path: root/man7/rtnetlink.7

'\" t
.\" SPDX-License-Identifier: Linux-man-pages-1-para
.\"
.\" This man page is Copyright (C) 1999 Andi Kleen <ak@muc.de>.
.\"
.\" Based on the original comments from Alexey Kuznetsov, written with
.\" help from Matthew Wilcox.
.\" $Id: rtnetlink.7,v 1.8 2000/01/22 01:55:04 freitag Exp $
.\"
.TH rtnetlink 7 2023-07-15 "Linux man-pages 6.05.01"
.SH NAME
rtnetlink \- Linux routing socket
.SH SYNOPSIS
.nf
.B #include <asm/types.h>
.B #include <linux/netlink.h>
.B #include <linux/rtnetlink.h>
.B #include <sys/socket.h>
.PP
.BI "rtnetlink_socket = socket(AF_NETLINK, int " socket_type ", NETLINK_ROUTE);"
.fi
.SH DESCRIPTION
Rtnetlink allows the kernel's routing tables to be read and altered.
It is used within the kernel to communicate between
various subsystems, though this usage is not documented here, and for
communication with user-space programs.
Network routes, IP addresses, link parameters, neighbor setups, queueing
disciplines, traffic classes and packet classifiers may all be controlled
through
.B NETLINK_ROUTE
sockets.
It is based on netlink messages; see
.BR netlink (7)
for more information.
.\" FIXME . ? all these macros could be moved to rtnetlink(3)
.SS Routing attributes
Some rtnetlink messages have optional attributes after the initial header:
.PP
.in +4n
.EX
struct rtattr {
    unsigned short rta_len;    /* Length of option */
    unsigned short rta_type;   /* Type of option */
    /* Data follows */
};
.EE
.in
.PP
These attributes should be manipulated using only the RTA_* macros
or libnetlink, see
.BR rtnetlink (3).
.SS Messages
Rtnetlink consists of these message types
(in addition to standard netlink messages):
.TP
.BR RTM_NEWLINK ", " RTM_DELLINK ", " RTM_GETLINK
Create, remove, or get information about a specific network interface.
These messages contain an
.I ifinfomsg
structure followed by a series of
.I rtattr
structures.
.IP
.EX
struct ifinfomsg {
    unsigned char  ifi_family; /* AF_UNSPEC */
    unsigned short ifi_type;   /* Device type */
    int            ifi_index;  /* Interface index */
    unsigned int   ifi_flags;  /* Device flags  */
    unsigned int   ifi_change; /* change mask */
};
.EE
.IP
.\" FIXME Document ifinfomsg.ifi_type
.I ifi_flags
contains the device flags, see
.BR netdevice (7);
.I ifi_index
is the unique interface index
(since Linux 3.7, it is possible to feed a nonzero value with the
.B RTM_NEWLINK
message, thus creating a link with the given
.IR ifindex );
.I ifi_change
is reserved for future use and should be always set to 0xFFFFFFFF.
.TS
tab(:);
c s s
lb l l.
Routing attributes
rta_type:Value type:Description
_
IFLA_UNSPEC:-:unspecified
IFLA_ADDRESS:hardware address:interface L2 address
IFLA_BROADCAST:hardware address:L2 broadcast address
IFLA_IFNAME:asciiz string:Device name
IFLA_MTU:unsigned int:MTU of the device
IFLA_LINK:int:Link type
IFLA_QDISC:asciiz string:Queueing discipline
IFLA_STATS:T{
see below
T}:Interface Statistics
IFLA_PERM_ADDRESS:hardware address:T{
hardware address provided by device (since Linux 5.5)
T}
.TE
.IP
The value type for
.B IFLA_STATS
is
.I struct rtnl_link_stats
.RI ( "struct net_device_stats"
in Linux 2.4 and earlier).
.TP
.BR RTM_NEWADDR ", " RTM_DELADDR ", " RTM_GETADDR
Add, remove, or receive information about an IP address associated with
an interface.
In Linux 2.2, an interface can carry multiple IP addresses,
this replaces the alias device concept in Linux 2.0.
In Linux 2.2, these messages
support IPv4 and IPv6 addresses.
They contain an
.I ifaddrmsg
structure, optionally followed by
.I rtattr
routing attributes.
.IP
.EX
struct ifaddrmsg {
    unsigned char ifa_family;    /* Address type */
    unsigned char ifa_prefixlen; /* Prefixlength of address */
    unsigned char ifa_flags;     /* Address flags */
    unsigned char ifa_scope;     /* Address scope */
    unsigned int  ifa_index;     /* Interface index */
};
.EE
.IP
.I ifa_family
is the address family type (currently
.B AF_INET
or
.BR AF_INET6 ),
.I ifa_prefixlen
is the length of the address mask of the address if defined for the
family (like for IPv4),
.I ifa_scope
is the address scope,
.I ifa_index
is the interface index of the interface the address is associated with.
.I ifa_flags
is a flag word of
.B IFA_F_SECONDARY
for secondary address (old alias interface),
.B IFA_F_PERMANENT
for a permanent address set by the user and other undocumented flags.
.TS
tab(:);
c s s
lb l l.
Attributes
rta_type:Value type:Description
_
IFA_UNSPEC:-:unspecified
IFA_ADDRESS:raw protocol address:interface address
IFA_LOCAL:raw protocol address:local address
IFA_LABEL:asciiz string:name of the interface
IFA_BROADCAST:raw protocol address:broadcast address
IFA_ANYCAST:raw protocol address:anycast address
IFA_CACHEINFO:struct ifa_cacheinfo:Address information
.TE
.\" FIXME Document struct ifa_cacheinfo
.TP
.BR RTM_NEWROUTE ", " RTM_DELROUTE ", " RTM_GETROUTE
Create, remove, or receive information about a network route.
These messages contain an
.I rtmsg
structure with an optional sequence of
.I rtattr
structures following.
For
.BR RTM_GETROUTE ,
setting
.I rtm_dst_len
and
.I rtm_src_len
to 0 means you get all entries for the specified routing table.
For the other fields, except
.I rtm_table
and
.IR rtm_protocol ,
0 is the wildcard.
.IP
.EX
struct rtmsg {
    unsigned char rtm_family;   /* Address family of route */
    unsigned char rtm_dst_len;  /* Length of destination */
    unsigned char rtm_src_len;  /* Length of source */
    unsigned char rtm_tos;      /* TOS filter */
    unsigned char rtm_table;    /* Routing table ID;
                                   see RTA_TABLE below */
    unsigned char rtm_protocol; /* Routing protocol; see below */
    unsigned char rtm_scope;    /* See below */
    unsigned char rtm_type;     /* See below */
\&
    unsigned int  rtm_flags;
};
.EE
.TS
tab(:);
lb l.
rtm_type:Route type
_
RTN_UNSPEC:unknown route
RTN_UNICAST:a gateway or direct route
RTN_LOCAL:a local interface route
RTN_BROADCAST:T{
a local broadcast route (sent as a broadcast)
T}
RTN_ANYCAST:T{
a local broadcast route (sent as a unicast)
T}
RTN_MULTICAST:a multicast route
RTN_BLACKHOLE:a packet dropping route
RTN_UNREACHABLE:an unreachable destination
RTN_PROHIBIT:a packet rejection route
RTN_THROW:continue routing lookup in another table
RTN_NAT:a network address translation rule
RTN_XRESOLVE:T{
refer to an external resolver (not implemented)
T}
.TE
.TS
tab(:);
lb l.
rtm_protocol:Route origin
_
RTPROT_UNSPEC:unknown
RTPROT_REDIRECT:T{
by an ICMP redirect (currently unused)
T}
RTPROT_KERNEL:by the kernel
RTPROT_BOOT:during boot
RTPROT_STATIC:by the administrator
.TE
.sp 1
Values larger than
.B RTPROT_STATIC
are not interpreted by the kernel, they are just for user information.
They may be used to tag the source of a routing information or to
distinguish between multiple routing daemons.
See
.I <linux/rtnetlink.h>
for the routing daemon identifiers which are already assigned.
.IP
.I rtm_scope
is the distance to the destination:
.TS
tab(:);
lb l.
RT_SCOPE_UNIVERSE:global route
RT_SCOPE_SITE:T{
interior route in the local autonomous system
T}
RT_SCOPE_LINK:route on this link
RT_SCOPE_HOST:route on the local host
RT_SCOPE_NOWHERE:destination doesn't exist
.TE
.sp 1
The values between
.B RT_SCOPE_UNIVERSE
and
.B RT_SCOPE_SITE
are available to the user.
.IP
The
.I rtm_flags
have the following meanings:
.TS
tab(:);
lb l.
RTM_F_NOTIFY:T{
if the route changes, notify the user via rtnetlink
T}
RTM_F_CLONED:route is cloned from another route
RTM_F_EQUALIZE:a multipath equalizer (not yet implemented)
.TE
.sp 1
.I rtm_table
specifies the routing table
.TS
tab(:);
lb l.
RT_TABLE_UNSPEC:an unspecified routing table
RT_TABLE_DEFAULT:the default table
RT_TABLE_MAIN:the main table
RT_TABLE_LOCAL:the local table
.TE
.sp 1
The user may assign arbitrary values between
.B RT_TABLE_UNSPEC
and
.BR RT_TABLE_DEFAULT .
.\" Keep table on same page
.bp +1
.TS
tab(:);
c s s
lb2 l2 l.
Attributes
rta_type:Value type:Description
_
RTA_UNSPEC:-:ignored
RTA_DST:protocol address:Route destination address
RTA_SRC:protocol address:Route source address
RTA_IIF:int:Input interface index
RTA_OIF:int:Output interface index
RTA_GATEWAY:protocol address:The gateway of the route
RTA_PRIORITY:int:Priority of route
RTA_PREFSRC:protocol address:Preferred source address
RTA_METRICS:int:Route metric
RTA_MULTIPATH::T{
Multipath nexthop data
br
(see below).
T}
RTA_PROTOINFO::No longer used
RTA_FLOW:int:Route realm
RTA_CACHEINFO:struct rta_cacheinfo:(see linux/rtnetlink.h)
RTA_SESSION::No longer used
RTA_MP_ALGO::No longer used
RTA_TABLE:int:T{
Routing table ID; if set,
.br
rtm_table is ignored
T}
RTA_MARK:int:
RTA_MFC_STATS:struct rta_mfc_stats:(see linux/rtnetlink.h)
RTA_VIA:struct rtvia:T{
Gateway in different AF
(see below)
T}
RTA_NEWDST:protocol address:T{
Change packet
destination address
T}
RTA_PREF:char:T{
RFC4191 IPv6 router
preference (see below)
T}
RTA_ENCAP_TYPE:short:T{
Encapsulation type for
.br
lwtunnels (see below)
T}
RTA_ENCAP::Defined by RTA_ENCAP_TYPE
RTA_EXPIRES:int:T{
Expire time for IPv6
routes (in seconds)
T}
.TE
.IP
.B RTA_MULTIPATH
contains several packed instances of
.I struct rtnexthop
together with nested RTAs
.RB ( RTA_GATEWAY ):
.IP
.in +4n
.EX
struct rtnexthop {
    unsigned short rtnh_len;     /* Length of struct + length
                                    of RTAs */
    unsigned char  rtnh_flags;   /* Flags (see
                                    linux/rtnetlink.h) */
    unsigned char  rtnh_hops;    /* Nexthop priority */
    int            rtnh_ifindex; /* Interface index for this
                                    nexthop */
}
.EE
.in
.IP
There exist a bunch of
.B RTNH_*
macros similar to
.B RTA_*
and
.B NLHDR_*
macros
useful to handle these structures.
.IP
.in +4n
.EX
struct rtvia {
    unsigned short rtvia_family;
    unsigned char  rtvia_addr[0];
};
.EE
.in
.IP
.I rtvia_addr
is the address,
.I rtvia_family
is its family type.
.IP
.B RTA_PREF
may contain values
.BR ICMPV6_ROUTER_PREF_LOW ,
.BR ICMPV6_ROUTER_PREF_MEDIUM ,
and
.B ICMPV6_ROUTER_PREF_HIGH
defined incw
.IR <linux/icmpv6.h> .
.IP
.B RTA_ENCAP_TYPE
may contain values
.BR LWTUNNEL_ENCAP_MPLS ,
.BR LWTUNNEL_ENCAP_IP ,
.BR LWTUNNEL_ENCAP_ILA ,
or
.B LWTUNNEL_ENCAP_IP6
defined in
.IR <linux/lwtunnel.h> .
.IP
.B Fill these values in!
.TP
.BR RTM_NEWNEIGH ", " RTM_DELNEIGH  ", " RTM_GETNEIGH
Add, remove, or receive information about a neighbor table
entry (e.g., an ARP entry).
The message contains an
.I ndmsg
structure.
.IP
.EX
struct ndmsg {
    unsigned char ndm_family;
    int           ndm_ifindex;  /* Interface index */
    __u16         ndm_state;    /* State */
    __u8          ndm_flags;    /* Flags */
    __u8          ndm_type;
};
\&
struct nda_cacheinfo {
    __u32         ndm_confirmed;
    __u32         ndm_used;
    __u32         ndm_updated;
    __u32         ndm_refcnt;
};
.EE
.IP
.I ndm_state
is a bit mask of the following states:
.TS
tab(:);
lb l.
NUD_INCOMPLETE:a currently resolving cache entry
NUD_REACHABLE:a confirmed working cache entry
NUD_STALE:an expired cache entry
NUD_DELAY:an entry waiting for a timer
NUD_PROBE:a cache entry that is currently reprobed
NUD_FAILED:an invalid cache entry
NUD_NOARP:a device with no destination cache
NUD_PERMANENT:a static entry
.TE
.sp 1
Valid
.I ndm_flags
are:
.TS
tab(:);
lb l.
NTF_PROXY:a proxy arp entry
NTF_ROUTER:an IPv6 router
.TE
.sp 1
.\" FIXME .
.\" document the members of the struct better
The
.I rtattr
struct has the following meanings for the
.I rta_type
field:
.TS
tab(:);
lb l.
NDA_UNSPEC:unknown type
NDA_DST:a neighbor cache n/w layer destination address
NDA_LLADDR:a neighbor cache link layer address
NDA_CACHEINFO:cache statistics
.TE
.sp 1
If the
.I rta_type
field is
.BR NDA_CACHEINFO ,
then a
.I struct nda_cacheinfo
header follows.
.TP
.BR RTM_NEWRULE ", " RTM_DELRULE ", " RTM_GETRULE
Add, delete, or retrieve a routing rule.
Carries a
.I struct rtmsg
.TP
.BR RTM_NEWQDISC ", " RTM_DELQDISC ", " RTM_GETQDISC
Add, remove, or get a queueing discipline.
The message contains a
.I struct tcmsg
and may be followed by a series of
attributes.
.IP
.EX
struct tcmsg {
    unsigned char    tcm_family;
    int              tcm_ifindex;   /* interface index */
    __u32            tcm_handle;    /* Qdisc handle */
    __u32            tcm_parent;    /* Parent qdisc */
    __u32            tcm_info;
};
.EE
.TS
tab(:);
c s s
lb2 l2 l.
Attributes
rta_type:Value type:Description
_
TCA_UNSPEC:-:unspecified
TCA_KIND:asciiz string:Name of queueing discipline
TCA_OPTIONS:byte sequence:Qdisc-specific options follow
TCA_STATS:struct tc_stats:Qdisc statistics
TCA_XSTATS:qdisc-specific:Module-specific statistics
TCA_RATE:struct tc_estimator:Rate limit
.TE
.sp 1
In addition, various other qdisc-module-specific attributes are allowed.
For more information see the appropriate include files.
.TP
.BR RTM_NEWTCLASS ", " RTM_DELTCLASS ", " RTM_GETTCLASS
Add, remove, or get a traffic class.
These messages contain a
.I struct tcmsg
as described above.
.TP
.BR RTM_NEWTFILTER ", " RTM_DELTFILTER ", " RTM_GETTFILTER
Add, remove, or receive information about a traffic filter.
These messages contain a
.I struct tcmsg
as described above.
.SH VERSIONS
.B rtnetlink
is a new feature of Linux 2.2.
.SH BUGS
This manual page is incomplete.
.SH SEE ALSO
.BR cmsg (3),
.BR rtnetlink (3),
.BR ip (7),
.BR netlink (7)