing and the destination is unknown through this port. When unset, the kernel's\ndefault will be used.\n\nAdded in version 242.\n\nMulticastToUnicast=\nTakes a boolean. Multicast to unicast works on top of the multicast snooping feature of\nthe bridge. Which means unicast copies are only delivered to hosts which are interested\nin it. When unset, the kernel's default will be used.\n\nAdded in version 240.\n\nNeighborSuppression=\nTakes a boolean. Configures whether ARP and ND neighbor suppression is enabled for this\nport. When unset, the kernel's default will be used.\n\nAdded in version 242.\n\nLearning=\nTakes a boolean. Configures whether MAC address learning is enabled for this port. When\nunset, the kernel's default will be used.\n\nAdded in version 242.\n\nHairPin=\nTakes a boolean. Configures whether traffic may be sent back out of the port on which it\nwas received. When this flag is false, then the bridge will not forward traffic back out\nof the receiving port. When unset, the kernel's default will be used.\n\nAdded in version 223.\n\nIsolated=\nTakes a boolean. Configures whether this port is isolated or not. Within a bridge,\nisolated ports can only communicate with non-isolated ports. When set to true, this port\ncan only communicate with other ports whose Isolated setting is false. When set to false,\nthis port can communicate with any other ports. When unset, the kernel's default will be\nused.\n\nAdded in version 251.\n\nUseBPDU=\nTakes a boolean. Configures whether STP Bridge Protocol Data Units will be processed by\nthe bridge port. When unset, the kernel's default will be used.\n\nAdded in version 223.\n\nFastLeave=\nTakes a boolean. This flag allows the bridge to immediately stop multicast traffic on a\nport that receives an IGMP Leave message. It is only used with IGMP snooping if enabled\non the bridge. When unset, the kernel's default will be used.\n\nAdded in version 223.\n\nAllowPortToBeRoot=\nTakes a boolean. Configures whether a given port is allowed to become a root port. Only\nused when STP is enabled on the bridge. When unset, the kernel's default will be used.\n\nAdded in version 223.\n\nProxyARP=\nTakes a boolean. Configures whether proxy ARP to be enabled on this port. When unset, the\nkernel's default will be used.\n\nAdded in version 243.\n\nProxyARPWiFi=\nTakes a boolean. Configures whether proxy ARP to be enabled on this port which meets\nextended requirements by IEEE 802.11 and Hotspot 2.0 specifications. When unset, the\nkernel's default will be used.\n\nAdded in version 243.\n\nMulticastRouter=\nConfigures this port for having multicast routers attached. A port with a multicast\nrouter will receive all multicast traffic. Takes one of \"no\" to disable multicast routers\non this port, \"query\" to let the system detect the presence of routers, \"permanent\" to\npermanently enable multicast traffic forwarding on this port, or \"temporary\" to enable\nmulticast routers temporarily on this port, not depending on incoming queries. When\nunset, the kernel's default will be used.\n\nAdded in version 243.\n\nCost=\nSets the \"cost\" of sending packets of this interface. Each port in a bridge may have a\ndifferent speed and the cost is used to decide which link to use. Faster interfaces\nshould have lower costs. It is an integer value between 1 and 65535.\n\nAdded in version 218.\n\nPriority=\nSets the \"priority\" of sending packets on this interface. Each port in a bridge may have\na different priority which is used to decide which link to use. Lower value means higher\npriority. It is an integer value between 0 to 63. Networkd does not set any default,\nmeaning the kernel default value of 32 is used.\n\nAdded in version 234.\n\n[BRIDGEFDB] SECTION OPTIONS\nThe [BridgeFDB] section manages the forwarding database table of a port and accepts the\nfollowing keys. Specify several [BridgeFDB] sections to configure several static MAC table\nentries.\n\nMACAddress=\nAs in the [Network] section. This key is mandatory.\n\nAdded in version 219.\n\nDestination=\nTakes an IP address of the destination VXLAN tunnel endpoint.\n\nAdded in version 243.\n\nVLANId=\nThe VLAN ID for the new static MAC table entry. If omitted, no VLAN ID information is\nappended to the new static MAC table entry.\n\nAdded in version 219.\n\nVNI=\nThe VXLAN Network Identifier (or VXLAN Segment ID) to use to connect to the remote VXLAN\ntunnel endpoint. Takes a number in the range 1...16777215. Defaults to unset.\n\nAdded in version 243.\n\nAssociatedWith=\nSpecifies where the address is associated with. Takes one of \"use\", \"self\", \"master\" or\n\"router\". \"use\" means the address is in use. User space can use this option to indicate\nto the kernel that the fdb entry is in use. \"self\" means the address is associated with\nthe port drivers fdb. Usually hardware. \"master\" means the address is associated with\nmaster devices fdb. \"router\" means the destination address is associated with a router.\nNote that it's valid if the referenced device is a VXLAN type device and has route\nshortcircuit enabled. Defaults to \"self\".\n\nAdded in version 243.\n\nOutgoingInterface=\nSpecifies the name or index of the outgoing interface for the VXLAN device driver to\nreach the remote VXLAN tunnel endpoint. Defaults to unset.\n\nAdded in version 249.\n\n[BRIDGEMDB] SECTION OPTIONS\nThe [BridgeMDB] section manages the multicast membership entries forwarding database table of\na port and accepts the following keys. Specify several [BridgeMDB] sections to configure\nseveral permanent multicast membership entries.\n\nMulticastGroupAddress=\nSpecifies the IPv4 or IPv6 multicast group address to add. This setting is mandatory.\n\nAdded in version 247.\n\nVLANId=\nThe VLAN ID for the new entry. Valid ranges are 0 (no VLAN) to 4094. Optional, defaults\nto 0.\n\nAdded in version 247.\n\n[LLDP] SECTION OPTIONS\nThe [LLDP] section manages the Link Layer Discovery Protocol (LLDP) and accepts the following\nkeys:\n\nMUDURL=\nWhen configured, the specified Manufacturer Usage Descriptions (MUD) URL will be sent in\nLLDP packets. The syntax and semantics are the same as for MUDURL= in the [DHCPv4]\nsection described above.\n\nThe MUD URLs received via LLDP packets are saved and can be read using the\nsdlldpneighborgetmudurl() function.\n\nAdded in version 246.\n\n[CAN] SECTION OPTIONS\nThe [CAN] section manages the Controller Area Network (CAN bus) and accepts the following\nkeys:\n\nBitRate=\nThe bitrate of CAN device in bits per second. The usual SI prefixes (K, M) with the base\nof 1000 can be used here. Takes a number in the range 1...4294967295.\n\nAdded in version 239.\n\nSamplePoint=\nOptional sample point in percent with one decimal (e.g. \"75%\", \"87.5%\") or permille\n(e.g. \"875‰\"). This will be ignored when BitRate= is unspecified.\n\nAdded in version 239.\n\nTimeQuantaNSec=, PropagationSegment=, PhaseBufferSegment1=, PhaseBufferSegment2=,\nSyncJumpWidth=\nSpecifies the time quanta, propagation segment, phase buffer segment 1 and 2, and the\nsynchronization jump width, which allow one to define the CAN bit-timing in a hardware\nindependent format as proposed by the Bosch CAN 2.0 Specification. TimeQuantaNSec= takes\na timespan in nanoseconds. PropagationSegment=, PhaseBufferSegment1=,\nPhaseBufferSegment2=, and SyncJumpWidth= take number of time quantum specified in\nTimeQuantaNSec= and must be an unsigned integer in the range 0...4294967295. These\nsettings except for SyncJumpWidth= will be ignored when BitRate= is specified.\n\nAdded in version 250.\n\nDataBitRate=, DataSamplePoint=\nThe bitrate and sample point for the data phase, if CAN-FD is used. These settings are\nanalogous to the BitRate= and SamplePoint= keys.\n\nAdded in version 246.\n\nDataTimeQuantaNSec=, DataPropagationSegment=, DataPhaseBufferSegment1=,\nDataPhaseBufferSegment2=, DataSyncJumpWidth=\nSpecifies the time quanta, propagation segment, phase buffer segment 1 and 2, and the\nsynchronization jump width for the data phase, if CAN-FD is used. These settings are\nanalogous to the TimeQuantaNSec= or related settings.\n\nAdded in version 250.\n\nFDMode=\nTakes a boolean. When \"yes\", CAN-FD mode is enabled for the interface. Note, that a\nbitrate and optional sample point should also be set for the CAN-FD data phase using the\nDataBitRate= and DataSamplePoint= keys, or DataTimeQuanta= and related settings.\n\nAdded in version 246.\n\nFDNonISO=\nTakes a boolean. When \"yes\", non-ISO CAN-FD mode is enabled for the interface. When\nunset, the kernel's default will be used.\n\nAdded in version 246.\n\nRestartSec=\nAutomatic restart delay time. If set to a non-zero value, a restart of the CAN controller\nwill be triggered automatically in case of a bus-off condition after the specified delay\ntime. Subsecond delays can be specified using decimals (e.g. \"0.1s\") or a \"ms\" or \"us\"\npostfix. Using \"infinity\" or \"0\" will turn the automatic restart off. By default\nautomatic restart is disabled.\n\nAdded in version 239.\n\nTermination=\nTakes a boolean or a termination resistor value in ohm in the range 0...65535. When\n\"yes\", the termination resistor is set to 120 ohm. When \"no\" or \"0\" is set, the\ntermination resistor is disabled. When unset, the kernel's default will be used.\n\nAdded in version 246.\n\nTripleSampling=\nTakes a boolean. When \"yes\", three samples (instead of one) are used to determine the\nvalue of a received bit by majority rule. When unset, the kernel's default will be used.\n\nAdded in version 242.\n\nBusErrorReporting=\nTakes a boolean. When \"yes\", reporting of CAN bus errors is activated (those include\nsingle bit, frame format, and bit stuffing errors, unable to send dominant bit, unable to\nsend recessive bit, bus overload, active error announcement, error occurred on\ntransmission). When unset, the kernel's default will be used. Note: in case of a CAN bus\nwith a single CAN device, sending a CAN frame may result in a huge number of CAN bus\nerrors.\n\nAdded in version 248.\n\nListenOnly=\nTakes a boolean. When \"yes\", listen-only mode is enabled. When the interface is in\nlisten-only mode, the interface neither transmit CAN frames nor send ACK bit. Listen-only\nmode is important to debug CAN networks without interfering with the communication or\nacknowledge the CAN frame. When unset, the kernel's default will be used.\n\nAdded in version 246.\n\nLoopback=\nTakes a boolean. When \"yes\", loopback mode is enabled. When the loopback mode is enabled,\nthe interface treats messages transmitted by itself as received messages. The loopback\nmode is important to debug CAN networks. When unset, the kernel's default will be used.\n\nAdded in version 250.\n\nOneShot=\nTakes a boolean. When \"yes\", one-shot mode is enabled. When unset, the kernel's default\nwill be used.\n\nAdded in version 250.\n\nPresumeAck=\nTakes a boolean. When \"yes\", the interface will ignore missing CAN ACKs. When unset, the\nkernel's default will be used.\n\nAdded in version 250.\n\nClassicDataLengthCode=\nTakes a boolean. When \"yes\", the interface will handle the 4bit data length code (DLC).\nWhen unset, the kernel's default will be used.\n\nAdded in version 250.\n\n[IPOIB] SECTION OPTIONS\nThe [IPoIB] section manages the IP over Infiniband and accepts the following keys:\n\nMode=\nTakes one of the special values \"datagram\" or \"connected\". Defaults to unset, and the\nkernel's default is used.\n\nWhen \"datagram\", the Infiniband unreliable datagram (UD) transport is used, and so the\ninterface MTU is equal to the IB L2 MTU minus the IPoIB encapsulation header (4 bytes).\nFor example, in a typical IB fabric with a 2K MTU, the IPoIB MTU will be 2048 - 4 = 2044\nbytes.\n\nWhen \"connected\", the Infiniband reliable connected (RC) transport is used. Connected\nmode takes advantage of the connected nature of the IB transport and allows an MTU up to\nthe maximal IP packet size of 64K, which reduces the number of IP packets needed for\nhandling large UDP datagrams, TCP segments, etc and increases the performance for large\nmessages.\n\nAdded in version 250.\n\nIgnoreUserspaceMulticastGroup=\nTakes an boolean value. When true, the kernel ignores multicast groups handled by\nuserspace. Defaults to unset, and the kernel's default is used.\n\nAdded in version 250.\n\n[QDISC] SECTION OPTIONS\nThe [QDisc] section manages the traffic control queueing discipline (qdisc).\n\nParent=\nSpecifies the parent Queueing Discipline (qdisc). Takes one of \"clsact\" or \"ingress\".\nThis is mandatory.\n\nAdded in version 244.\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\n[NETWORKEMULATOR] SECTION OPTIONS\nThe [NetworkEmulator] section manages the queueing discipline (qdisc) of the network\nemulator. It can be used to configure the kernel packet scheduler and simulate packet delay\nand loss for UDP or TCP applications, or limit the bandwidth usage of a particular service to\nsimulate internet connections.\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nDelaySec=\nSpecifies the fixed amount of delay to be added to all packets going out of the\ninterface. Defaults to unset.\n\nAdded in version 245.\n\nDelayJitterSec=\nSpecifies the chosen delay to be added to the packets outgoing to the network interface.\nDefaults to unset.\n\nAdded in version 245.\n\nPacketLimit=\nSpecifies the maximum number of packets the qdisc may hold queued at a time. An unsigned\ninteger in the range 0...4294967294. Defaults to 1000.\n\nAdded in version 245.\n\nLossRate=\nSpecifies an independent loss probability to be added to the packets outgoing from the\nnetwork interface. Takes a percentage value, suffixed with \"%\". Defaults to unset.\n\nAdded in version 245.\n\nDuplicateRate=\nSpecifies that the chosen percent of packets is duplicated before queuing them. Takes a\npercentage value, suffixed with \"%\". Defaults to unset.\n\nAdded in version 245.\n\n[TOKENBUCKETFILTER] SECTION OPTIONS\nThe [TokenBucketFilter] section manages the queueing discipline (qdisc) of token bucket\nfilter (tbf).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nLatencySec=\nSpecifies the latency parameter, which specifies the maximum amount of time a packet can\nsit in the Token Bucket Filter (TBF). Defaults to unset.\n\nAdded in version 245.\n\nLimitBytes=\nTakes the number of bytes that can be queued waiting for tokens to become available. When\nthe size is suffixed with K, M, or G, it is parsed as Kilobytes, Megabytes, or Gigabytes,\nrespectively, to the base of 1024. Defaults to unset.\n\nAdded in version 246.\n\nBurstBytes=\nSpecifies the size of the bucket. This is the maximum amount of bytes that tokens can be\navailable for instantaneous transfer. When the size is suffixed with K, M, or G, it is\nparsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of 1024. Defaults\nto unset.\n\nAdded in version 246.\n\nRate=\nSpecifies the device specific bandwidth. When suffixed with K, M, or G, the specified\nbandwidth is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of\n1000. Defaults to unset.\n\nAdded in version 245.\n\nMPUBytes=\nThe Minimum Packet Unit (MPU) determines the minimal token usage (specified in bytes) for\na packet. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,\nMegabytes, or Gigabytes, respectively, to the base of 1024. Defaults to zero.\n\nAdded in version 245.\n\nPeakRate=\nTakes the maximum depletion rate of the bucket. When suffixed with K, M, or G, the\nspecified size is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of\n1000. Defaults to unset.\n\nAdded in version 245.\n\nMTUBytes=\nSpecifies the size of the peakrate bucket. When suffixed with K, M, or G, the specified\nsize is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of 1024.\nDefaults to unset.\n\nAdded in version 245.\n\n[PIE] SECTION OPTIONS\nThe [PIE] section manages the queueing discipline (qdisc) of Proportional Integral\ncontroller-Enhanced (PIE).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the queue size in number of packets. When this limit is\nreached, incoming packets are dropped. An unsigned integer in the range 1...4294967294.\nDefaults to unset and kernel's default is used.\n\nAdded in version 246.\n\n[FLOWQUEUEPIE] SECTION OPTIONS\nThe \"[FlowQueuePIE]\" section manages the queueing discipline (qdisc) of Flow Queue\nProportional Integral controller-Enhanced (fqpie).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the queue size in number of packets. When this limit is\nreached, incoming packets are dropped. An unsigned integer ranges 1 to 4294967294.\nDefaults to unset and kernel's default is used.\n\nAdded in version 247.\n\n[STOCHASTICFAIRBLUE] SECTION OPTIONS\nThe [StochasticFairBlue] section manages the queueing discipline (qdisc) of stochastic fair\nblue (sfb).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the queue size in number of packets. When this limit is\nreached, incoming packets are dropped. An unsigned integer in the range 0...4294967294.\nDefaults to unset and kernel's default is used.\n\nAdded in version 246.\n\n[STOCHASTICFAIRNESSQUEUEING] SECTION OPTIONS\nThe [StochasticFairnessQueueing] section manages the queueing discipline (qdisc) of\nstochastic fairness queueing (sfq).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPerturbPeriodSec=\nSpecifies the interval in seconds for queue algorithm perturbation. Defaults to unset.\n\nAdded in version 245.\n\n[BFIFO] SECTION OPTIONS\nThe [BFIFO] section manages the queueing discipline (qdisc) of Byte limited Packet First In\nFirst Out (bfifo).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nLimitBytes=\nSpecifies the hard limit in bytes on the FIFO buffer size. The size limit prevents\noverflow in case the kernel is unable to dequeue packets as quickly as it receives them.\nWhen this limit is reached, incoming packets are dropped. When suffixed with K, M, or G,\nthe specified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the\nbase of 1024. Defaults to unset and kernel default is used.\n\nAdded in version 246.\n\n[PFIFO] SECTION OPTIONS\nThe [PFIFO] section manages the queueing discipline (qdisc) of Packet First In First Out\n(pfifo).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the number of packets in the FIFO queue. The size limit\nprevents overflow in case the kernel is unable to dequeue packets as quickly as it\nreceives them. When this limit is reached, incoming packets are dropped. An unsigned\ninteger in the range 0...4294967294. Defaults to unset and kernel's default is used.\n\nAdded in version 246.\n\n[PFIFOHEADDROP] SECTION OPTIONS\nThe [PFIFOHeadDrop] section manages the queueing discipline (qdisc) of Packet First In First\nOut Head Drop (pfifoheaddrop).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nAs in [PFIFO] section.\n\nAdded in version 246.\n\n[PFIFOFAST] SECTION OPTIONS\nThe [PFIFOFast] section manages the queueing discipline (qdisc) of Packet First In First Out\nFast (pfifofast).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\n[CAKE] SECTION OPTIONS\nThe [CAKE] section manages the queueing discipline (qdisc) of Common Applications Kept\nEnhanced (CAKE).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nBandwidth=\nSpecifies the shaper bandwidth. When suffixed with K, M, or G, the specified size is\nparsed as Kilobits, Megabits, or Gigabits, respectively, to the base of 1000. Defaults to\nunset and kernel's default is used.\n\nAdded in version 246.\n\nAutoRateIngress=\nTakes a boolean value. Enables automatic capacity estimation based on traffic arriving at\nthis qdisc. This is most likely to be useful with cellular links, which tend to change\nquality randomly. If this setting is enabled, the Bandwidth= setting is used as an\ninitial estimate. Defaults to unset, and the kernel's default is used.\n\nAdded in version 250.\n\nOverheadBytes=\nSpecifies that bytes to be addeded to the size of each packet. Bytes may be negative.\nTakes an integer in the range -64...256. Defaults to unset and kernel's default is used.\n\nAdded in version 246.\n\nMPUBytes=\nRounds each packet (including overhead) up to the specified bytes. Takes an integer in\nthe range 1...256. Defaults to unset and kernel's default is used.\n\nAdded in version 250.\n\nCompensationMode=\nTakes one of \"none\", \"atm\", or \"ptm\". Specifies the compensation mode for overhead\ncalculation. When \"none\", no compensation is taken into account. When \"atm\", enables the\ncompensation for ATM cell framing, which is normally found on ADSL links. When \"ptm\",\nenables the compensation for PTM encoding, which is normally found on VDSL2 links and\nuses a 64b/65b encoding scheme. Defaults to unset and the kernel's default is used.\n\nAdded in version 250.\n\nUseRawPacketSize=\nTakes a boolean value. When true, the packet size reported by the Linux kernel will be\nused, instead of the underlying IP packet size. Defaults to unset, and the kernel's\ndefault is used.\n\nAdded in version 250.\n\nFlowIsolationMode=\nCAKE places packets from different flows into different queues, then packets from each\nqueue are delivered fairly. This specifies whether the fairness is based on source\naddress, destination address, individual flows, or any combination of those. The\navailable values are:\n\nnone\nThe flow isolation is disabled, and all traffic passes through a single queue.\n\nAdded in version 250.\n\nsrc-host\nFlows are defined only by source address. Equivalent to the \"srchost\" option for tc\nqdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\ndst-host\nFlows are defined only by destination address. Equivalent to the \"dsthost\" option for\ntc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\nhosts\nFlows are defined by source-destination host pairs. Equivalent to the same option for\ntc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\nflows\nFlows are defined by the entire 5-tuple of source address, destination address,\ntransport protocol, source port and destination port. Equivalent to the same option\nfor tc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\ndual-src-host\nFlows are defined by the 5-tuple (see \"flows\" in the above), and fairness is applied\nfirst over source addresses, then over individual flows. Equivalent to the\n\"dual-srchost\" option for tc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\ndual-dst-host\nFlows are defined by the 5-tuple (see \"flows\" in the above), and fairness is applied\nfirst over destination addresses, then over individual flows. Equivalent to the\n\"dual-dsthost\" option for tc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\ntriple\nFlows are defined by the 5-tuple (see \"flows\"), and fairness is applied over source\nand destination addresses, and also over individual flows. Equivalent to the\n\"triple-isolate\" option for tc qdisc command. See also tc-cake(8).\n\nAdded in version 250.\n\nDefaults to unset and the kernel's default is used.\n\nAdded in version 250.\n\nNAT=\nTakes a boolean value. When true, CAKE performs a NAT lookup before applying\nflow-isolation rules, to determine the true addresses and port numbers of the packet, to\nimprove fairness between hosts inside the NAT. This has no practical effect when\nFlowIsolationMode= is \"none\" or \"flows\", or if NAT is performed on a different host.\nDefaults to unset, and the kernel's default is used.\n\nAdded in version 250.\n\nPriorityQueueingPreset=\nCAKE divides traffic into \"tins\", and each tin has its own independent set of\nflow-isolation queues, bandwidth threshold, and priority. This specifies the preset of\ntin profiles. The available values are:\n\nbesteffort\nDisables priority queueing by placing all traffic in one tin.\n\nAdded in version 250.\n\nprecedence\nEnables priority queueing based on the legacy interpretation of TOS \"Precedence\"\nfield. Use of this preset on the modern Internet is firmly discouraged.\n\nAdded in version 250.\n\ndiffserv8\nEnables priority queueing based on the Differentiated Service (\"DiffServ\") field with\neight tins: Background Traffic, High Throughput, Best Effort, Video Streaming, Low\nLatency Transactions, Interactive Shell, Minimum Latency, and Network Control.\n\nAdded in version 250.\n\ndiffserv4\nEnables priority queueing based on the Differentiated Service (\"DiffServ\") field with\nfour tins: Background Traffic, Best Effort, Streaming Media, and Latency Sensitive.\n\nAdded in version 250.\n\ndiffserv3\nEnables priority queueing based on the Differentiated Service (\"DiffServ\") field with\nthree tins: Background Traffic, Best Effort, and Latency Sensitive.\n\nAdded in version 250.\n\nDefaults to unset, and the kernel's default is used.\n\nAdded in version 250.\n\nFirewallMark=\nTakes an integer in the range 1...4294967295. When specified, firewall-mark-based\noverriding of CAKE's tin selection is enabled. Defaults to unset, and the kernel's\ndefault is used.\n\nAdded in version 250.\n\nWash=\nTakes a boolean value. When true, CAKE clears the DSCP fields, except for ECN bits, of\nany packet passing through CAKE. Defaults to unset, and the kernel's default is used.\n\nAdded in version 250.\n\nSplitGSO=\nTakes a boolean value. When true, CAKE will split General Segmentation Offload (GSO)\nsuper-packets into their on-the-wire components and dequeue them individually. Defaults\nto unset, and the kernel's default is used.\n\nAdded in version 250.\n\nRTTSec=\nSpecifies the RTT for the filter. Takes a timespan. Typical values are e.g. 100us for\nextremely high-performance 10GigE+ networks like datacentre, 1ms for non-WiFi LAN\nconnections, 100ms for typical internet connections. Defaults to unset, and the kernel's\ndefault will be used.\n\nAdded in version 253.\n\nAckFilter=\nTakes a boolean value, or special value \"aggressive\". If enabled, ACKs in each flow are\nqueued and redundant ACKs to the upstream are dropped. If yes, the filter will always\nkeep at least two redundant ACKs in the queue, while in \"aggressive\" mode, it will filter\ndown to a single ACK. This may improve download throughput on links with very\nasymmetrical rate limits. Defaults to unset, and the kernel's default will be used.\n\nAdded in version 253.\n\n[CONTROLLEDDELAY] SECTION OPTIONS\nThe [ControlledDelay] section manages the queueing discipline (qdisc) of controlled delay\n(CoDel).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the queue size in number of packets. When this limit is\nreached, incoming packets are dropped. An unsigned integer in the range 0...4294967294.\nDefaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nTargetSec=\nTakes a timespan. Specifies the acceptable minimum standing/persistent queue delay.\nDefaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nIntervalSec=\nTakes a timespan. This is used to ensure that the measured minimum delay does not become\ntoo stale. Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nECN=\nTakes a boolean. This can be used to mark packets instead of dropping them. Defaults to\nunset and kernel's default is used.\n\nAdded in version 245.\n\nCEThresholdSec=\nTakes a timespan. This sets a threshold above which all packets are marked with ECN\nCongestion Experienced (CE). Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\n[DEFICITROUNDROBINSCHEDULER] SECTION OPTIONS\nThe [DeficitRoundRobinScheduler] section manages the queueing discipline (qdisc) of Deficit\nRound Robin Scheduler (DRR).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\n[DEFICITROUNDROBINSCHEDULERCLASS] SECTION OPTIONS\nThe [DeficitRoundRobinSchedulerClass] section manages the traffic control class of Deficit\nRound Robin Scheduler (DRR).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", or a qdisc\nidentifier. The qdisc identifier is specified as the major and minor numbers in\nhexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\"). Defaults to\n\"root\".\n\nClassId=\nConfigures the unique identifier of the class. It is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to unset.\n\nQuantumBytes=\nSpecifies the amount of bytes a flow is allowed to dequeue before the scheduler moves to\nthe next class. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,\nMegabytes, or Gigabytes, respectively, to the base of 1024. Defaults to the MTU of the\ninterface.\n\nAdded in version 246.\n\n[ENHANCEDTRANSMISSIONSELECTION] SECTION OPTIONS\nThe [EnhancedTransmissionSelection] section manages the queueing discipline (qdisc) of\nEnhanced Transmission Selection (ETS).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nBands=\nSpecifies the number of bands. An unsigned integer in the range 1...16. This value has to\nbe at least large enough to cover the strict bands specified through the StrictBands= and\nbandwidth-sharing bands specified in QuantumBytes=.\n\nAdded in version 246.\n\nStrictBands=\nSpecifies the number of bands that should be created in strict mode. An unsigned integer\nin the range 1...16.\n\nAdded in version 246.\n\nQuantumBytes=\nSpecifies the white-space separated list of quantum used in band-sharing bands. When\nsuffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or\nGigabytes, respectively, to the base of 1024. This setting can be specified multiple\ntimes. If an empty string is assigned, then the all previous assignments are cleared.\n\nAdded in version 246.\n\nPriorityMap=\nThe priority map maps the priority of a packet to a band. The argument is a whitespace\nseparated list of numbers. The first number indicates which band the packets with\npriority 0 should be put to, the second is for priority 1, and so on. There can be up to\n16 numbers in the list. If there are fewer, the default band that traffic with one of the\nunmentioned priorities goes to is the last one. Each band number must be in the range\n0...255. This setting can be specified multiple times. If an empty string is assigned,\nthen the all previous assignments are cleared.\n\nAdded in version 246.\n\n[GENERICRANDOMEARLYDETECTION] SECTION OPTIONS\nThe [GenericRandomEarlyDetection] section manages the queueing discipline (qdisc) of Generic\nRandom Early Detection (GRED).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nVirtualQueues=\nSpecifies the number of virtual queues. Takes an integer in the range 1...16. Defaults to\nunset and kernel's default is used.\n\nAdded in version 246.\n\nDefaultVirtualQueue=\nSpecifies the number of default virtual queue. This must be less than VirtualQueue=.\nDefaults to unset and kernel's default is used.\n\nAdded in version 246.\n\nGenericRIO=\nTakes a boolean. It turns on the RIO-like buffering scheme. Defaults to unset and\nkernel's default is used.\n\nAdded in version 246.\n\n[FAIRQUEUEINGCONTROLLEDDELAY] SECTION OPTIONS\nThe [FairQueueingControlledDelay] section manages the queueing discipline (qdisc) of fair\nqueuing controlled delay (FQ-CoDel).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the real queue size. When this limit is reached, incoming\npackets are dropped. Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nMemoryLimitBytes=\nSpecifies the limit on the total number of bytes that can be queued in this FQ-CoDel\ninstance. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,\nMegabytes, or Gigabytes, respectively, to the base of 1024. Defaults to unset and\nkernel's default is used.\n\nAdded in version 246.\n\nFlows=\nSpecifies the number of flows into which the incoming packets are classified. Defaults to\nunset and kernel's default is used.\n\nAdded in version 245.\n\nTargetSec=\nTakes a timespan. Specifies the acceptable minimum standing/persistent queue delay.\nDefaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nIntervalSec=\nTakes a timespan. This is used to ensure that the measured minimum delay does not become\ntoo stale. Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nQuantumBytes=\nSpecifies the number of bytes used as the \"deficit\" in the fair queuing algorithm\ntimespan. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,\nMegabytes, or Gigabytes, respectively, to the base of 1024. Defaults to unset and\nkernel's default is used.\n\nAdded in version 246.\n\nECN=\nTakes a boolean. This can be used to mark packets instead of dropping them. Defaults to\nunset and kernel's default is used.\n\nAdded in version 245.\n\nCEThresholdSec=\nTakes a timespan. This sets a threshold above which all packets are marked with ECN\nCongestion Experienced (CE). Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\n[FAIRQUEUEING] SECTION OPTIONS\nThe [FairQueueing] section manages the queueing discipline (qdisc) of fair queue traffic\npolicing (FQ).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the real queue size. When this limit is reached, incoming\npackets are dropped. Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nFlowLimit=\nSpecifies the hard limit on the maximum number of packets queued per flow. Defaults to\nunset and kernel's default is used.\n\nAdded in version 245.\n\nQuantumBytes=\nSpecifies the credit per dequeue RR round, i.e. the amount of bytes a flow is allowed to\ndequeue at once. When suffixed with K, M, or G, the specified size is parsed as\nKilobytes, Megabytes, or Gigabytes, respectively, to the base of 1024. Defaults to unset\nand kernel's default is used.\n\nAdded in version 246.\n\nInitialQuantumBytes=\nSpecifies the initial sending rate credit, i.e. the amount of bytes a new flow is allowed\nto dequeue initially. When suffixed with K, M, or G, the specified size is parsed as\nKilobytes, Megabytes, or Gigabytes, respectively, to the base of 1024. Defaults to unset\nand kernel's default is used.\n\nAdded in version 245.\n\nMaximumRate=\nSpecifies the maximum sending rate of a flow. When suffixed with K, M, or G, the\nspecified size is parsed as Kilobits, Megabits, or Gigabits, respectively, to the base of\n1000. Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\nBuckets=\nSpecifies the size of the hash table used for flow lookups. Defaults to unset and\nkernel's default is used.\n\nAdded in version 245.\n\nOrphanMask=\nTakes an unsigned integer. For packets not owned by a socket, fq is able to mask a part\nof hash and reduce number of buckets associated with the traffic. Defaults to unset and\nkernel's default is used.\n\nAdded in version 245.\n\nPacing=\nTakes a boolean, and enables or disables flow pacing. Defaults to unset and kernel's\ndefault is used.\n\nAdded in version 245.\n\nCEThresholdSec=\nTakes a timespan. This sets a threshold above which all packets are marked with ECN\nCongestion Experienced (CE). Defaults to unset and kernel's default is used.\n\nAdded in version 245.\n\n[TRIVIALLINKEQUALIZER] SECTION OPTIONS\nThe [TrivialLinkEqualizer] section manages the queueing discipline (qdisc) of trivial link\nequalizer (teql).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nId=\nSpecifies the interface ID \"N\" of teql. Defaults to \"0\". Note that when teql is used,\ncurrently, the module schteql with maxequalizers=N+1 option must be loaded before\nsystemd-networkd is started.\n\nAdded in version 245.\n\n[HIERARCHYTOKENBUCKET] SECTION OPTIONS\nThe [HierarchyTokenBucket] section manages the queueing discipline (qdisc) of hierarchy token\nbucket (htb).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nDefaultClass=\nTakes the minor id in hexadecimal of the default class. Unclassified traffic gets sent to\nthe class. Defaults to unset.\n\nAdded in version 246.\n\nRateToQuantum=\nTakes an unsigned integer. The DRR quantums are calculated by dividing the value\nconfigured in Rate= by RateToQuantum=.\n\nAdded in version 246.\n\n[HIERARCHYTOKENBUCKETCLASS] SECTION OPTIONS\nThe [HierarchyTokenBucketClass] section manages the traffic control class of hierarchy token\nbucket (htb).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", or a qdisc\nidentifier. The qdisc identifier is specified as the major and minor numbers in\nhexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\"). Defaults to\n\"root\".\n\nClassId=\nConfigures the unique identifier of the class. It is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to unset.\n\nPriority=\nSpecifies the priority of the class. In the round-robin process, classes with the lowest\npriority field are tried for packets first.\n\nAdded in version 246.\n\nQuantumBytes=\nSpecifies how many bytes to serve from leaf at once. When suffixed with K, M, or G, the\nspecified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base\nof 1024.\n\nAdded in version 246.\n\nMTUBytes=\nSpecifies the maximum packet size we create. When suffixed with K, M, or G, the specified\nsize is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the base of 1024.\n\nAdded in version 246.\n\nOverheadBytes=\nTakes an unsigned integer which specifies per-packet size overhead used in rate\ncomputations. When suffixed with K, M, or G, the specified size is parsed as Kilobytes,\nMegabytes, or Gigabytes, respectively, to the base of 1024.\n\nAdded in version 246.\n\nRate=\nSpecifies the maximum rate this class and all its children are guaranteed. When suffixed\nwith K, M, or G, the specified size is parsed as Kilobits, Megabits, or Gigabits,\nrespectively, to the base of 1000. This setting is mandatory.\n\nAdded in version 246.\n\nCeilRate=\nSpecifies the maximum rate at which a class can send, if its parent has bandwidth to\nspare. When suffixed with K, M, or G, the specified size is parsed as Kilobits, Megabits,\nor Gigabits, respectively, to the base of 1000. When unset, the value specified with\nRate= is used.\n\nAdded in version 246.\n\nBufferBytes=\nSpecifies the maximum bytes burst which can be accumulated during idle period. When\nsuffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or\nGigabytes, respectively, to the base of 1024.\n\nAdded in version 246.\n\nCeilBufferBytes=\nSpecifies the maximum bytes burst for ceil which can be accumulated during idle period.\nWhen suffixed with K, M, or G, the specified size is parsed as Kilobytes, Megabytes, or\nGigabytes, respectively, to the base of 1024.\n\nAdded in version 246.\n\n[HEAVYHITTERFILTER] SECTION OPTIONS\nThe [HeavyHitterFilter] section manages the queueing discipline (qdisc) of Heavy Hitter\nFilter (hhf).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\nPacketLimit=\nSpecifies the hard limit on the queue size in number of packets. When this limit is\nreached, incoming packets are dropped. An unsigned integer in the range 0...4294967294.\nDefaults to unset and kernel's default is used.\n\nAdded in version 246.\n\n[QUICKFAIRQUEUEING] SECTION OPTIONS\nThe [QuickFairQueueing] section manages the queueing discipline (qdisc) of Quick Fair\nQueueing (QFQ).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", \"clsact\",\n\"ingress\" or a class identifier. The class identifier is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to \"root\".\n\nHandle=\nConfigures the major number of unique identifier of the qdisc, known as the handle. Takes\na hexadecimal number in the range 0x1–0xffff. Defaults to unset.\n\n[QUICKFAIRQUEUEINGCLASS] SECTION OPTIONS\nThe [QuickFairQueueingClass] section manages the traffic control class of Quick Fair Queueing\n(qfq).\n\nParent=\nConfigures the parent Queueing Discipline (qdisc). Takes one of \"root\", or a qdisc\nidentifier. The qdisc identifier is specified as the major and minor numbers in\nhexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\"). Defaults to\n\"root\".\n\nClassId=\nConfigures the unique identifier of the class. It is specified as the major and minor\nnumbers in hexadecimal in the range 0x1–0xffff separated with a colon (\"major:minor\").\nDefaults to unset.\n\nWeight=\nSpecifies the weight of the class. Takes an integer in the range 1...1023. Defaults to\nunset in which case the kernel default is used.\n\nAdded in version 246.\n\nMaxPacketBytes=\nSpecifies the maximum packet size in bytes for the class. When suffixed with K, M, or G,\nthe specified size is parsed as Kilobytes, Megabytes, or Gigabytes, respectively, to the\nbase of 1024. When unset, the kernel default is used.\n\nAdded in version 246.\n\n[BRIDGEVLAN] SECTION OPTIONS\nThe [BridgeVLAN] section manages the VLAN ID configuration of a bridge port and accepts the\nfollowing keys. Specify several [BridgeVLAN] sections to configure several VLAN entries. The\nVLANFiltering= option has to be enabled, see the [Bridge] section in systemd.netdev(5).\n\nVLAN=\nThe VLAN ID allowed on the port. This can be either a single ID or a range M-N. Takes an\ninteger in the range 1...4094.\n\nAdded in version 231.\n\nEgressUntagged=\nThe VLAN ID specified here will be used to untag frames on egress. Configuring\nEgressUntagged= implicates the use of VLAN= above and will enable the VLAN ID for ingress\nas well. This can be either a single ID or a range M-N.\n\nAdded in version 231.\n\nPVID=\nThe Port VLAN ID specified here is assigned to all untagged frames at ingress. PVID= can\nbe used only once. Configuring PVID= implicates the use of VLAN= above and will enable\nthe VLAN ID for ingress as well.\n\nAdded in version 231.\n", "subsections": [] }, "EXAMPLES": { "content": "", "subsections": [ { "name": "Example 1. Static network configuration", "content": "# /etc/systemd/network/50-static.network\n[Match]\nName=enp2s0\n\n[Network]\nAddress=192.168.0.15/24\nGateway=192.168.0.1\n\nThis brings interface \"enp2s0\" up with a static address. The specified gateway will be used\nfor a default route.\n" }, { "name": "Example 2. DHCP on ethernet links", "content": "# /etc/systemd/network/80-dhcp.network\n[Match]\nName=en*\n\n[Network]\nDHCP=yes\n\nThis will enable DHCPv4 and DHCPv6 on all interfaces with names starting with \"en\" (i.e.\nethernet interfaces).\n" }, { "name": "Example 3. IPv6 Prefix Delegation (DHCPv6 PD)", "content": "# /etc/systemd/network/55-dhcpv6-pd-upstream.network\n[Match]\nName=enp1s0\n\n[Network]\nDHCP=ipv6\n\n# The below setting is optional, to also assign an address in the delegated prefix\n# to the upstream interface. If not necessary, then comment out the line below and\n# the [DHCPPrefixDelegation] section.\nDHCPPrefixDelegation=yes\n\n# If the upstream network provides Router Advertisement with Managed bit set,\n# then comment out the line below and WithoutRA= setting in the [DHCPv6] section.\nIPv6AcceptRA=no\n\n[DHCPv6]\nWithoutRA=solicit\n\n[DHCPPrefixDelegation]\nUplinkInterface=:self\nSubnetId=0\nAnnounce=no\n\n# /etc/systemd/network/55-dhcpv6-pd-downstream.network\n[Match]\nName=enp2s0\n\n[Network]\nDHCPPrefixDelegation=yes\nIPv6SendRA=yes\n\n# It is expected that the host is acting as a router. So, usually it is not\n# necessary to receive Router Advertisement from other hosts in the downstream network.\nIPv6AcceptRA=no\n\n[DHCPPrefixDelegation]\nUplinkInterface=enp1s0\nSubnetId=1\nAnnounce=yes\n\nThis will enable DHCPv6-PD on the interface enp1s0 as an upstream interface where the DHCPv6\nclient is running and enp2s0 as a downstream interface where the prefix is delegated to. The\ndelegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.\n" }, { "name": "Example 4. IPv6 Prefix Delegation (DHCPv4 6RD)", "content": "# /etc/systemd/network/55-dhcpv4-6rd-upstream.network\n[Match]\nName=enp1s0\n\n[Network]\nDHCP=ipv4\n\n# When DHCPv4-6RD is used, the upstream network does not support IPv6.\n# Hence, it is not necessary to wait for Router Advertisement, which is enabled by default.\nIPv6AcceptRA=no\n\n[DHCPv4]\nUse6RD=yes\n\n# /etc/systemd/network/55-dhcpv4-6rd-downstream.network\n[Match]\nName=enp2s0\n\n[Network]\nDHCPPrefixDelegation=yes\nIPv6SendRA=yes\n\n# It is expected that the host is acting as a router. So, usually it is not\n# necessary to receive Router Advertisement from other hosts in the downstream network.\nIPv6AcceptRA=no\n\n[DHCPPrefixDelegation]\nUplinkInterface=enp1s0\nSubnetId=1\nAnnounce=yes\n\nThis will enable DHCPv4-6RD on the interface enp1s0 as an upstream interface where the DHCPv4\nclient is running and enp2s0 as a downstream interface where the prefix is delegated to. The\ndelegated prefixes are distributed by IPv6 Router Advertisement on the downstream network.\n" }, { "name": "Example 5. A bridge with two enslaved links", "content": "# /etc/systemd/network/25-bridge-static.netdev\n[NetDev]\nName=bridge0\nKind=bridge\n\n# /etc/systemd/network/25-bridge-static.network\n[Match]\nName=bridge0\n\n[Network]\nAddress=192.168.0.15/24\nGateway=192.168.0.1\nDNS=192.168.0.1\n\n# /etc/systemd/network/25-bridge-slave-interface-1.network\n[Match]\nName=enp2s0\n\n[Network]\nBridge=bridge0\n\n# /etc/systemd/network/25-bridge-slave-interface-2.network\n[Match]\nName=wlp3s0\n\n[Network]\nBridge=bridge0\n\nThis creates a bridge and attaches devices \"enp2s0\" and \"wlp3s0\" to it. The bridge will have\nthe specified static address and network assigned, and a default route via the specified\ngateway will be added. The specified DNS server will be added to the global list of DNS\nresolvers.\n" }, { "name": "Example 6. Bridge port with VLAN forwarding", "content": "# /etc/systemd/network/25-bridge-slave-interface-1.network\n[Match]\nName=enp2s0\n\n[Network]\nBridge=bridge0\n\n[BridgeVLAN]\nVLAN=1-32\nPVID=42\nEgressUntagged=42\n\n[BridgeVLAN]\nVLAN=100-200\n\n[BridgeVLAN]\nEgressUntagged=300-400\n\nThis overrides the configuration specified in the previous example for the interface\n\"enp2s0\", and enables VLAN on that bridge port. VLAN IDs 1-32, 42, 100-400 will be allowed.\nPackets tagged with VLAN IDs 42, 300-400 will be untagged when they leave on this interface.\nUntagged packets which arrive on this interface will be assigned VLAN ID 42.\n" }, { "name": "Example 7. Various tunnels", "content": "/etc/systemd/network/25-tunnels.network\n[Match]\nName=ens1\n\n[Network]\nTunnel=ipip-tun\nTunnel=sit-tun\nTunnel=gre-tun\nTunnel=vti-tun\n\n\n/etc/systemd/network/25-tunnel-ipip.netdev\n[NetDev]\nName=ipip-tun\nKind=ipip\n\n\n/etc/systemd/network/25-tunnel-sit.netdev\n[NetDev]\nName=sit-tun\nKind=sit\n\n\n/etc/systemd/network/25-tunnel-gre.netdev\n[NetDev]\nName=gre-tun\nKind=gre\n\n\n/etc/systemd/network/25-tunnel-vti.netdev\n[NetDev]\nName=vti-tun\nKind=vti\n\n\nThis will bring interface \"ens1\" up and create an IPIP tunnel, a SIT tunnel, a GRE tunnel,\nand a VTI tunnel using it.\n" }, { "name": "Example 8. A bond device", "content": "# /etc/systemd/network/30-bond1.network\n[Match]\nName=bond1\n\n[Network]\nDHCP=ipv6\n\n# /etc/systemd/network/30-bond1.netdev\n[NetDev]\nName=bond1\nKind=bond\n\n# /etc/systemd/network/30-bond1-dev1.network\n[Match]\nMACAddress=52:54:00:e9:64:41\n\n[Network]\nBond=bond1\n\n# /etc/systemd/network/30-bond1-dev2.network\n[Match]\nMACAddress=52:54:00:e9:64:42\n\n[Network]\nBond=bond1\n\nThis will create a bond device \"bond1\" and enslave the two devices with MAC addresses\n52:54:00:e9:64:41 and 52:54:00:e9:64:42 to it. IPv6 DHCP will be used to acquire an address.\n" }, { "name": "Example 9. Virtual Routing and Forwarding (VRF)", "content": "Add the \"bond1\" interface to the VRF master interface \"vrf1\". This will redirect routes\ngenerated on this interface to be within the routing table defined during VRF creation. For\nkernels before 4.8 traffic won't be redirected towards the VRFs routing table unless specific\nip-rules are added.\n\n# /etc/systemd/network/25-vrf.network\n[Match]\nName=bond1\n\n[Network]\nVRF=vrf1\n" }, { "name": "Example 10. MacVTap", "content": "This brings up a network interface \"macvtap-test\" and attaches it to \"enp0s25\".\n\n# /usr/lib/systemd/network/25-macvtap.network\n[Match]\nName=enp0s25\n\n[Network]\nMACVTAP=macvtap-test\n" }, { "name": "Example 11. A Xfrm interface with physical underlying device.", "content": "# /etc/systemd/network/27-xfrm.netdev\n[NetDev]\nName=xfrm0\nKind=xfrm\n\n[Xfrm]\nInterfaceId=7\n\n# /etc/systemd/network/27-eth0.network\n[Match]\nName=eth0\n\n[Network]\nXfrm=xfrm0\n\nThis creates a \"xfrm0\" interface and binds it to the \"eth0\" device. This allows hardware\nbased ipsec offloading to the \"eth0\" nic. If offloading is not needed, xfrm interfaces can be\nassigned to the \"lo\" device.\n" } ] }, "SEE ALSO": { "content": "systemd(1), systemd-networkd.service(8), systemd.link(5), systemd.netdev(5), systemd-network-\ngenerator.service(8), systemd-resolved.service(8)\n", "subsections": [] }, "NOTES": { "content": "1. System and Service Credentials\nhttps://systemd.io/CREDENTIALS\n\n2. Link-Local Multicast Name Resolution\nhttps://tools.ietf.org/html/rfc4795\n\n3. Multicast DNS\nhttps://tools.ietf.org/html/rfc6762\n\n4. DNS-over-TLS\nhttps://tools.ietf.org/html/rfc7858\n\n5. DNSSEC\nhttps://tools.ietf.org/html/rfc4033\n\n6. IEEE 802.1AB-2016\nhttps://standards.ieee.org/findstds/standard/802.1AB-2016.html\n\n7. IP Sysctl\nhttps://docs.kernel.org/networking/ip-sysctl.html\n\n8. RFC 4941\nhttps://tools.ietf.org/html/rfc4941\n\n9. RFC 3704\nhttps://tools.ietf.org/html/rfc1027\n\n10. RFC 6275\nhttps://tools.ietf.org/html/rfc6275\n\n11. RFC 5227\nhttps://tools.ietf.org/html/rfc5227\n\n12. RFC 4862\nhttps://tools.ietf.org/html/rfc4862\n\n13. RFC 3041\nhttps://tools.ietf.org/html/rfc3041\n\n14. NetLabel\nhttps://docs.kernel.org/netlabel/index.html\n\n15. Linux Security Modules (LSMs)\nhttps://en.wikipedia.org/wiki/LinuxSecurityModules\n\n16. NetLabel Fallback Peer Labeling\nhttps://github.com/SELinuxProject/selinux-notebook/blob/main/src/networksupport.md\n\n17. NFT\nhttps://netfilter.org/projects/nftables/index.html\n\n18. RFC 3484\nhttps://tools.ietf.org/html/rfc3484\n\n19. Type of Service\nhttps://en.wikipedia.org/wiki/Typeofservice\n\n20. Differentiated services\nhttps://en.wikipedia.org/wiki/Differentiatedservices\n\n21. RFC 4191\nhttps://tools.ietf.org/html/rfc4191\n\n22. RFC 8520\nhttps://tools.ietf.org/html/rfc8520\n\n23. RFC 4039\nhttps://tools.ietf.org/html/rfc4039\n\n24. RFC 7844\nhttps://tools.ietf.org/html/rfc7844\n\n25. C-style escapes\nhttps://en.wikipedia.org/wiki/EscapesequencesinC#Tableofescapesequences\n\n26. RFC 3442\nhttps://datatracker.ietf.org/doc/html/rfc3442\n\n27. RFC 5969\nhttps://tools.ietf.org/html/rfc5969\n\n28. RFC 8925\nhttps://tools.ietf.org/html/rfc8925\n\n29. RFC 3315\nhttps://tools.ietf.org/html/rfc3315#section-17.2.1\n\n30. RFC 8415\nhttps://www.rfc-editor.org/rfc/rfc8415.html#section-6.3\n\n31. RFC 4291\nhttps://tools.ietf.org/html/rfc4291#section-2.5.4\n\n32. RFC 7217\nhttps://tools.ietf.org/html/rfc7217\n\n33. RFC 8781\nhttps://tools.ietf.org/html/rfc8781\n\n34. RFC 2131\nhttps://www.rfc-editor.org/rfc/rfc2131.html\n\n35. RFC 2132\nhttps://www.rfc-editor.org/rfc/rfc2132.html\n\n36. RFC 1542\nhttps://tools.ietf.org/html/rfc1542\n\n37. RFC 4039\nhttps://datatracker.ietf.org/doc/html/rfc4039\n\n38. RFC 4861\nhttps://tools.ietf.org/html/rfc4861\n\nsystemd 255 SYSTEMD.NETWORK(5)", "subsections": [] } } } }