1IP-ROUTE(8) Linux IP-ROUTE(8)
2
6 ip-route - routing table management
7
9 ip [ ip-OPTIONS ] route { COMMAND | help }
10 ip route { show | flush } SELECTOR
13 ip route save SELECTOR
15 ip route restore
17 ip route get ROUTE_GET_FLAGS ADDRESS [ from ADDRESS iif STRING ] [ oif
19 STRING ] [ mark MARK ] [ tos TOS ] [ vrf NAME ] [ ipproto PRO‐
20 TOCOL ] [ sport NUMBER ] [ dport NUMBER ]
21 ip route { add | del | change | append | replace } ROUTE
23 SELECTOR := [ root PREFIX ] [ match PREFIX ] [ exact PREFIX ] [ table
25 TABLE_ID ] [ vrf NAME ] [ proto RTPROTO ] [ type TYPE ] [ scope
26 SCOPE ]
27 ROUTE := NODE_SPEC [ INFO_SPEC ]
29 NODE_SPEC := [ TYPE ] PREFIX [ tos TOS ] [ table TABLE_ID ] [ proto
31 RTPROTO ] [ scope SCOPE ] [ metric METRIC ] [ ttl-propagate {
32 enabled | disabled } ]
33 INFO_SPEC := { NH | nhid ID } OPTIONS FLAGS [ nexthop NH ] ...
35 NH := [ encap ENCAP ] [ via [ FAMILY ] ADDRESS ] [ dev STRING ] [
37 weight NUMBER ] NHFLAGS
38 FAMILY := [ inet | inet6 | mpls | bridge | link ]
40 OPTIONS := FLAGS [ mtu NUMBER ] [ advmss NUMBER ] [ as [ to ] ADDRESS ]
42 rtt TIME ] [ rttvar TIME ] [ reordering NUMBER ] [ window NUM‐
43 BER ] [ cwnd NUMBER ] [ ssthresh NUMBER ] [ realms REALM ] [
44 rto_min TIME ] [ initcwnd NUMBER ] [ initrwnd NUMBER ] [ fea‐
45 tures FEATURES ] [ quickack BOOL ] [ congctl NAME ] [ pref PREF
46 ] [ expires TIME ] [ fastopen_no_cookie BOOL ]
47 TYPE := [ unicast | local | broadcast | multicast | throw | unreachable
49 | prohibit | blackhole | nat ]
50 TABLE_ID := [ local| main | default | all | NUMBER ]
52 SCOPE := [ host | link | global | NUMBER ]
54 NHFLAGS := [ onlink | pervasive ]
56 RTPROTO := [ kernel | boot | static | NUMBER ]
58 FEATURES := [ ecn | ]
60 PREF := [ low | medium | high ]
62 ENCAP := [ ENCAP_MPLS | ENCAP_IP | ENCAP_BPF | ENCAP_SEG6 |
64 ENCAP_SEG6LOCAL ]
65 ENCAP_MPLS := mpls [ LABEL ] [ ttl TTL ]
67 ENCAP_IP := ip id TUNNEL_ID dst REMOTE_IP [ src SRC ] [ tos TOS ] [ ttl
69 TTL ]
70 ENCAP_BPF := bpf [ in PROG ] [ out PROG ] [ xmit PROG ] [ headroom SIZE
72 ]
73 ENCAP_SEG6 := seg6 mode [ encap | inline | l2encap ] segs SEGMENTS [
75 hmac KEYID ]
76 ENCAP_SEG6LOCAL := seg6local action SEG6_ACTION [ SEG6_ACTION_PARAM ]
78 ROUTE_GET_FLAGS := [ fibmatch ]
80
83 ip route is used to manipulate entries in the kernel routing tables.
84 Route types:
86 unicast - the route entry describes real paths to the destina‐
88 tions covered by the route prefix.
89 unreachable - these destinations are unreachable. Packets are
92 discarded and the ICMP message host unreachable is generated.
93 The local senders get an EHOSTUNREACH error.
94 blackhole - these destinations are unreachable. Packets are
97 discarded silently. The local senders get an EINVAL error.
98 prohibit - these destinations are unreachable. Packets are dis‐
101 carded and the ICMP message communication administratively pro‐
102 hibited is generated. The local senders get an EACCES error.
103 local - the destinations are assigned to this host. The packets
106 are looped back and delivered locally.
107 broadcast - the destinations are broadcast addresses. The pack‐
110 ets are sent as link broadcasts.
111 throw - a special control route used together with policy
114 rules. If such a route is selected, lookup in this table is
115 terminated pretending that no route was found. Without policy
116 routing it is equivalent to the absence of the route in the
117 routing table. The packets are dropped and the ICMP message net
118 unreachable is generated. The local senders get an ENETUNREACH
119 error.
120 nat - a special NAT route. Destinations covered by the prefix
123 are considered to be dummy (or external) addresses which
124 require translation to real (or internal) ones before forward‐
125 ing. The addresses to translate to are selected with the
126 attribute via. Warning: Route NAT is no longer supported in
127 Linux 2.6.
128 anycast - not implemented the destinations are anycast
131 addresses assigned to this host. They are mainly equivalent to
132 local with one difference: such addresses are invalid when used
133 as the source address of any packet.
134 multicast - a special type used for multicast routing. It is
137 not present in normal routing tables.
138 Route tables: Linux-2.x can pack routes into several routing tables
141 identified by a number in the range from 1 to 2^32-1 or by name from
142 the file /etc/iproute2/rt_tables By default all normal routes are
143 inserted into the main table (ID 254) and the kernel only uses this ta‐
144 ble when calculating routes. Values (0, 253, 254, and 255) are
145 reserved for built-in use.
146 Actually, one other table always exists, which is invisible but even
149 more important. It is the local table (ID 255). This table consists of
150 routes for local and broadcast addresses. The kernel maintains this ta‐
151 ble automatically and the administrator usually need not modify it or
152 even look at it.
153 The multiple routing tables enter the game when policy routing is used.
155 ip route add
158 add new route
159 ip route change
161 change route
162 ip route replace
164 change or add new one
165 to TYPE PREFIX (default)
167 the destination prefix of the route. If TYPE is omitted,
168 ip assumes type unicast. Other values of TYPE are listed
169 above. PREFIX is an IP or IPv6 address optionally fol‐
170 lowed by a slash and the prefix length. If the length of
171 the prefix is missing, ip assumes a full-length host
172 route. There is also a special PREFIX default - which is
173 equivalent to IP 0/0 or to IPv6 ::/0.
174 tos TOS
177 dsfield TOS
179 the Type Of Service (TOS) key. This key has no associated
180 mask and the longest match is understood as: First, com‐
181 pare the TOS of the route and of the packet. If they are
182 not equal, then the packet may still match a route with a
183 zero TOS. TOS is either an 8 bit hexadecimal number or
184 an identifier from /etc/iproute2/rt_dsfield.
185 metric NUMBER
188 preference NUMBER
190 the preference value of the route. NUMBER is an arbi‐
191 trary 32bit number, where routes with lower values are
192 preferred.
193 table TABLEID
196 the table to add this route to. TABLEID may be a number
197 or a string from the file /etc/iproute2/rt_tables. If
198 this parameter is omitted, ip assumes the main table,
199 with the exception of local, broadcast and nat routes,
200 which are put into the local table by default.
201 vrf NAME
204 the vrf name to add this route to. Implicitly means the
205 table associated with the VRF.
206 dev NAME
209 the output device name.
210 via [ FAMILY ] ADDRESS
213 the address of the nexthop router, in the address family
214 FAMILY. Actually, the sense of this field depends on the
215 route type. For normal unicast routes it is either the
216 true next hop router or, if it is a direct route
217 installed in BSD compatibility mode, it can be a local
218 address of the interface. For NAT routes it is the first
219 address of the block of translated IP destinations.
220 src ADDRESS
223 the source address to prefer when sending to the destina‐
224 tions covered by the route prefix.
225 realm REALMID
228 the realm to which this route is assigned. REALMID may
229 be a number or a string from the file
230 /etc/iproute2/rt_realms.
231 mtu MTU
234 mtu lock MTU
236 the MTU along the path to the destination. If the modi‐
237 fier lock is not used, the MTU may be updated by the ker‐
238 nel due to Path MTU Discovery. If the modifier lock is
239 used, no path MTU discovery will be tried, all packets
240 will be sent without the DF bit in IPv4 case or frag‐
241 mented to MTU for IPv6.
242 window NUMBER
245 the maximal window for TCP to advertise to these destina‐
246 tions, measured in bytes. It limits maximal data bursts
247 that our TCP peers are allowed to send to us.
248 rtt TIME
251 the initial RTT ('Round Trip Time') estimate. If no suf‐
252 fix is specified the units are raw values passed directly
253 to the routing code to maintain compatibility with previ‐
254 ous releases. Otherwise if a suffix of s, sec or secs is
255 used to specify seconds and ms, msec or msecs to specify
256 milliseconds.
257 rttvar TIME (Linux 2.3.15+ only)
261 the initial RTT variance estimate. Values are specified
262 as with rtt above.
263 rto_min TIME (Linux 2.6.23+ only)
266 the minimum TCP Retransmission TimeOut to use when commu‐
267 nicating with this destination. Values are specified as
268 with rtt above.
269 ssthresh NUMBER (Linux 2.3.15+ only)
272 an estimate for the initial slow start threshold.
273 cwnd NUMBER (Linux 2.3.15+ only)
276 the clamp for congestion window. It is ignored if the
277 lock flag is not used.
278 initcwnd NUMBER (Linux 2.5.70+ only)
281 the initial congestion window size for connections to
282 this destination. Actual window size is this value mul‐
283 tiplied by the MSS (``Maximal Segment Size'') for same
284 connection. The default is zero, meaning to use the val‐
285 ues specified in RFC2414.
286 initrwnd NUMBER (Linux 2.6.33+ only)
289 the initial receive window size for connections to this
290 destination. Actual window size is this value multiplied
291 by the MSS of the connection. The default value is zero,
292 meaning to use Slow Start value.
293 features FEATURES (Linux3.18+only)
296 Enable or disable per-route features. Only available fea‐
297 ture at this time is ecn to enable explicit congestion
298 notification when initiating connections to the given
299 destination network. When responding to a connection
300 request from the given network, ecn will also be used
301 even if the net.ipv4.tcp_ecn sysctl is set to 0.
302 quickack BOOL (Linux 3.11+ only)
305 Enable or disable quick ack for connections to this des‐
306 tination.
307 fastopen_no_cookie BOOL (Linux 4.15+ only)
310 Enable TCP Fastopen without a cookie for connections to
311 this destination.
312 congctl NAME (Linux 3.20+ only)
315 congctl lock NAME (Linux 3.20+ only)
317 Sets a specific TCP congestion control algorithm only for
318 a given destination. If not specified, Linux keeps the
319 current global default TCP congestion control algorithm,
320 or the one set from the application. If the modifier lock
321 is not used, an application may nevertheless overwrite
322 the suggested congestion control algorithm for that des‐
323 tination. If the modifier lock is used, then an applica‐
324 tion is not allowed to overwrite the specified congestion
325 control algorithm for that destination, thus it will be
326 enforced/guaranteed to use the proposed algorithm.
327 advmss NUMBER (Linux 2.3.15+ only)
330 the MSS ('Maximal Segment Size') to advertise to these
331 destinations when establishing TCP connections. If it is
332 not given, Linux uses a default value calculated from the
333 first hop device MTU. (If the path to these destination
334 is asymmetric, this guess may be wrong.)
335 reordering NUMBER (Linux 2.3.15+ only)
338 Maximal reordering on the path to this destination. If
339 it is not given, Linux uses the value selected with
340 sysctl variable net/ipv4/tcp_reordering.
341 nexthop NEXTHOP
344 the nexthop of a multipath route. NEXTHOP is a complex
345 value with its own syntax similar to the top level argu‐
346 ment lists:
347 via [ FAMILY ] ADDRESS - is the nexthop router.
349 dev NAME - is the output device.
352 weight NUMBER - is a weight for this element of a
355 multipath route reflecting its relative bandwidth
356 or quality.
357 The internal buffer used in iproute2 limits the maximum
359 number of nexthops that may be specified in one go. If
360 only ADDRESS is given, the current buffer size allows for
361 144 IPv6 nexthops and 253 IPv4 ones. For IPv4, this
362 effectively limits the number of nexthops possible per
363 route. With IPv6, further nexthops may be appended to the
364 same route via ip route append command.
365 scope SCOPE_VAL
368 the scope of the destinations covered by the route pre‐
369 fix. SCOPE_VAL may be a number or a string from the file
370 /etc/iproute2/rt_scopes. If this parameter is omitted,
371 ip assumes scope global for all gatewayed unicast routes,
372 scope link for direct unicast and broadcast routes and
373 scope host for local routes.
374 protocol RTPROTO
377 the routing protocol identifier of this route. RTPROTO
378 may be a number or a string from the file
379 /etc/iproute2/rt_protos. If the routing protocol ID is
380 not given, ip assumes protocol boot (i.e. it assumes the
381 route was added by someone who doesn't understand what
382 they are doing). Several protocol values have a fixed
383 interpretation. Namely:
384 redirect - the route was installed due to an ICMP
386 redirect.
387 kernel - the route was installed by the kernel
390 during autoconfiguration.
391 boot - the route was installed during the bootup
394 sequence. If a routing daemon starts, it will
395 purge all of them.
396 static - the route was installed by the adminis‐
399 trator to override dynamic routing. Routing dae‐
400 mon will respect them and, probably, even adver‐
401 tise them to its peers.
402 ra - the route was installed by Router Discovery
405 protocol.
406 The rest of the values are not reserved and the adminis‐
409 trator is free to assign (or not to assign) protocol
410 tags.
411 onlink pretend that the nexthop is directly attached to this
414 link, even if it does not match any interface prefix.
415 pref PREF
418 the IPv6 route preference. PREF is a string specifying
419 the route preference as defined in RFC4191 for Router
420 Discovery messages. Namely:
421 low - the route has a lowest priority
423 medium - the route has a default priority
426 high - the route has a highest priority
429 nhid ID
433 use nexthop object with given id as nexthop specifica‐
434 tion.
435 encap ENCAPTYPE ENCAPHDR
438 attach tunnel encapsulation attributes to this route.
439 ENCAPTYPE is a string specifying the supported encapsula‐
441 tion type. Namely:
442 mpls - encapsulation type MPLS
444 ip - IP encapsulation (Geneve, GRE, VXLAN, ...)
446 bpf - Execution of BPF program
448 seg6 - encapsulation type IPv6 Segment Routing
450 seg6local - local SRv6 segment processing
452 ENCAPHDR is a set of encapsulation attributes specific to
454 the ENCAPTYPE.
455 mpls
457 MPLSLABEL - mpls label stack with labels sepa‐
458 rated by /
459 ttl TTL - TTL to use for MPLS header or 0 to
462 inherit from IP header
463 ip
466 id TUNNEL_ID dst REMOTE_IP [ src SRC ] [ tos
467 TOS ] [ ttl TTL ] [ key ] [ csum ] [ seq ]
468 bpf
471 in PROG - BPF program to execute for incoming
472 packets
473 out PROG - BPF program to execute for outgoing
476 packets
477 xmit PROG - BPF program to execute for trans‐
480 mitted packets
481 headroom SIZE - Size of header BPF program will
484 attach (xmit)
485 seg6
488 mode inline - Directly insert Segment Routing
489 Header after IPv6 header
490 mode encap - Encapsulate packet in an outer
493 IPv6 header with SRH
494 mode l2encap - Encapsulate ingress L2 frame
497 within an outer IPv6 header and SRH
498 SEGMENTS - List of comma-separated IPv6
501 addresses
502 KEYID - Numerical value in decimal representa‐
505 tion. See ip-sr(8).
506 seg6local
509 SEG6_ACTION [ SEG6_ACTION_PARAM ] - Operation
510 to perform on matching packets. The following
511 actions are currently supported (Linux 4.14+
512 only).
513 End - Regular SRv6 processing as intermediate
515 segment endpoint. This action only accepts
516 packets with a non-zero Segments Left value.
517 Other matching packets are dropped.
518 End.X nh6 NEXTHOP - Regular SRv6 processing
520 as intermediate segment endpoint. Addition‐
521 ally, forward processed packets to given
522 next-hop. This action only accepts packets
523 with a non-zero Segments Left value. Other
524 matching packets are dropped.
525 End.DX6 nh6 NEXTHOP - Decapsulate inner IPv6
527 packet and forward it to the specified next-
528 hop. If the argument is set to ::, then the
529 next-hop is selected according to the local
530 selection rules. This action only accepts
531 packets with either a zero Segments Left
532 value or no SRH at all, and an inner IPv6
533 packet. Other matching packets are dropped.
534 End.B6 srh segs SEGMENTS [ hmac KEYID ] -
536 Insert the specified SRH immediately after
537 the IPv6 header, update the DA with the first
538 segment of the newly inserted SRH, then for‐
539 ward the resulting packet. The original SRH
540 is not modified. This action only accepts
541 packets with a non-zero Segments Left value.
542 Other matching packets are dropped.
543 End.B6.Encaps srh segs SEGMENTS [ hmac KEYID
545 ] - Regular SRv6 processing as intermediate
546 segment endpoint. Additionally, encapsulate
547 the matching packet within an outer IPv6
548 header followed by the specified SRH. The
549 destination address of the outer IPv6 header
550 is set to the first segment of the new SRH.
551 The source address is set as described in ip-
552 sr(8).
553 expires TIME (Linux 4.4+ only)
557 the route will be deleted after the expires time. Only
558 support IPv6 at present.
559 ttl-propagate { enabled | disabled }
562 Control whether TTL should be propagated from any encap
563 into the un-encapsulated packet, overriding any global
564 configuration. Only supported for MPLS at present.
565 ip route delete
568 delete route
569 ip route del has the same arguments as ip route add, but their
570 semantics are a bit different.
571 Key values (to, tos, preference and table) select the route to
573 delete. If optional attributes are present, ip verifies that
574 they coincide with the attributes of the route to delete. If no
575 route with the given key and attributes was found, ip route del
576 fails.
577 ip route show
580 list routes
581 the command displays the contents of the routing tables or the
582 route(s) selected by some criteria.
583 to SELECTOR (default)
586 only select routes from the given range of destinations.
587 SELECTOR consists of an optional modifier (root, match or
588 exact) and a prefix. root PREFIX selects routes with
589 prefixes not shorter than PREFIX. F.e. root 0/0 selects
590 the entire routing table. match PREFIX selects routes
591 with prefixes not longer than PREFIX. F.e. match
592 10.0/16 selects 10.0/16, 10/8 and 0/0, but it does not
593 select 10.1/16 and 10.0.0/24. And exact PREFIX (or just
594 PREFIX) selects routes with this exact prefix. If neither
595 of these options are present, ip assumes root 0/0 i.e. it
596 lists the entire table.
597 tos TOS
600 dsfield TOS
602 only select routes with the given TOS.
603 table TABLEID
606 show the routes from this table(s). The default setting
607 is to show table main. TABLEID may either be the ID of a
608 real table or one of the special values:
609 all - list all of the tables.
611 cache - dump the routing cache.
613 vrf NAME
616 show the routes for the table associated with the vrf
617 name
618 cloned
621 cached list cloned routes i.e. routes which were dynamically
623 forked from other routes because some route attribute
624 (f.e. MTU) was updated. Actually, it is equivalent to
625 table cache.
626 from SELECTOR
629 the same syntax as for to, but it binds the source
630 address range rather than destinations. Note that the
631 from option only works with cloned routes.
632 protocol RTPROTO
635 only list routes of this protocol.
636 scope SCOPE_VAL
639 only list routes with this scope.
640 type TYPE
643 only list routes of this type.
644 dev NAME
647 only list routes going via this device.
648 via [ FAMILY ] PREFIX
651 only list routes going via the nexthop routers selected
652 by PREFIX.
653 src PREFIX
656 only list routes with preferred source addresses selected
657 by PREFIX.
658 realm REALMID
661 realms FROMREALM/TOREALM
663 only list routes with these realms.
664 ip route flush
667 flush routing tables
668 this command flushes routes selected by some criteria.
669 The arguments have the same syntax and semantics as the argu‐
672 ments of ip route show, but routing tables are not listed but
673 purged. The only difference is the default action: show dumps
674 all the IP main routing table but flush prints the helper page.
675 With the -statistics option, the command becomes verbose. It
678 prints out the number of deleted routes and the number of rounds
679 made to flush the routing table. If the option is given twice,
680 ip route flush also dumps all the deleted routes in the format
681 described in the previous subsection.
682 ip route get
685 get a single route
686 this command gets a single route to a destination and prints its
687 contents exactly as the kernel sees it.
688 fibmatch
691 Return full fib lookup matched route. Default is to
692 return the resolved dst entry
693 to ADDRESS (default)
696 the destination address.
697 from ADDRESS
700 the source address.
701 tos TOS
704 dsfield TOS
706 the Type Of Service.
707 iif NAME
710 the device from which this packet is expected to arrive.
711 oif NAME
714 force the output device on which this packet will be
715 routed.
716 mark MARK
719 the firewall mark (fwmark)
720 vrf NAME
723 force the vrf device on which this packet will be routed.
724 ipproto PROTOCOL
727 ip protocol as seen by the route lookup
728 sport NUMBER
731 source port as seen by the route lookup
732 dport NUMBER
735 destination port as seen by the route lookup
736 connected
739 if no source address (option from) was given, relookup
740 the route with the source set to the preferred address
741 received from the first lookup. If policy routing is
742 used, it may be a different route.
743 Note that this operation is not equivalent to ip route show.
746 show shows existing routes. get resolves them and creates new
747 clones if necessary. Essentially, get is equivalent to sending a
748 packet along this path. If the iif argument is not given, the
749 kernel creates a route to output packets towards the requested
750 destination. This is equivalent to pinging the destination with
751 a subsequent ip route ls cache, however, no packets are actually
752 sent. With the iif argument, the kernel pretends that a packet
753 arrived from this interface and searches for a path to forward
754 the packet.
755 ip route save
758 save routing table information to stdout
759 This command behaves like ip route show except that the output
760 is raw data suitable for passing to ip route restore.
761 ip route restore
764 restore routing table information from stdin
765 This command expects to read a data stream as returned from ip
766 route save. It will attempt to restore the routing table infor‐
767 mation exactly as it was at the time of the save, so any trans‐
768 lation of information in the stream (such as device indexes)
769 must be done first. Any existing routes are left unchanged. Any
770 routes specified in the data stream that already exist in the
771 table will be ignored.
772
775 Starting with Linux kernel version 3.6, there is no routing cache for
776 IPv4 anymore. Hence ip route show cached will never print any entries
777 on systems with this or newer kernel versions.
778
781 ip ro
782 Show all route entries in the kernel.
783 ip route add default via 192.168.1.1 dev eth0
785 Adds a default route (for all addresses) via the local gateway
786 192.168.1.1 that can be reached on device eth0.
787 ip route add 10.1.1.0/30 encap mpls 200/300 via 10.1.1.1 dev eth0
789 Adds an ipv4 route with mpls encapsulation attributes attached to
790 it.
791 ip -6 route add 2001:db8:1::/64 encap seg6 mode encap segs
793 2001:db8:42::1,2001:db8:ffff::2 dev eth0
794 Adds an IPv6 route with SRv6 encapsulation and two segments
795 attached.
796 ip route add 10.1.1.0/30 nhid 10
798 Adds an ipv4 route using nexthop object with id 10.
799
801 ip(8)
802
805 Original Manpage by Michail Litvak <mci@owl.openwall.com>
806iproute2 13 Dec 2012 IP-ROUTE(8)