7 :abbr:`BGP` stands for Border Gateway Protocol. The latest BGP version is 4.
8 BGP-4 is one of the Exterior Gateway Protocols and the de facto standard
9 interdomain routing protocol. BGP-4 is described in :rfc:`1771` and updated by
10 :rfc:`4271`. :rfc:`2858` adds multiprotocol support to BGP-4.
17 The default configuration file of *bgpd* is :file:`bgpd.conf`. *bgpd* searches
18 the current directory first, followed by |INSTALL_PREFIX_ETC|/bgpd.conf. All of
19 *bgpd*'s commands must be configured in :file:`bgpd.conf` when the integrated
20 config is not being used.
22 *bgpd* specific invocation options are described below. Common options may also
23 be specified (:ref:`common-invocation-options`).
27 .. option:: -p, --bgp_port <port>
29 Set the bgp protocol's port number. When port number is 0, that means do not
32 .. option:: -l, --listenon
34 Specify a specific IP address for bgpd to listen on, rather than its default
35 of ``0.0.0.0`` / ``::``. This can be useful to constrain bgpd to an internal
36 address, or to run multiple bgpd processes on one host.
38 .. _bgp-basic-concepts:
43 .. _bgp-autonomous-systems:
50 An AS is a connected group of one or more IP prefixes run by one or more
51 network operators which has a SINGLE and CLEARLY DEFINED routing policy.
53 Each AS has an identifying number associated with it called an :abbr:`ASN
54 (Autonomous System Number)`. This is a two octet value ranging in value from 1
55 to 65535. The AS numbers 64512 through 65535 are defined as private AS numbers.
56 Private AS numbers must not be advertised on the global Internet.
58 The :abbr:`ASN (Autonomous System Number)` is one of the essential elements of
59 BGP. BGP is a distance vector routing protocol, and the AS-Path framework
60 provides distance vector metric and loop detection to BGP.
62 .. seealso:: :rfc:`1930`
64 .. _bgp-address-families:
69 Multiprotocol extensions enable BGP to carry routing information for multiple
70 network layer protocols. BGP supports an Address Family Identifier (AFI) for
71 IPv4 and IPv6. Support is also provided for multiple sets of per-AFI
72 information via the BGP Subsequent Address Family Identifier (SAFI). FRR
73 supports SAFIs for unicast information, labeled information (:rfc:`3107` and
74 :rfc:`8277`), and Layer 3 VPN information (:rfc:`4364` and :rfc:`4659`).
76 .. _bgp-route-selection:
81 The route selection process used by FRR's BGP implementation uses the following
82 decision criterion, starting at the top of the list and going towards the
83 bottom until one of the factors can be used.
87 Prefer higher local weight routes to lower routes.
89 2. **Local preference check**
91 Prefer higher local preference routes to lower.
93 3. **Local route check**
95 Prefer local routes (statics, aggregates, redistributed) to received routes.
97 4. **AS path length check**
99 Prefer shortest hop-count AS_PATHs.
103 Prefer the lowest origin type route. That is, prefer IGP origin routes to
104 EGP, to Incomplete routes.
108 Where routes with a MED were received from the same AS, prefer the route
109 with the lowest MED. :ref:`bgp-med`.
111 7. **External check**
113 Prefer the route received from an external, eBGP peer over routes received
114 from other types of peers.
116 8. **IGP cost check**
118 Prefer the route with the lower IGP cost.
120 9. **Multi-path check**
122 If multi-pathing is enabled, then check whether the routes not yet
123 distinguished in preference may be considered equal. If
124 :clicmd:`bgp bestpath as-path multipath-relax` is set, all such routes are
125 considered equal, otherwise routes received via iBGP with identical AS_PATHs
126 or routes received from eBGP neighbours in the same AS are considered equal.
128 10. **Already-selected external check**
130 Where both routes were received from eBGP peers, then prefer the route
131 which is already selected. Note that this check is not applied if
132 :clicmd:`bgp bestpath compare-routerid` is configured. This check can
133 prevent some cases of oscillation.
135 11. **Router-ID check**
137 Prefer the route with the lowest `router-ID`. If the route has an
138 `ORIGINATOR_ID` attribute, through iBGP reflection, then that router ID is
139 used, otherwise the `router-ID` of the peer the route was received from is
142 12. **Cluster-List length check**
144 The route with the shortest cluster-list length is used. The cluster-list
145 reflects the iBGP reflection path the route has taken.
149 Prefer the route received from the peer with the higher transport layer
150 address, as a last-resort tie-breaker.
152 .. _bgp-capability-negotiation:
154 Capability Negotiation
155 ----------------------
157 When adding IPv6 routing information exchange feature to BGP. There were some
158 proposals. :abbr:`IETF (Internet Engineering Task Force)`
159 :abbr:`IDR (Inter Domain Routing)` adopted a proposal called Multiprotocol
160 Extension for BGP. The specification is described in :rfc:`2283`. The protocol
161 does not define new protocols. It defines new attributes to existing BGP. When
162 it is used exchanging IPv6 routing information it is called BGP-4+. When it is
163 used for exchanging multicast routing information it is called MBGP.
165 *bgpd* supports Multiprotocol Extension for BGP. So if a remote peer supports
166 the protocol, *bgpd* can exchange IPv6 and/or multicast routing information.
168 Traditional BGP did not have the feature to detect a remote peer's
169 capabilities, e.g. whether it can handle prefix types other than IPv4 unicast
170 routes. This was a big problem using Multiprotocol Extension for BGP in an
171 operational network. :rfc:`2842` adopted a feature called Capability
172 Negotiation. *bgpd* use this Capability Negotiation to detect the remote peer's
173 capabilities. If a peer is only configured as an IPv4 unicast neighbor, *bgpd*
174 does not send these Capability Negotiation packets (at least not unless other
175 optional BGP features require capability negotiation).
177 By default, FRR will bring up peering with minimal common capability for the
178 both sides. For example, if the local router has unicast and multicast
179 capabilities and the remote router only has unicast capability the local router
180 will establish the connection with unicast only capability. When there are no
181 common capabilities, FRR sends Unsupported Capability error and then resets the
184 .. _bgp-concepts-vrfs:
186 VRFs: Virtual Routing and Forwarding
187 ------------------------------------
189 *bgpd* supports :abbr:`L3VPN (Layer 3 Virtual Private Networks)` :abbr:`VRFs
190 (Virtual Routing and Forwarding tables)` for IPv4 :rfc:`4364` and IPv6
191 :rfc:`4659`. L3VPN routes, and their associated VRF MPLS labels, can be
192 distributed to VPN SAFI neighbors in the *default*, i.e., non VRF, BGP
193 instance. VRF MPLS labels are reached using *core* MPLS labels which are
194 distributed using LDP or BGP labeled unicast. *bgpd* also supports inter-VRF
195 route leaking. General information on FRR's VRF support can be found in
198 .. _bgp-router-configuration:
200 BGP Router Configuration
201 ========================
206 First of all you must configure BGP router with the :clicmd:`router bgp ASN`
207 command. The AS number is an identifier for the autonomous system. The BGP
208 protocol uses the AS number for detecting whether the BGP connection is
209 internal or external.
211 .. index:: router bgp ASN
212 .. clicmd:: router bgp ASN
214 Enable a BGP protocol process with the specified ASN. After
215 this statement you can input any `BGP Commands`.
217 .. index:: no router bgp ASN
218 .. clicmd:: no router bgp ASN
220 Destroy a BGP protocol process with the specified ASN.
222 .. index:: bgp router-id A.B.C.D
223 .. clicmd:: bgp router-id A.B.C.D
225 This command specifies the router-ID. If *bgpd* connects to *zebra* it gets
226 interface and address information. In that case default router ID value is
227 selected as the largest IP Address of the interfaces. When `router zebra` is
228 not enabled *bgpd* can't get interface information so `router-id` is set to
229 0.0.0.0. So please set router-id by hand.
234 .. index:: bgp bestpath as-path confed
235 .. clicmd:: bgp bestpath as-path confed
237 This command specifies that the length of confederation path sets and
238 sequences should should be taken into account during the BGP best path
241 .. index:: bgp bestpath as-path multipath-relax
242 .. clicmd:: bgp bestpath as-path multipath-relax
244 This command specifies that BGP decision process should consider paths
245 of equal AS_PATH length candidates for multipath computation. Without
246 the knob, the entire AS_PATH must match for multipath computation.
248 .. clicmd:: bgp bestpath compare-routerid
250 Ensure that when comparing routes where both are equal on most metrics,
251 including local-pref, AS_PATH length, IGP cost, MED, that the tie is broken
254 If this option is enabled, then the already-selected check, where
255 already selected eBGP routes are preferred, is skipped.
257 If a route has an `ORIGINATOR_ID` attribute because it has been reflected,
258 that `ORIGINATOR_ID` will be used. Otherwise, the router-ID of the peer the
259 route was received from will be used.
261 The advantage of this is that the route-selection (at this point) will be
262 more deterministic. The disadvantage is that a few or even one lowest-ID
263 router may attract all traffic to otherwise-equal paths because of this
264 check. It may increase the possibility of MED or IGP oscillation, unless
265 other measures were taken to avoid these. The exact behaviour will be
266 sensitive to the iBGP and reflection topology.
270 Administrative Distance Metrics
271 -------------------------------
273 .. index:: distance bgp (1-255) (1-255) (1-255)
274 .. clicmd:: distance bgp (1-255) (1-255) (1-255)
276 This command change distance value of BGP. The arguments are the distance
277 values for for external routes, internal routes and local routes
280 .. index:: distance (1-255) A.B.C.D/M
281 .. clicmd:: distance (1-255) A.B.C.D/M
283 .. index:: distance (1-255) A.B.C.D/M WORD
284 .. clicmd:: distance (1-255) A.B.C.D/M WORD
286 Sets the administrative distance for a particular route.
288 .. _bgp-route-flap-dampening:
293 .. clicmd:: bgp dampening (1-45) (1-20000) (1-20000) (1-255)
295 This command enables BGP route-flap dampening and specifies dampening parameters.
298 Half-life time for the penalty
301 Value to start reusing a route
304 Value to start suppressing a route
307 Maximum duration to suppress a stable route
309 The route-flap damping algorithm is compatible with :rfc:`2439`. The use of
310 this command is not recommended nowadays.
313 https://www.ripe.net/publications/docs/ripe-378
317 Multi-Exit Discriminator
318 ------------------------
320 The BGP :abbr:`MED (Multi-Exit Discriminator)` attribute has properties which
321 can cause subtle convergence problems in BGP. These properties and problems
322 have proven to be hard to understand, at least historically, and may still not
323 be widely understood. The following attempts to collect together and present
324 what is known about MED, to help operators and FRR users in designing and
325 configuring their networks.
327 The BGP :abbr:`MED` attribute is intended to allow one AS to indicate its
328 preferences for its ingress points to another AS. The MED attribute will not be
329 propagated on to another AS by the receiving AS - it is 'non-transitive' in the
332 E.g., if AS X and AS Y have 2 different BGP peering points, then AS X might set
333 a MED of 100 on routes advertised at one and a MED of 200 at the other. When AS
334 Y selects between otherwise equal routes to or via AS X, AS Y should prefer to
335 take the path via the lower MED peering of 100 with AS X. Setting the MED
336 allows an AS to influence the routing taken to it within another, neighbouring
339 In this use of MED it is not really meaningful to compare the MED value on
340 routes where the next AS on the paths differs. E.g., if AS Y also had a route
341 for some destination via AS Z in addition to the routes from AS X, and AS Z had
342 also set a MED, it wouldn't make sense for AS Y to compare AS Z's MED values to
343 those of AS X. The MED values have been set by different administrators, with
344 different frames of reference.
346 The default behaviour of BGP therefore is to not compare MED values across
347 routes received from different neighbouring ASes. In FRR this is done by
348 comparing the neighbouring, left-most AS in the received AS_PATHs of the routes
349 and only comparing MED if those are the same.
351 Unfortunately, this behaviour of MED, of sometimes being compared across routes
352 and sometimes not, depending on the properties of those other routes, means MED
353 can cause the order of preference over all the routes to be undefined. That is,
354 given routes A, B, and C, if A is preferred to B, and B is preferred to C, then
355 a well-defined order should mean the preference is transitive (in the sense of
356 orders [#med-transitivity-rant]_) and that A would be preferred to C.
358 However, when MED is involved this need not be the case. With MED it is
359 possible that C is actually preferred over A. So A is preferred to B, B is
360 preferred to C, but C is preferred to A. This can be true even where BGP
361 defines a deterministic 'most preferred' route out of the full set of A,B,C.
362 With MED, for any given set of routes there may be a deterministically
363 preferred route, but there need not be any way to arrange them into any order
364 of preference. With unmodified MED, the order of preference of routes literally
367 That MED can induce non-transitive preferences over routes can cause issues.
368 Firstly, it may be perceived to cause routing table churn locally at speakers;
369 secondly, and more seriously, it may cause routing instability in iBGP
370 topologies, where sets of speakers continually oscillate between different
373 The first issue arises from how speakers often implement routing decisions.
374 Though BGP defines a selection process that will deterministically select the
375 same route as best at any given speaker, even with MED, that process requires
376 evaluating all routes together. For performance and ease of implementation
377 reasons, many implementations evaluate route preferences in a pair-wise fashion
378 instead. Given there is no well-defined order when MED is involved, the best
379 route that will be chosen becomes subject to implementation details, such as
380 the order the routes are stored in. That may be (locally) non-deterministic,
381 e.g.: it may be the order the routes were received in.
383 This indeterminism may be considered undesirable, though it need not cause
384 problems. It may mean additional routing churn is perceived, as sometimes more
385 updates may be produced than at other times in reaction to some event .
387 This first issue can be fixed with a more deterministic route selection that
388 ensures routes are ordered by the neighbouring AS during selection.
389 :clicmd:`bgp deterministic-med`. This may reduce the number of updates as routes
390 are received, and may in some cases reduce routing churn. Though, it could
391 equally deterministically produce the largest possible set of updates in
392 response to the most common sequence of received updates.
394 A deterministic order of evaluation tends to imply an additional overhead of
395 sorting over any set of n routes to a destination. The implementation of
396 deterministic MED in FRR scales significantly worse than most sorting
397 algorithms at present, with the number of paths to a given destination. That
398 number is often low enough to not cause any issues, but where there are many
399 paths, the deterministic comparison may quickly become increasingly expensive
402 Deterministic local evaluation can *not* fix the second, more major, issue of
403 MED however. Which is that the non-transitive preference of routes MED can
404 cause may lead to routing instability or oscillation across multiple speakers
405 in iBGP topologies. This can occur with full-mesh iBGP, but is particularly
406 problematic in non-full-mesh iBGP topologies that further reduce the routing
407 information known to each speaker. This has primarily been documented with iBGP
408 route-reflection topologies. However, any route-hiding technologies potentially
409 could also exacerbate oscillation with MED.
411 This second issue occurs where speakers each have only a subset of routes, and
412 there are cycles in the preferences between different combinations of routes -
413 as the undefined order of preference of MED allows - and the routes are
414 distributed in a way that causes the BGP speakers to 'chase' those cycles. This
415 can occur even if all speakers use a deterministic order of evaluation in route
418 E.g., speaker 4 in AS A might receive a route from speaker 2 in AS X, and from
419 speaker 3 in AS Y; while speaker 5 in AS A might receive that route from
420 speaker 1 in AS Y. AS Y might set a MED of 200 at speaker 1, and 100 at speaker
421 3. I.e, using ASN:ID:MED to label the speakers:
427 X:2------|--A:4-------A:5--|-Y:1:200
433 Assuming all other metrics are equal (AS_PATH, ORIGIN, 0 IGP costs), then based
434 on the RFC4271 decision process speaker 4 will choose X:2 over Y:3:100, based
435 on the lower ID of 2. Speaker 4 advertises X:2 to speaker 5. Speaker 5 will
436 continue to prefer Y:1:200 based on the ID, and advertise this to speaker 4.
437 Speaker 4 will now have the full set of routes, and the Y:1:200 it receives
438 from 5 will beat X:2, but when speaker 4 compares Y:1:200 to Y:3:100 the MED
439 check now becomes active as the ASes match, and now Y:3:100 is preferred.
440 Speaker 4 therefore now advertises Y:3:100 to 5, which will also agrees that
441 Y:3:100 is preferred to Y:1:200, and so withdraws the latter route from 4.
442 Speaker 4 now has only X:2 and Y:3:100, and X:2 beats Y:3:100, and so speaker 4
443 implicitly updates its route to speaker 5 to X:2. Speaker 5 sees that Y:1:200
444 beats X:2 based on the ID, and advertises Y:1:200 to speaker 4, and the cycle
447 The root cause is the lack of a clear order of preference caused by how MED
448 sometimes is and sometimes is not compared, leading to this cycle in the
449 preferences between the routes:
454 /---> X:2 ---beats---> Y:3:100 --\\
457 \\---beats--- Y:1:200 <---beats---/
461 This particular type of oscillation in full-mesh iBGP topologies can be
462 avoided by speakers preferring already selected, external routes rather than
463 choosing to update to new a route based on a post-MED metric (e.g. router-ID),
464 at the cost of a non-deterministic selection process. FRR implements this, as
465 do many other implementations, so long as it is not overridden by setting
466 :clicmd:`bgp bestpath compare-routerid`, and see also
467 :ref:`bgp-route-selection`.
469 However, more complex and insidious cycles of oscillation are possible with
470 iBGP route-reflection, which are not so easily avoided. These have been
471 documented in various places. See, e.g.:
473 - [bgp-route-osci-cond]_
474 - [stable-flexible-ibgp]_
475 - [ibgp-correctness]_
477 for concrete examples and further references.
479 There is as of this writing *no* known way to use MED for its original purpose;
480 *and* reduce routing information in iBGP topologies; *and* be sure to avoid the
481 instability problems of MED due the non-transitive routing preferences it can
482 induce; in general on arbitrary networks.
484 There may be iBGP topology specific ways to reduce the instability risks, even
485 while using MED, e.g.: by constraining the reflection topology and by tuning
486 IGP costs between route-reflector clusters, see :rfc:`3345` for details. In the
487 near future, the Add-Path extension to BGP may also solve MED oscillation while
488 still allowing MED to be used as intended, by distributing "best-paths per
489 neighbour AS". This would be at the cost of distributing at least as many
490 routes to all speakers as a full-mesh iBGP would, if not more, while also
491 imposing similar CPU overheads as the "Deterministic MED" feature at each
494 More generally, the instability problems that MED can introduce on more
495 complex, non-full-mesh, iBGP topologies may be avoided either by:
497 - Setting :clicmd:`bgp always-compare-med`, however this allows MED to be compared
498 across values set by different neighbour ASes, which may not produce
499 coherent desirable results, of itself.
500 - Effectively ignoring MED by setting MED to the same value (e.g.: 0) using
501 :clicmd:`set metric METRIC` on all received routes, in combination with
502 setting :clicmd:`bgp always-compare-med` on all speakers. This is the simplest
503 and most performant way to avoid MED oscillation issues, where an AS is happy
504 not to allow neighbours to inject this problematic metric.
506 As MED is evaluated after the AS_PATH length check, another possible use for
507 MED is for intra-AS steering of routes with equal AS_PATH length, as an
508 extension of the last case above. As MED is evaluated before IGP metric, this
509 can allow cold-potato routing to be implemented to send traffic to preferred
510 hand-offs with neighbours, rather than the closest hand-off according to the
513 Note that even if action is taken to address the MED non-transitivity issues,
514 other oscillations may still be possible. E.g., on IGP cost if iBGP and IGP
515 topologies are at cross-purposes with each other - see the Flavel and Roughan
516 paper above for an example. Hence the guideline that the iBGP topology should
517 follow the IGP topology.
519 .. index:: bgp deterministic-med
520 .. clicmd:: bgp deterministic-med
522 Carry out route-selection in way that produces deterministic answers
523 locally, even in the face of MED and the lack of a well-defined order of
524 preference it can induce on routes. Without this option the preferred route
525 with MED may be determined largely by the order that routes were received
528 Setting this option will have a performance cost that may be noticeable when
529 there are many routes for each destination. Currently in FRR it is
530 implemented in a way that scales poorly as the number of routes per
531 destination increases.
533 The default is that this option is not set.
535 Note that there are other sources of indeterminism in the route selection
536 process, specifically, the preference for older and already selected routes
537 from eBGP peers, :ref:`bgp-route-selection`.
539 .. index:: bgp always-compare-med
540 .. clicmd:: bgp always-compare-med
542 Always compare the MED on routes, even when they were received from
543 different neighbouring ASes. Setting this option makes the order of
544 preference of routes more defined, and should eliminate MED induced
547 If using this option, it may also be desirable to use
548 :clicmd:`set metric METRIC` to set MED to 0 on routes received from external
551 This option can be used, together with :clicmd:`set metric METRIC` to use
552 MED as an intra-AS metric to steer equal-length AS_PATH routes to, e.g.,
560 .. index:: network A.B.C.D/M
561 .. clicmd:: network A.B.C.D/M
563 This command adds the announcement network.
568 address-family ipv4 unicast
572 This configuration example says that network 10.0.0.0/8 will be
573 announced to all neighbors. Some vendors' routers don't advertise
574 routes if they aren't present in their IGP routing tables; `bgpd`
575 doesn't care about IGP routes when announcing its routes.
577 .. index:: no network A.B.C.D/M
578 .. clicmd:: no network A.B.C.D/M
580 .. _bgp-route-aggregation:
585 .. index:: aggregate-address A.B.C.D/M
586 .. clicmd:: aggregate-address A.B.C.D/M
588 This command specifies an aggregate address.
590 .. index:: aggregate-address A.B.C.D/M as-set
591 .. clicmd:: aggregate-address A.B.C.D/M as-set
593 This command specifies an aggregate address. Resulting routes include
596 .. index:: aggregate-address A.B.C.D/M summary-only
597 .. clicmd:: aggregate-address A.B.C.D/M summary-only
599 This command specifies an aggregate address. Aggregated routes will
602 .. index:: no aggregate-address A.B.C.D/M
603 .. clicmd:: no aggregate-address A.B.C.D/M
605 .. _bgp-redistribute-to-bgp:
610 .. index:: redistribute kernel
611 .. clicmd:: redistribute kernel
613 Redistribute kernel route to BGP process.
615 .. index:: redistribute static
616 .. clicmd:: redistribute static
618 Redistribute static route to BGP process.
620 .. index:: redistribute connected
621 .. clicmd:: redistribute connected
623 Redistribute connected route to BGP process.
625 .. index:: redistribute rip
626 .. clicmd:: redistribute rip
628 Redistribute RIP route to BGP process.
630 .. index:: redistribute ospf
631 .. clicmd:: redistribute ospf
633 Redistribute OSPF route to BGP process.
635 .. index:: redistribute vpn
636 .. clicmd:: redistribute vpn
638 Redistribute VNC routes to BGP process.
640 .. index:: update-delay MAX-DELAY
641 .. clicmd:: update-delay MAX-DELAY
643 .. index:: update-delay MAX-DELAY ESTABLISH-WAIT
644 .. clicmd:: update-delay MAX-DELAY ESTABLISH-WAIT
646 This feature is used to enable read-only mode on BGP process restart or when
647 BGP process is cleared using 'clear ip bgp \*'. When applicable, read-only
648 mode would begin as soon as the first peer reaches Established status and a
649 timer for max-delay seconds is started.
651 During this mode BGP doesn't run any best-path or generate any updates to its
652 peers. This mode continues until:
654 1. All the configured peers, except the shutdown peers, have sent explicit EOR
655 (End-Of-RIB) or an implicit-EOR. The first keep-alive after BGP has reached
656 Established is considered an implicit-EOR.
657 If the establish-wait optional value is given, then BGP will wait for
658 peers to reach established from the beginning of the update-delay till the
659 establish-wait period is over, i.e. the minimum set of established peers for
660 which EOR is expected would be peers established during the establish-wait
661 window, not necessarily all the configured neighbors.
662 2. max-delay period is over.
664 On hitting any of the above two conditions, BGP resumes the decision process
665 and generates updates to its peers.
667 Default max-delay is 0, i.e. the feature is off by default.
669 .. index:: table-map ROUTE-MAP-NAME
670 .. clicmd:: table-map ROUTE-MAP-NAME
672 This feature is used to apply a route-map on route updates from BGP to
673 Zebra. All the applicable match operations are allowed, such as match on
674 prefix, next-hop, communities, etc. Set operations for this attach-point are
675 limited to metric and next-hop only. Any operation of this feature does not
676 affect BGPs internal RIB.
678 Supported for ipv4 and ipv6 address families. It works on multi-paths as
679 well, however, metric setting is based on the best-path only.
686 .. _bgp-defining-peers:
691 .. index:: neighbor PEER remote-as ASN
692 .. clicmd:: neighbor PEER remote-as ASN
694 Creates a new neighbor whose remote-as is ASN. PEER can be an IPv4 address
695 or an IPv6 address or an interface to use for the connection.
700 neighbor 10.0.0.1 remote-as 2
702 In this case my router, in AS-1, is trying to peer with AS-2 at 10.0.0.1.
704 This command must be the first command used when configuring a neighbor. If
705 the remote-as is not specified, *bgpd* will complain like this: ::
707 can't find neighbor 10.0.0.1
709 .. index:: neighbor PEER remote-as internal
710 .. clicmd:: neighbor PEER remote-as internal
712 Create a peer as you would when you specify an ASN, except that if the
713 peers ASN is different than mine as specified under the :clicmd:`router bgp ASN`
714 command the connection will be denied.
716 .. index:: neighbor PEER remote-as external
717 .. clicmd:: neighbor PEER remote-as external
719 Create a peer as you would when you specify an ASN, except that if the
720 peers ASN is the same as mine as specified under the :clicmd:`router bgp ASN`
721 command the connection will be denied.
723 .. _bgp-configuring-peers:
728 .. index:: [no] neighbor PEER shutdown
729 .. clicmd:: [no] neighbor PEER shutdown
731 Shutdown the peer. We can delete the neighbor's configuration by
732 ``no neighbor PEER remote-as ASN`` but all configuration of the neighbor
733 will be deleted. When you want to preserve the configuration, but want to
734 drop the BGP peer, use this syntax.
736 .. index:: [no] neighbor PEER disable-connected-check
737 .. clicmd:: [no] neighbor PEER disable-connected-check
739 Allow peerings between directly connected eBGP peers using loopback
742 .. index:: [no] neighbor PEER ebgp-multihop
743 .. clicmd:: [no] neighbor PEER ebgp-multihop
745 .. index:: [no] neighbor PEER description ...
746 .. clicmd:: [no] neighbor PEER description ...
748 Set description of the peer.
750 .. index:: [no] neighbor PEER version VERSION
751 .. clicmd:: [no] neighbor PEER version VERSION
753 Set up the neighbor's BGP version. `version` can be `4`, `4+` or `4-`. BGP
754 version `4` is the default value used for BGP peering. BGP version `4+`
755 means that the neighbor supports Multiprotocol Extensions for BGP-4. BGP
756 version `4-` is similar but the neighbor speaks the old Internet-Draft
757 revision 00's Multiprotocol Extensions for BGP-4. Some routing software is
758 still using this version.
760 .. index:: [no] neighbor PEER interface IFNAME
761 .. clicmd:: [no] neighbor PEER interface IFNAME
763 When you connect to a BGP peer over an IPv6 link-local address, you have to
764 specify the IFNAME of the interface used for the connection. To specify
765 IPv4 session addresses, see the ``neighbor PEER update-source`` command
768 This command is deprecated and may be removed in a future release. Its use
771 .. index:: [no] neighbor PEER next-hop-self [all]
772 .. clicmd:: [no] neighbor PEER next-hop-self [all]
774 This command specifies an announced route's nexthop as being equivalent to
775 the address of the bgp router if it is learned via eBGP. If the optional
776 keyword `all` is specified the modification is done also for routes learned
779 .. index:: [no] neighbor PEER update-source <IFNAME|ADDRESS>
780 .. clicmd:: [no] neighbor PEER update-source <IFNAME|ADDRESS>
782 Specify the IPv4 source address to use for the :abbr:`BGP` session to this
783 neighbour, may be specified as either an IPv4 address directly or as an
784 interface name (in which case the *zebra* daemon MUST be running in order
785 for *bgpd* to be able to retrieve interface state).
790 neighbor foo update-source 192.168.0.1
791 neighbor bar update-source lo0
794 .. index:: [no] neighbor PEER default-originate
795 .. clicmd:: [no] neighbor PEER default-originate
797 *bgpd*'s default is to not announce the default route (0.0.0.0/0) even if it
798 is in routing table. When you want to announce default routes to the peer,
801 .. index:: neighbor PEER port PORT
802 .. clicmd:: neighbor PEER port PORT
804 .. index:: neighbor PEER send-community
805 .. clicmd:: neighbor PEER send-community
807 .. index:: [no] neighbor PEER weight WEIGHT
808 .. clicmd:: [no] neighbor PEER weight WEIGHT
810 This command specifies a default `weight` value for the neighbor's routes.
812 .. index:: [no] neighbor PEER maximum-prefix NUMBER
813 .. clicmd:: [no] neighbor PEER maximum-prefix NUMBER
815 .. index:: [no] neighbor PEER local-as AS-NUMBER no-prepend
816 .. clicmd:: [no] neighbor PEER local-as AS-NUMBER no-prepend
818 .. index:: [no] neighbor PEER local-as AS-NUMBER no-prepend replace-as
819 .. clicmd:: [no] neighbor PEER local-as AS-NUMBER no-prepend replace-as
821 .. index:: [no] neighbor PEER local-as AS-NUMBER
822 .. clicmd:: [no] neighbor PEER local-as AS-NUMBER
824 Specify an alternate AS for this BGP process when interacting with the
825 specified peer. With no modifiers, the specified local-as is prepended to
826 the received AS_PATH when receiving routing updates from the peer, and
827 prepended to the outgoing AS_PATH (after the process local AS) when
828 transmitting local routes to the peer.
830 If the no-prepend attribute is specified, then the supplied local-as is not
831 prepended to the received AS_PATH.
833 If the replace-as attribute is specified, then only the supplied local-as is
834 prepended to the AS_PATH when transmitting local-route updates to this peer.
836 Note that replace-as can only be specified if no-prepend is.
838 This command is only allowed for eBGP peers.
840 .. index:: [no] neighbor PEER ttl-security hops NUMBER
841 .. clicmd:: [no] neighbor PEER ttl-security hops NUMBER
843 This command enforces Generalized TTL Security Mechanism (GTSM), as
844 specified in RFC 5082. With this command, only neighbors that are the
845 specified number of hops away will be allowed to become neighbors. This
846 command is mutually exclusive with *ebgp-multihop*.
848 .. index:: [no] neighbor PEER capability extended-nexthop
849 .. clicmd:: [no] neighbor PEER capability extended-nexthop
851 Allow bgp to negotiate the extended-nexthop capability with it's peer.
852 If you are peering over a v6 LL address then this capability is turned
853 on automatically. If you are peering over a v6 Global Address then
854 turning on this command will allow BGP to install v4 routes with
855 v6 nexthops if you do not have v4 configured on interfaces.
857 .. index:: [no] bgp fast-external-failover
858 .. clicmd:: [no] bgp fast-external-failover
860 This command causes bgp to not take down ebgp peers immediately
861 when a link flaps. `bgp fast-external-failover` is the default
862 and will not be displayed as part of a `show run`. The no form
863 of the command turns off this ability.
865 .. _bgp-peer-filtering:
870 .. index:: neighbor PEER distribute-list NAME [in|out]
871 .. clicmd:: neighbor PEER distribute-list NAME [in|out]
873 This command specifies a distribute-list for the peer. `direct` is
876 .. index:: neighbor PEER prefix-list NAME [in|out]
877 .. clicmd:: neighbor PEER prefix-list NAME [in|out]
879 .. index:: neighbor PEER filter-list NAME [in|out]
880 .. clicmd:: neighbor PEER filter-list NAME [in|out]
882 .. index:: neighbor PEER route-map NAME [in|out]
883 .. clicmd:: neighbor PEER route-map NAME [in|out]
885 Apply a route-map on the neighbor. `direct` must be `in` or `out`.
887 .. index:: bgp route-reflector allow-outbound-policy
888 .. clicmd:: bgp route-reflector allow-outbound-policy
890 By default, attribute modification via route-map policy out is not reflected
891 on reflected routes. This option allows the modifications to be reflected as
892 well. Once enabled, it affects all reflected routes.
899 Peer groups are used to help improve scaling by generating the same
900 update information to all members of a peer group. Note that this means
901 that the routes generated by a member of a peer group will be sent back
902 to that originating peer with the originator identifier attribute set to
903 indicated the originating peer. All peers not associated with a
904 specific peer group are treated as belonging to a default peer group,
905 and will share updates.
907 .. index:: neighbor WORD peer-group
908 .. clicmd:: neighbor WORD peer-group
910 This command defines a new peer group.
912 .. index:: neighbor PEER peer-group WORD
913 .. clicmd:: neighbor PEER peer-group WORD
915 This command bind specific peer to peer group WORD.
917 .. index:: neighbor PEER solo
918 .. clicmd:: neighbor PEER solo
920 This command is used to indicate that routes advertised by the peer
921 should not be reflected back to the peer. This command only is only
922 meaningful when there is a single peer defined in the peer-group.
924 Capability Negotiation
925 ^^^^^^^^^^^^^^^^^^^^^^
927 .. index:: neighbor PEER strict-capability-match
928 .. clicmd:: neighbor PEER strict-capability-match
930 .. index:: no neighbor PEER strict-capability-match
931 .. clicmd:: no neighbor PEER strict-capability-match
933 Strictly compares remote capabilities and local capabilities. If
934 capabilities are different, send Unsupported Capability error then reset
937 You may want to disable sending Capability Negotiation OPEN message optional
938 parameter to the peer when remote peer does not implement Capability
939 Negotiation. Please use *dont-capability-negotiate* command to disable the
942 .. index:: neighbor PEER dont-capability-negotiate
943 .. clicmd:: neighbor PEER dont-capability-negotiate
945 .. index:: no neighbor PEER dont-capability-negotiate
946 .. clicmd:: no neighbor PEER dont-capability-negotiate
948 Suppress sending Capability Negotiation as OPEN message optional parameter
949 to the peer. This command only affects the peer is configured other than
950 IPv4 unicast configuration.
952 When remote peer does not have capability negotiation feature, remote peer
953 will not send any capabilities at all. In that case, bgp configures the peer
954 with configured capabilities.
956 You may prefer locally configured capabilities more than the negotiated
957 capabilities even though remote peer sends capabilities. If the peer is
958 configured by *override-capability*, *bgpd* ignores received capabilities
959 then override negotiated capabilities with configured values.
961 .. index:: neighbor PEER override-capability
962 .. clicmd:: neighbor PEER override-capability
964 .. index:: no neighbor PEER override-capability
965 .. clicmd:: no neighbor PEER override-capability
967 Override the result of Capability Negotiation with local configuration.
968 Ignore remote peer's capability value.
970 .. _bgp-as-path-access-lists:
975 AS path access list is user defined AS path.
977 .. index:: ip as-path access-list WORD permit|deny LINE
978 .. clicmd:: ip as-path access-list WORD permit|deny LINE
980 This command defines a new AS path access list.
982 .. index:: no ip as-path access-list WORD
983 .. clicmd:: no ip as-path access-list WORD
985 .. index:: no ip as-path access-list WORD permit|deny LINE
986 .. clicmd:: no ip as-path access-list WORD permit|deny LINE
988 .. _bgp-using-as-path-in-route-map:
990 Using AS Path in Route Map
991 --------------------------
993 .. index:: match as-path WORD
994 .. clicmd:: match as-path WORD
997 .. index:: set as-path prepend AS-PATH
998 .. clicmd:: set as-path prepend AS-PATH
1000 Prepend the given string of AS numbers to the AS_PATH.
1002 .. index:: set as-path prepend last-as NUM
1003 .. clicmd:: set as-path prepend last-as NUM
1005 Prepend the existing last AS number (the leftmost ASN) to the AS_PATH.
1007 .. _bgp-communities-attribute:
1009 Communities Attribute
1010 ---------------------
1012 The BGP communities attribute is widely used for implementing policy routing.
1013 Network operators can manipulate BGP communities attribute based on their
1014 network policy. BGP communities attribute is defined in :rfc:`1997` and
1015 :rfc:`1998`. It is an optional transitive attribute, therefore local policy can
1016 travel through different autonomous system.
1018 The communities attribute is a set of communities values. Each community value
1019 is 4 octet long. The following format is used to define the community value.
1022 This format represents 4 octet communities value. ``AS`` is high order 2
1023 octet in digit format. ``VAL`` is low order 2 octet in digit format. This
1024 format is useful to define AS oriented policy value. For example,
1025 ``7675:80`` can be used when AS 7675 wants to pass local policy value 80 to
1029 ``internet`` represents well-known communities value 0.
1031 ``graceful-shutdown``
1032 ``graceful-shutdown`` represents well-known communities value
1033 ``GRACEFUL_SHUTDOWN`` ``0xFFFF0000`` ``65535:0``. :rfc:`8326` implements
1034 the purpose Graceful BGP Session Shutdown to reduce the amount of
1035 lost traffic when taking BGP sessions down for maintainance. The use
1036 of the community needs to be supported from your peers side to
1037 actually have any effect.
1040 ``accept-own`` represents well-known communities value ``ACCEPT_OWN``
1041 ``0xFFFF0001`` ``65535:1``. :rfc:`7611` implements a way to signal
1042 to a router to accept routes with a local nexthop address. This
1043 can be the case when doing policing and having traffic having a
1044 nexthop located in another VRF but still local interface to the
1045 router. It is recommended to read the RFC for full details.
1047 ``route-filter-translated-v4``
1048 ``route-filter-translated-v4`` represents well-known communities value
1049 ``ROUTE_FILTER_TRANSLATED_v4`` ``0xFFFF0002`` ``65535:2``.
1052 ``route-filter-v4`` represents well-known communities value
1053 ``ROUTE_FILTER_v4`` ``0xFFFF0003`` ``65535:3``.
1055 ``route-filter-translated-v6``
1056 ``route-filter-translated-v6`` represents well-known communities value
1057 ``ROUTE_FILTER_TRANSLATED_v6`` ``0xFFFF0004`` ``65535:4``.
1060 ``route-filter-v6`` represents well-known communities value
1061 ``ROUTE_FILTER_v6`` ``0xFFFF0005`` ``65535:5``.
1064 ``llgr-stale`` represents well-known communities value ``LLGR_STALE``
1065 ``0xFFFF0006`` ``65535:6``.
1066 Assigned and intented only for use with routers supporting the
1067 Long-lived Graceful Restart Capability as described in
1068 :rfc:`draft-uttaro-idr-bgp-persistence`.
1069 Routers recieving routes with this community may (depending on
1070 implementation) choose allow to reject or modify routes on the
1071 presence or absence of this community.
1074 ``no-llgr`` represents well-known communities value ``NO_LLGR``
1075 ``0xFFFF0007`` ``65535:7``.
1076 Assigned and intented only for use with routers supporting the
1077 Long-lived Graceful Restart Capability as described in
1078 :rfc:`draft-uttaro-idr-bgp-persistence`.
1079 Routers recieving routes with this community may (depending on
1080 implementation) choose allow to reject or modify routes on the
1081 presence or absence of this community.
1083 ``accept-own-nexthop``
1084 ``accept-own-nexthop`` represents well-known communities value
1085 ``accept-own-nexthop`` ``0xFFFF0008`` ``65535:8``.
1086 :rfc:`draft-agrewal-idr-accept-own-nexthop` describes
1087 how to tag and label VPN routes to be able to send traffic between VRFs
1088 via an internal layer 2 domain on the same PE device. Refer to
1089 :rfc:`draft-agrewal-idr-accept-own-nexthop` for full details.
1092 ``blackhole`` represents well-known communities value ``BLACKHOLE``
1093 ``0xFFFF029A`` ``65535:666``. :rfc:`7999` documents sending prefixes to
1094 EBGP peers and upstream for the purpose of blackholing traffic.
1095 Prefixes tagged with the this community should normally not be
1096 re-advertised from neighbors of the originating network. It is
1097 recommended upon receiving prefixes tagged with this community to
1098 add ``NO_EXPORT`` and ``NO_ADVERTISE``.
1101 ``no-export`` represents well-known communities value ``NO_EXPORT``
1102 ``0xFFFFFF01``. All routes carry this value must not be advertised to
1103 outside a BGP confederation boundary. If neighboring BGP peer is part of BGP
1104 confederation, the peer is considered as inside a BGP confederation
1105 boundary, so the route will be announced to the peer.
1108 ``no-advertise`` represents well-known communities value ``NO_ADVERTISE``
1109 ``0xFFFFFF02``. All routes carry this value must not be advertise to other
1113 ``local-AS`` represents well-known communities value ``NO_EXPORT_SUBCONFED``
1114 ``0xFFFFFF03``. All routes carry this value must not be advertised to
1115 external BGP peers. Even if the neighboring router is part of confederation,
1116 it is considered as external BGP peer, so the route will not be announced to
1120 ``no-peer`` represents well-known communities value ``NOPEER``
1121 ``0xFFFFFF04`` ``65535:65284``. :rfc:`3765` is used to communicate to
1122 another network how the originating network want the prefix propagated.
1124 When the communities attribute is received duplicate community values in the
1125 attribute are ignored and value is sorted in numerical order.
1127 .. _bgp-community-lists:
1131 Community lists are user defined lists of community attribute values. These
1132 lists can be used for matching or manipulating the communities attribute in
1135 There are two types of community list:
1138 This type accepts an explicit value for the atttribute.
1141 This type accepts a regular expression. Because the regex must be
1142 interpreted on each use expanded community lists are slower than standard
1145 .. index:: ip community-list standard NAME permit|deny COMMUNITY
1146 .. clicmd:: ip community-list standard NAME permit|deny COMMUNITY
1148 This command defines a new standard community list. ``COMMUNITY`` is
1149 communities value. The ``COMMUNITY`` is compiled into community structure.
1150 We can define multiple community list under same name. In that case match
1151 will happen user defined order. Once the community list matches to
1152 communities attribute in BGP updates it return permit or deny by the
1153 community list definition. When there is no matched entry, deny will be
1154 returned. When ``COMMUNITY`` is empty it matches to any routes.
1156 .. index:: ip community-list expanded NAME permit|deny COMMUNITY
1157 .. clicmd:: ip community-list expanded NAME permit|deny COMMUNITY
1159 This command defines a new expanded community list. ``COMMUNITY`` is a
1160 string expression of communities attribute. ``COMMUNITY`` can be a regular
1161 expression (:ref:`bgp-regular-expressions`) to match the communities
1162 attribute in BGP updates.
1165 It is recommended to use the more explicit versions of this command.
1167 .. index:: ip community-list NAME permit|deny COMMUNITY
1168 .. clicmd:: ip community-list NAME permit|deny COMMUNITY
1170 When the community list type is not specified, the community list type is
1171 automatically detected. If ``COMMUNITY`` can be compiled into communities
1172 attribute, the community list is defined as a standard community list.
1173 Otherwise it is defined as an expanded community list. This feature is left
1174 for backward compatibility. Use of this feature is not recommended.
1177 .. index:: no ip community-list [standard|expanded] NAME
1178 .. clicmd:: no ip community-list [standard|expanded] NAME
1180 Deletes the community list specified by ``NAME``. All community lists share
1181 the same namespace, so it's not necessary to specify ``standard`` or
1182 ``expanded``; these modifiers are purely aesthetic.
1184 .. index:: show ip community-list [NAME]
1185 .. clicmd:: show ip community-list [NAME]
1187 Displays community list information. When ``NAME`` is specified the
1188 specified community list's information is shown.
1192 # show ip community-list
1193 Named Community standard list CLIST
1194 permit 7675:80 7675:100 no-export
1196 Named Community expanded list EXPAND
1199 # show ip community-list CLIST
1200 Named Community standard list CLIST
1201 permit 7675:80 7675:100 no-export
1205 .. _bgp-numbered-community-lists:
1207 Numbered Community Lists
1208 ^^^^^^^^^^^^^^^^^^^^^^^^
1210 When number is used for BGP community list name, the number has
1211 special meanings. Community list number in the range from 1 and 99 is
1212 standard community list. Community list number in the range from 100
1213 to 199 is expanded community list. These community lists are called
1214 as numbered community lists. On the other hand normal community lists
1215 is called as named community lists.
1217 .. index:: ip community-list (1-99) permit|deny COMMUNITY
1218 .. clicmd:: ip community-list (1-99) permit|deny COMMUNITY
1220 This command defines a new community list. The argument to (1-99) defines
1221 the list identifier.
1223 .. index:: ip community-list (100-199) permit|deny COMMUNITY
1224 .. clicmd:: ip community-list (100-199) permit|deny COMMUNITY
1226 This command defines a new expanded community list. The argument to
1227 (100-199) defines the list identifier.
1229 .. _bgp-using-communities-in-route-map:
1231 Using Communities in Route Maps
1232 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1234 In :ref:`route-map` we can match on or set the BGP communities attribute. Using
1235 this feature network operator can implement their network policy based on BGP
1236 communities attribute.
1238 The ollowing commands can be used in route maps:
1240 .. index:: match community WORD exact-match [exact-match]
1241 .. clicmd:: match community WORD exact-match [exact-match]
1243 This command perform match to BGP updates using community list WORD. When
1244 the one of BGP communities value match to the one of communities value in
1245 community list, it is match. When `exact-match` keyword is specified, match
1246 happen only when BGP updates have completely same communities value
1247 specified in the community list.
1249 .. index:: set community <none|COMMUNITY> additive
1250 .. clicmd:: set community <none|COMMUNITY> additive
1252 This command sets the community value in BGP updates. If the attribute is
1253 already configured, the newly provided value replaces the old one unless the
1254 ``additive`` keyword is specified, in which case the new value is appended
1255 to the existing value.
1257 If ``none`` is specified as the community value, the communities attribute
1260 .. index:: set comm-list WORD delete
1261 .. clicmd:: set comm-list WORD delete
1263 This command remove communities value from BGP communities attribute. The
1264 ``word`` is community list name. When BGP route's communities value matches
1265 to the community list ``word``, the communities value is removed. When all
1266 of communities value is removed eventually, the BGP update's communities
1267 attribute is completely removed.
1269 .. _bgp-communities-example:
1271 Example Configuration
1272 ^^^^^^^^^^^^^^^^^^^^^
1274 The following configuration is exemplary of the most typical usage of BGP
1275 communities attribute. In the example, AS 7675 provides an upstream Internet
1276 connection to AS 100. When the following configuration exists in AS 7675, the
1277 network operator of AS 100 can set local preference in AS 7675 network by
1278 setting BGP communities attribute to the updates.
1283 neighbor 192.168.0.1 remote-as 100
1284 address-family ipv4 unicast
1285 neighbor 192.168.0.1 route-map RMAP in
1288 ip community-list 70 permit 7675:70
1289 ip community-list 70 deny
1290 ip community-list 80 permit 7675:80
1291 ip community-list 80 deny
1292 ip community-list 90 permit 7675:90
1293 ip community-list 90 deny
1295 route-map RMAP permit 10
1297 set local-preference 70
1299 route-map RMAP permit 20
1301 set local-preference 80
1303 route-map RMAP permit 30
1305 set local-preference 90
1308 The following configuration announces ``10.0.0.0/8`` from AS 100 to AS 7675.
1309 The route has communities value ``7675:80`` so when above configuration exists
1310 in AS 7675, the announced routes' local preference value will be set to 80.
1316 neighbor 192.168.0.2 remote-as 7675
1317 address-family ipv4 unicast
1318 neighbor 192.168.0.2 route-map RMAP out
1321 ip prefix-list PLIST permit 10.0.0.0/8
1323 route-map RMAP permit 10
1324 match ip address prefix-list PLIST
1325 set community 7675:80
1328 The following configuration is an example of BGP route filtering using
1329 communities attribute. This configuration only permit BGP routes which has BGP
1330 communities value ``0:80`` or ``0:90``. The network operator can set special
1331 internal communities value at BGP border router, then limit the BGP route
1332 announcements into the internal network.
1337 neighbor 192.168.0.1 remote-as 100
1338 address-family ipv4 unicast
1339 neighbor 192.168.0.1 route-map RMAP in
1342 ip community-list 1 permit 0:80 0:90
1344 route-map RMAP permit in
1348 The following example filters BGP routes which have a community value of
1349 ``1:1``. When there is no match community-list returns ``deny``. To avoid
1350 filtering all routes, a ``permit`` line is set at the end of the
1356 neighbor 192.168.0.1 remote-as 100
1357 address-family ipv4 unicast
1358 neighbor 192.168.0.1 route-map RMAP in
1361 ip community-list standard FILTER deny 1:1
1362 ip community-list standard FILTER permit
1364 route-map RMAP permit 10
1365 match community FILTER
1368 The communities value keyword ``internet`` has special meanings in standard
1369 community lists. In the below example ``internet`` matches all BGP routes even
1370 if the route does not have communities attribute at all. So community list
1371 ``INTERNET`` is the same as ``FILTER`` in the previous example.
1375 ip community-list standard INTERNET deny 1:1
1376 ip community-list standard INTERNET permit internet
1379 The following configuration is an example of communities value deletion. With
1380 this configuration the community values ``100:1`` and ``100:2`` are removed
1381 from BGP updates. For communities value deletion, only ``permit``
1382 community-list is used. ``deny`` community-list is ignored.
1387 neighbor 192.168.0.1 remote-as 100
1388 address-family ipv4 unicast
1389 neighbor 192.168.0.1 route-map RMAP in
1392 ip community-list standard DEL permit 100:1 100:2
1394 route-map RMAP permit 10
1395 set comm-list DEL delete
1398 .. _bgp-extended-communities-attribute:
1400 Extended Communities Attribute
1401 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1403 BGP extended communities attribute is introduced with MPLS VPN/BGP technology.
1404 MPLS VPN/BGP expands capability of network infrastructure to provide VPN
1405 functionality. At the same time it requires a new framework for policy routing.
1406 With BGP Extended Communities Attribute we can use Route Target or Site of
1407 Origin for implementing network policy for MPLS VPN/BGP.
1409 BGP Extended Communities Attribute is similar to BGP Communities Attribute. It
1410 is an optional transitive attribute. BGP Extended Communities Attribute can
1411 carry multiple Extended Community value. Each Extended Community value is
1414 BGP Extended Communities Attribute provides an extended range compared with BGP
1415 Communities Attribute. Adding to that there is a type field in each value to
1416 provides community space structure.
1418 There are two format to define Extended Community value. One is AS based format
1419 the other is IP address based format.
1422 This is a format to define AS based Extended Community value. ``AS`` part
1423 is 2 octets Global Administrator subfield in Extended Community value.
1424 ``VAL`` part is 4 octets Local Administrator subfield. ``7675:100``
1425 represents AS 7675 policy value 100.
1428 This is a format to define IP address based Extended Community value.
1429 ``IP-Address`` part is 4 octets Global Administrator subfield. ``VAL`` part
1430 is 2 octets Local Administrator subfield.
1432 .. _bgp-extended-community-lists:
1434 Extended Community Lists
1435 ^^^^^^^^^^^^^^^^^^^^^^^^
1437 .. index:: ip extcommunity-list standard NAME permit|deny EXTCOMMUNITY
1438 .. clicmd:: ip extcommunity-list standard NAME permit|deny EXTCOMMUNITY
1440 This command defines a new standard extcommunity-list. `extcommunity` is
1441 extended communities value. The `extcommunity` is compiled into extended
1442 community structure. We can define multiple extcommunity-list under same
1443 name. In that case match will happen user defined order. Once the
1444 extcommunity-list matches to extended communities attribute in BGP updates
1445 it return permit or deny based upon the extcommunity-list definition. When
1446 there is no matched entry, deny will be returned. When `extcommunity` is
1447 empty it matches to any routes.
1449 .. index:: ip extcommunity-list expanded NAME permit|deny LINE
1450 .. clicmd:: ip extcommunity-list expanded NAME permit|deny LINE
1452 This command defines a new expanded extcommunity-list. `line` is a string
1453 expression of extended communities attribute. `line` can be a regular
1454 expression (:ref:`bgp-regular-expressions`) to match an extended communities
1455 attribute in BGP updates.
1457 .. index:: no ip extcommunity-list NAME
1458 .. clicmd:: no ip extcommunity-list NAME
1460 .. index:: no ip extcommunity-list standard NAME
1461 .. clicmd:: no ip extcommunity-list standard NAME
1463 .. index:: no ip extcommunity-list expanded NAME
1464 .. clicmd:: no ip extcommunity-list expanded NAME
1466 These commands delete extended community lists specified by `name`. All of
1467 extended community lists shares a single name space. So extended community
1468 lists can be removed simply specifying the name.
1470 .. index:: show ip extcommunity-list
1471 .. clicmd:: show ip extcommunity-list
1473 .. index:: show ip extcommunity-list NAME
1474 .. clicmd:: show ip extcommunity-list NAME
1476 This command displays current extcommunity-list information. When `name` is
1477 specified the community list's information is shown.::
1479 # show ip extcommunity-list
1482 .. _bgp-extended-communities-in-route-map:
1484 BGP Extended Communities in Route Map
1485 """""""""""""""""""""""""""""""""""""
1487 .. index:: match extcommunity WORD
1488 .. clicmd:: match extcommunity WORD
1490 .. index:: set extcommunity rt EXTCOMMUNITY
1491 .. clicmd:: set extcommunity rt EXTCOMMUNITY
1493 This command set Route Target value.
1495 .. index:: set extcommunity soo EXTCOMMUNITY
1496 .. clicmd:: set extcommunity soo EXTCOMMUNITY
1498 This command set Site of Origin value.
1500 .. _bgp-large-communities-attribute:
1502 Large Communities Attribute
1503 ^^^^^^^^^^^^^^^^^^^^^^^^^^^
1505 The BGP Large Communities attribute was introduced in Feb 2017 with
1508 The BGP Large Communities Attribute is similar to the BGP Communities Attribute
1509 except that it has 3 components instead of two and each of which are 4 octets
1510 in length. Large Communities bring additional functionality and convenience
1511 over traditional communities, specifically the fact that the ``GLOBAL`` part
1512 below is now 4 octets wide allowing seamless use in networks using 4-byte ASNs.
1514 ``GLOBAL:LOCAL1:LOCAL2``
1515 This is the format to define Large Community values. Referencing :rfc:`8195`
1516 the values are commonly referred to as follows:
1518 - The ``GLOBAL`` part is a 4 octet Global Administrator field, commonly used
1519 as the operators AS number.
1520 - The ``LOCAL1`` part is a 4 octet Local Data Part 1 subfield referred to as
1522 - The ``LOCAL2`` part is a 4 octet Local Data Part 2 field and referred to
1523 as the parameter subfield.
1525 As an example, ``65551:1:10`` represents AS 65551 function 1 and parameter
1526 10. The referenced RFC above gives some guidelines on recommended usage.
1528 .. _bgp-large-community-lists:
1530 Large Community Lists
1531 """""""""""""""""""""
1533 Two types of large community lists are supported, namely `standard` and
1536 .. index:: ip large-community-list standard NAME permit|deny LARGE-COMMUNITY
1537 .. clicmd:: ip large-community-list standard NAME permit|deny LARGE-COMMUNITY
1539 This command defines a new standard large-community-list. `large-community`
1540 is the Large Community value. We can add multiple large communities under
1541 same name. In that case the match will happen in the user defined order.
1542 Once the large-community-list matches the Large Communities attribute in BGP
1543 updates it will return permit or deny based upon the large-community-list
1544 definition. When there is no matched entry, a deny will be returned. When
1545 `large-community` is empty it matches any routes.
1547 .. index:: ip large-community-list expanded NAME permit|deny LINE
1548 .. clicmd:: ip large-community-list expanded NAME permit|deny LINE
1550 This command defines a new expanded large-community-list. Where `line` is a
1551 string matching expression, it will be compared to the entire Large
1552 Communities attribute as a string, with each large-community in order from
1553 lowest to highest. `line` can also be a regular expression which matches
1554 this Large Community attribute.
1556 .. index:: no ip large-community-list NAME
1557 .. clicmd:: no ip large-community-list NAME
1559 .. index:: no ip large-community-list standard NAME
1560 .. clicmd:: no ip large-community-list standard NAME
1562 .. index:: no ip large-community-list expanded NAME
1563 .. clicmd:: no ip large-community-list expanded NAME
1565 These commands delete Large Community lists specified by `name`. All Large
1566 Community lists share a single namespace. This means Large Community lists
1567 can be removed by simply specifying the name.
1569 .. index:: show ip large-community-list
1570 .. clicmd:: show ip large-community-list
1572 .. index:: show ip large-community-list NAME
1573 .. clicmd:: show ip large-community-list NAME
1575 This command display current large-community-list information. When
1576 `name` is specified the community list information is shown.
1578 .. index:: show ip bgp large-community-info
1579 .. clicmd:: show ip bgp large-community-info
1581 This command displays the current large communities in use.
1583 .. _bgp-large-communities-in-route-map:
1585 Large Communities in Route Map
1586 """"""""""""""""""""""""""""""
1588 .. index:: match large-community LINE
1589 .. clicmd:: match large-community LINE
1591 Where `line` can be a simple string to match, or a regular expression. It
1592 is very important to note that this match occurs on the entire
1593 large-community string as a whole, where each large-community is ordered
1594 from lowest to highest.
1596 .. index:: set large-community LARGE-COMMUNITY
1597 .. clicmd:: set large-community LARGE-COMMUNITY
1599 .. index:: set large-community LARGE-COMMUNITY LARGE-COMMUNITY
1600 .. clicmd:: set large-community LARGE-COMMUNITY LARGE-COMMUNITY
1602 .. index:: set large-community LARGE-COMMUNITY additive
1603 .. clicmd:: set large-community LARGE-COMMUNITY additive
1605 These commands are used for setting large-community values. The first
1606 command will overwrite any large-communities currently present.
1607 The second specifies two large-communities, which overwrites the current
1608 large-community list. The third will add a large-community value without
1609 overwriting other values. Multiple large-community values can be specified.
1617 BGP supports multiple VRF instances with the following command:
1619 .. index:: router bgp ASN vrf VRFNAME
1620 .. clicmd:: router bgp ASN vrf VRFNAME
1622 ``VRFNAME`` is matched against VRFs configured in the kernel. When
1623 ``vrf VRFNAME`` is not specified, the BGP protocol process belongs to the
1626 With VRF, you can isolate networking information. Having BGP VRF allows you to
1627 have several BGP instances on the same system process. This solution solves
1628 scalabiliy issues where the network administrator had previously to run
1629 separately several BGP processes on each namespace. Now, not only BGP VRF
1630 solves this, but also this method applies to both kind of VRFs backend: default
1631 VRF from Linux kernel or network namespaces. Also, having separate BGP
1632 instances does not imply that the AS number has to be different. For internal
1633 purposes, it is possible to do iBGP peering from two differents network
1639 BGP routes may be leaked (i.e. copied) between a unicast VRF RIB and the VPN
1640 SAFI RIB of the default VRF for use in MPLS-based L3VPNs. Unicast routes may
1641 also be leaked between any VRFs (including the unicast RIB of the default BGP
1642 instanced). A shortcut syntax is also available for specifying leaking from one
1643 VRF to another VRF using the default instance's VPN RIB as the intemediary. A
1644 common application of the VRF-VRF feature is to connect a customer's private
1645 routing domain to a provider's VPN service. Leaking is configured from the
1646 point of view of an individual VRF: ``import`` refers to routes leaked from VPN
1647 to a unicast VRF, whereas ``export`` refers to routes leaked from a unicast VRF
1653 Routes exported from a unicast VRF to the VPN RIB must be augmented by two
1656 - an :abbr:`RD (Route Distinguisher)`
1657 - an :abbr:`RTLIST (Route-target List)`
1659 Configuration for these exported routes must, at a minimum, specify these two
1662 Routes imported from the VPN RIB to a unicast VRF are selected according to
1663 their RTLISTs. Routes whose RTLIST contains at least one route-target in
1664 common with the configured import RTLIST are leaked. Configuration for these
1665 imported routes must specify an RTLIST to be matched.
1667 The RD, which carries no semantic value, is intended to make the route unique
1668 in the VPN RIB among all routes of its prefix that originate from all the
1669 customers and sites that are attached to the provider's VPN service.
1670 Accordingly, each site of each customer is typically assigned an RD that is
1671 unique across the entire provider network.
1673 The RTLIST is a set of route-target extended community values whose purpose is
1674 to specify route-leaking policy. Typically, a customer is assigned a single
1675 route-target value for import and export to be used at all customer sites. This
1676 configuration specifies a simple topology wherein a customer has a single
1677 routing domain which is shared across all its sites. More complex routing
1678 topologies are possible through use of additional route-targets to augment the
1679 leaking of sets of routes in various ways.
1681 When using the shortcut syntax for vrf-to-vrf leaking, the RD and RT are
1684 General configuration
1685 """""""""""""""""""""
1687 Configuration of route leaking between a unicast VRF RIB and the VPN SAFI RIB
1688 of the default VRF is accomplished via commands in the context of a VRF
1691 .. index:: rd vpn export AS:NN|IP:nn
1692 .. clicmd:: rd vpn export AS:NN|IP:nn
1694 Specifies the route distinguisher to be added to a route exported from the
1695 current unicast VRF to VPN.
1697 .. index:: no rd vpn export [AS:NN|IP:nn]
1698 .. clicmd:: no rd vpn export [AS:NN|IP:nn]
1700 Deletes any previously-configured export route distinguisher.
1702 .. index:: rt vpn import|export|both RTLIST...
1703 .. clicmd:: rt vpn import|export|both RTLIST...
1705 Specifies the route-target list to be attached to a route (export) or the
1706 route-target list to match against (import) when exporting/importing between
1707 the current unicast VRF and VPN.
1709 The RTLIST is a space-separated list of route-targets, which are BGP
1710 extended community values as described in
1711 :ref:`bgp-extended-communities-attribute`.
1713 .. index:: no rt vpn import|export|both [RTLIST...]
1714 .. clicmd:: no rt vpn import|export|both [RTLIST...]
1716 Deletes any previously-configured import or export route-target list.
1718 .. index:: label vpn export (0..1048575)|auto
1719 .. clicmd:: label vpn export (0..1048575)|auto
1721 Specifies an optional MPLS label to be attached to a route exported from the
1722 current unicast VRF to VPN. If label is specified as ``auto``, the label
1723 value is automatically assigned from a pool maintained by the zebra
1724 daemon. If zebra is not running, automatic label assignment will not
1725 complete, which will block corresponding route export.
1727 .. index:: no label vpn export [(0..1048575)|auto]
1728 .. clicmd:: no label vpn export [(0..1048575)|auto]
1730 Deletes any previously-configured export label.
1732 .. index:: nexthop vpn export A.B.C.D|X:X::X:X
1733 .. clicmd:: nexthop vpn export A.B.C.D|X:X::X:X
1735 Specifies an optional nexthop value to be assigned to a route exported from
1736 the current unicast VRF to VPN. If left unspecified, the nexthop will be set
1737 to 0.0.0.0 or 0:0::0:0 (self).
1739 .. index:: no nexthop vpn export [A.B.C.D|X:X::X:X]
1740 .. clicmd:: no nexthop vpn export [A.B.C.D|X:X::X:X]
1742 Deletes any previously-configured export nexthop.
1744 .. index:: route-map vpn import|export MAP
1745 .. clicmd:: route-map vpn import|export MAP
1747 Specifies an optional route-map to be applied to routes imported or exported
1748 between the current unicast VRF and VPN.
1750 .. index:: no route-map vpn import|export [MAP]
1751 .. clicmd:: no route-map vpn import|export [MAP]
1753 Deletes any previously-configured import or export route-map.
1755 .. index:: import|export vpn
1756 .. clicmd:: import|export vpn
1758 Enables import or export of routes between the current unicast VRF and VPN.
1760 .. index:: no import|export vpn
1761 .. clicmd:: no import|export vpn
1763 Disables import or export of routes between the current unicast VRF and VPN.
1765 .. index:: import vrf VRFNAME
1766 .. clicmd:: import vrf VRFNAME
1768 Shortcut syntax for specifying automatic leaking from vrf VRFNAME to
1769 the current VRF using the VPN RIB as intermediary. The RD and RT
1770 are auto derived and should not be specified explicitly for either the
1771 source or destination VRF's.
1773 This shortcut syntax mode is not compatible with the explicit
1774 `import vpn` and `export vpn` statements for the two VRF's involved.
1775 The CLI will disallow attempts to configure incompatible leaking
1778 .. index:: no import vrf VRFNAME
1779 .. clicmd:: no import vrf VRFNAME
1781 Disables automatic leaking from vrf VRFNAME to the current VRF using
1782 the VPN RIB as intermediary.
1784 .. _bgp-instances-and-views:
1789 A BGP *instance* is a normal BGP process. Routes selected by BGP are installed
1790 into the kernel routing table.
1793 In previous versions of FRR, running multiple AS's from the same BGP process
1794 was not supported; in order to run multiple AS's it was necessary to run
1795 multiple BGP processes. This had to be explicitly configured with the
1796 ``bgp multiple-instance`` command. Recent versions of FRR support multiple
1797 BGP AS's within the same process by simply defining multiple
1798 ``router bgp X`` blocks, so the ``multiple-instance`` command is now
1799 unnecessary and deprecated.
1801 .. index:: router bgp AS-NUMBER
1802 .. clicmd:: router bgp AS-NUMBER
1804 Make a new BGP instance. You can use an arbitrary word for the `name`.
1809 neighbor 10.0.0.1 remote-as 2
1810 neighbor 10.0.0.2 remote-as 3
1813 neighbor 10.0.0.3 remote-as 4
1814 neighbor 10.0.0.4 remote-as 5
1817 This command does nothing and can be safely removed.
1819 .. index:: bgp multiple-instance
1820 .. clicmd:: bgp multiple-instance
1822 Enable BGP multiple instance feature. Because this is now the default
1823 configuration this command will not be displayed in the running
1827 This command does nothing and can be safely removed.
1829 .. index:: no bgp multiple-instance
1830 .. clicmd:: no bgp multiple-instance
1832 In previous versions of FRR, this command disabled the BGP multiple instance
1833 feature. This functionality is automatically turned on when BGP multiple
1834 instances or views exist so this command no longer does anything.
1836 BGP views are almost same as normal BGP processes, except that routes selected
1837 by BGP are not installed into the kernel routing table. The view functionality
1838 allows the exchange of BGP routing information only without affecting the
1839 kernel routing tables.
1841 .. index:: router bgp AS-NUMBER view NAME
1842 .. clicmd:: router bgp AS-NUMBER view NAME
1844 Make a new BGP view. You can use arbitrary word for the ``NAME``. Routes selected by the view are not installed into the kernel routing table.
1845 view's route selection result does not go to the kernel routing table.
1847 With this command, you can setup Route Server like below.
1853 neighbor 10.0.0.1 remote-as 2
1854 neighbor 10.0.0.2 remote-as 3
1857 neighbor 10.0.0.3 remote-as 4
1858 neighbor 10.0.0.4 remote-as 5
1860 .. index:: show [ip] bgp view NAME
1861 .. clicmd:: show [ip] bgp view NAME
1863 Display the routing table of BGP view ``NAME``.
1865 .. _bgp-cisco-compatibility:
1870 FRR has commands that change some configuration syntax and default behavior to
1871 behave more closely to Cisco conventions. These are deprecated and will be
1872 removed in a future version of FRR.
1875 Please transition to using the FRR specific syntax for your configuration.
1877 .. index:: bgp config-type cisco
1878 .. clicmd:: bgp config-type cisco
1880 Cisco compatible BGP configuration output.
1882 When this configuration line is specified:
1884 - ``no synchronization`` is displayed. This command does nothing and is for
1885 display purposes only.
1886 - ``no auto-summary`` is displayed.
1887 - The ``network`` and ``aggregate-address`` arguments are displayed as:
1893 FRR: network 10.0.0.0/8
1894 Cisco: network 10.0.0.0
1896 FRR: aggregate-address 192.168.0.0/24
1897 Cisco: aggregate-address 192.168.0.0 255.255.255.0
1899 Community attribute handling is also different. If no configuration is
1900 specified community attribute and extended community attribute are sent to
1901 the neighbor. If a user manually disables the feature, the community
1902 attribute is not sent to the neighbor. When ``bgp config-type cisco`` is
1903 specified, the community attribute is not sent to the neighbor by default.
1904 To send the community attribute user has to specify
1905 :clicmd:`neighbor A.B.C.D send-community` like so:
1911 neighbor 10.0.0.1 remote-as 1
1912 address-family ipv4 unicast
1913 no neighbor 10.0.0.1 send-community
1917 neighbor 10.0.0.1 remote-as 1
1918 address-family ipv4 unicast
1919 neighbor 10.0.0.1 send-community
1924 Please transition to using the FRR specific syntax for your configuration.
1926 .. index:: bgp config-type zebra
1927 .. clicmd:: bgp config-type zebra
1929 FRR style BGP configuration. This is the default.
1936 .. index:: show debug
1937 .. clicmd:: show debug
1939 Show all enabled debugs.
1941 .. index:: [no] debug bgp neighbor-events
1942 .. clicmd:: [no] debug bgp neighbor-events
1944 Enable or disable debugging for neighbor events. This provides general
1945 information on BGP events such as peer connection / disconnection, session
1946 establishment / teardown, and capability negotiation.
1948 .. index:: [no] debug bgp updates
1949 .. clicmd:: [no] debug bgp updates
1951 Enable or disable debugging for BGP updates. This provides information on
1952 BGP UPDATE messages transmitted and received between local and remote
1955 .. index:: [no] debug bgp keepalives
1956 .. clicmd:: [no] debug bgp keepalives
1958 Enable or disable debugging for BGP keepalives. This provides information on
1959 BGP KEEPALIVE messages transmitted and received between local and remote
1962 .. index:: [no] debug bgp bestpath <A.B.C.D/M|X:X::X:X/M>
1963 .. clicmd:: [no] debug bgp bestpath <A.B.C.D/M|X:X::X:X/M>
1965 Enable or disable debugging for bestpath selection on the specified prefix.
1967 .. index:: [no] debug bgp nht
1968 .. clicmd:: [no] debug bgp nht
1970 Enable or disable debugging of BGP nexthop tracking.
1972 .. index:: [no] debug bgp update-groups
1973 .. clicmd:: [no] debug bgp update-groups
1975 Enable or disable debugging of dynamic update groups. This provides general
1976 information on group creation, deletion, join and prune events.
1978 .. index:: [no] debug bgp zebra
1979 .. clicmd:: [no] debug bgp zebra
1981 Enable or disable debugging of communications between *bgpd* and *zebra*.
1983 Dumping Messages and Routing Tables
1984 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
1986 .. index:: dump bgp all PATH [INTERVAL]
1987 .. clicmd:: dump bgp all PATH [INTERVAL]
1989 .. index:: dump bgp all-et PATH [INTERVAL]
1990 .. clicmd:: dump bgp all-et PATH [INTERVAL]
1992 .. index:: no dump bgp all [PATH] [INTERVAL]
1993 .. clicmd:: no dump bgp all [PATH] [INTERVAL]
1995 Dump all BGP packet and events to `path` file.
1996 If `interval` is set, a new file will be created for echo `interval` of
1997 seconds. The path `path` can be set with date and time formatting
1998 (strftime). The type ‘all-et’ enables support for Extended Timestamp Header
1999 (:ref:`packet-binary-dump-format`).
2001 .. index:: dump bgp updates PATH [INTERVAL]
2002 .. clicmd:: dump bgp updates PATH [INTERVAL]
2004 .. index:: dump bgp updates-et PATH [INTERVAL]
2005 .. clicmd:: dump bgp updates-et PATH [INTERVAL]
2007 .. index:: no dump bgp updates [PATH] [INTERVAL]
2008 .. clicmd:: no dump bgp updates [PATH] [INTERVAL]
2010 Dump only BGP updates messages to `path` file.
2011 If `interval` is set, a new file will be created for echo `interval` of
2012 seconds. The path `path` can be set with date and time formatting
2013 (strftime). The type ‘updates-et’ enables support for Extended Timestamp
2014 Header (:ref:`packet-binary-dump-format`).
2016 .. index:: dump bgp routes-mrt PATH
2017 .. clicmd:: dump bgp routes-mrt PATH
2019 .. index:: dump bgp routes-mrt PATH INTERVAL
2020 .. clicmd:: dump bgp routes-mrt PATH INTERVAL
2022 .. index:: no dump bgp route-mrt [PATH] [INTERVAL]
2023 .. clicmd:: no dump bgp route-mrt [PATH] [INTERVAL]
2025 Dump whole BGP routing table to `path`. This is heavy process. The path
2026 `path` can be set with date and time formatting (strftime). If `interval` is
2027 set, a new file will be created for echo `interval` of seconds.
2029 Note: the interval variable can also be set using hours and minutes: 04h20m00.
2032 .. _bgp-other-commands:
2037 .. index:: clear bgp ipv4|ipv6 \*
2038 .. clicmd:: clear bgp ipv4|ipv6 \*
2040 Clear all address family peers.
2042 .. index:: clear bgp ipv4|ipv6 PEER
2043 .. clicmd:: clear bgp ipv4|ipv6 PEER
2045 Clear peers which have addresses of X.X.X.X
2047 .. index:: clear bgp ipv4|ipv6 PEER soft in
2048 .. clicmd:: clear bgp ipv4|ipv6 PEER soft in
2050 Clear peer using soft reconfiguration.
2053 .. _bgp-displaying-bgp-information:
2055 Displaying BGP Information
2056 ==========================
2058 The following four commands display the IPv6 and IPv4 routing tables, depending
2059 on whether or not the ``ip`` keyword is used.
2060 Actually, :clicmd:`show ip bgp` command was used on older `Quagga` routing
2061 daemon project, while :clicmd:`show bgp` command is the new format. The choice
2062 has been done to keep old format with IPv4 routing table, while new format
2063 displays IPv6 routing table.
2065 .. index:: show ip bgp
2066 .. clicmd:: show ip bgp
2068 .. index:: show ip bgp A.B.C.D
2069 .. clicmd:: show ip bgp A.B.C.D
2072 .. clicmd:: show bgp
2074 .. index:: show bgp X:X::X:X
2075 .. clicmd:: show bgp X:X::X:X
2077 These commands display BGP routes. When no route is specified, the default
2078 is to display all BGP routes.
2082 BGP table version is 0, local router ID is 10.1.1.1
2083 Status codes: s suppressed, d damped, h history, * valid, > best, i - internal
2084 Origin codes: i - IGP, e - EGP, ? - incomplete
2086 Network Next Hop Metric LocPrf Weight Path
2087 \*> 1.1.1.1/32 0.0.0.0 0 32768 i
2089 Total number of prefixes 1
2091 Some other commands provide additional options for filtering the output.
2093 .. index:: show [ip] bgp regexp LINE
2094 .. clicmd:: show [ip] bgp regexp LINE
2096 This command displays BGP routes using AS path regular expression
2097 (:ref:`bgp-regular-expressions`).
2099 .. index:: show [ip] bgp summary
2100 .. clicmd:: show [ip] bgp summary
2102 Show a bgp peer summary for the specified address family.
2104 The old command structure :clicmd:`show ip bgp` may be removed in the future
2105 and should no longer be used. In order to reach the other BGP routing tables
2106 other than the IPv6 routing table given by :clicmd:`show bgp`, the new command
2107 structure is extended with :clicmd:`show bgp [afi] [safi]`.
2109 .. index:: show bgp [afi] [safi]
2110 .. clicmd:: show bgp [afi] [safi]
2112 .. index:: show bgp <ipv4|ipv6> <unicast|multicast|vpn|labeled-unicast>
2113 .. clicmd:: show bgp <ipv4|ipv6> <unicast|multicast|vpn|labeled-unicast>
2115 These commands display BGP routes for the specific routing table indicated by
2116 the selected afi and the selected safi. If no afi and no safi value is given,
2117 the command falls back to the default IPv6 routing table
2119 .. index:: show bgp [afi] [safi] summary
2120 .. clicmd:: show bgp [afi] [safi] summary
2122 Show a bgp peer summary for the specified address family, and subsequent
2125 .. index:: show bgp [afi] [safi] neighbor [PEER]
2126 .. clicmd:: show bgp [afi] [safi] neighbor [PEER]
2128 This command shows information on a specific BGP peer of the relevant
2129 afi and safi selected.
2131 .. index:: show bgp [afi] [safi] dampening dampened-paths
2132 .. clicmd:: show bgp [afi] [safi] dampening dampened-paths
2134 Display paths suppressed due to dampening of the selected afi and safi
2137 .. index:: show bgp [afi] [safi] dampening flap-statistics
2138 .. clicmd:: show bgp [afi] [safi] dampening flap-statistics
2140 Display flap statistics of routes of the selected afi and safi selected.
2142 .. _bgp-display-routes-by-community:
2144 Displaying Routes by Community Attribute
2145 ----------------------------------------
2147 The following commands allow displaying routes based on their community
2150 .. index:: show [ip] bgp <ipv4|ipv6> community
2151 .. clicmd:: show [ip] bgp <ipv4|ipv6> community
2153 .. index:: show [ip] bgp <ipv4|ipv6> community COMMUNITY
2154 .. clicmd:: show [ip] bgp <ipv4|ipv6> community COMMUNITY
2156 .. index:: show [ip] bgp <ipv4|ipv6> community COMMUNITY exact-match
2157 .. clicmd:: show [ip] bgp <ipv4|ipv6> community COMMUNITY exact-match
2159 These commands display BGP routes which have the community attribute.
2160 attribute. When ``COMMUNITY`` is specified, BGP routes that match that
2161 community are displayed. When `exact-match` is specified, it display only
2162 routes that have an exact match.
2164 .. index:: show [ip] bgp <ipv4|ipv6> community-list WORD
2165 .. clicmd:: show [ip] bgp <ipv4|ipv6> community-list WORD
2167 .. index:: show [ip] bgp <ipv4|ipv6> community-list WORD exact-match
2168 .. clicmd:: show [ip] bgp <ipv4|ipv6> community-list WORD exact-match
2170 These commands display BGP routes for the address family specified that
2171 match the specified community list. When `exact-match` is specified, it
2172 displays only routes that have an exact match.
2174 .. _bgp-display-routes-by-as-path:
2176 Displaying Routes by AS Path
2177 ----------------------------
2179 .. index:: show bgp ipv4|ipv6 regexp LINE
2180 .. clicmd:: show bgp ipv4|ipv6 regexp LINE
2182 This commands displays BGP routes that matches a regular
2183 expression `line` (:ref:`bgp-regular-expressions`).
2185 .. index:: show [ip] bgp ipv4 vpn
2186 .. clicmd:: show [ip] bgp ipv4 vpn
2188 .. index:: show [ip] bgp ipv6 vpn
2189 .. clicmd:: show [ip] bgp ipv6 vpn
2191 Print active IPV4 or IPV6 routes advertised via the VPN SAFI.
2193 .. index:: show bgp ipv4 vpn summary
2194 .. clicmd:: show bgp ipv4 vpn summary
2196 .. index:: show bgp ipv6 vpn summary
2197 .. clicmd:: show bgp ipv6 vpn summary
2199 Print a summary of neighbor connections for the specified AFI/SAFI combination.
2202 .. _bgp-route-reflector:
2207 .. note:: This documentation is woefully incomplete.
2209 .. index:: bgp cluster-id A.B.C.D
2210 .. clicmd:: bgp cluster-id A.B.C.D
2212 .. index:: neighbor PEER route-reflector-client
2213 .. clicmd:: neighbor PEER route-reflector-client
2215 .. index:: no neighbor PEER route-reflector-client
2216 .. clicmd:: no neighbor PEER route-reflector-client
2224 You can set different routing policy for a peer. For example, you can set
2225 different filter for a peer.
2229 bgp multiple-instance
2232 neighbor 10.0.0.1 remote-as 2
2233 address-family ipv4 unicast
2234 neighbor 10.0.0.1 distribute-list 1 in
2238 neighbor 10.0.0.1 remote-as 2
2239 address-family ipv4 unicast
2240 neighbor 10.0.0.1 distribute-list 2 in
2243 This means BGP update from a peer 10.0.0.1 goes to both BGP view 1 and view 2.
2244 When the update is inserted into view 1, distribute-list 1 is applied. On the
2245 other hand, when the update is inserted into view 2, distribute-list 2 is
2249 .. _bgp-regular-expressions:
2251 BGP Regular Expressions
2252 =======================
2254 BGP regular expressions are based on :t:`POSIX 1003.2` regular expressions. The
2255 following description is just a quick subset of the POSIX regular expressions.
2259 Matches any single character.
2262 Matches 0 or more occurrences of pattern.
2265 Matches 1 or more occurrences of pattern.
2268 Match 0 or 1 occurrences of pattern.
2271 Matches the beginning of the line.
2274 Matches the end of the line.
2277 The ``_`` character has special meanings in BGP regular expressions. It
2278 matches to space and comma , and AS set delimiter ``{`` and ``}`` and AS
2279 confederation delimiter ``(`` and ``)``. And it also matches to the
2280 beginning of the line and the end of the line. So ``_`` can be used for AS
2281 value boundaries match. This character technically evaluates to
2285 .. _bgp-configuration-examples:
2287 Miscellaneous Configuration Examples
2288 ====================================
2290 Example of a session to an upstream, advertising only one prefix to it.
2295 bgp router-id 10.236.87.1
2296 neighbor upstream peer-group
2297 neighbor upstream remote-as 64515
2298 neighbor upstream capability dynamic
2299 neighbor 10.1.1.1 peer-group upstream
2300 neighbor 10.1.1.1 description ACME ISP
2302 address-family ipv4 unicast
2303 network 10.236.87.0/24
2304 neighbor upstream prefix-list pl-allowed-adv out
2307 ip prefix-list pl-allowed-adv seq 5 permit 82.195.133.0/25
2308 ip prefix-list pl-allowed-adv seq 10 deny any
2310 A more complex example including upstream, peer and customer sessions
2311 advertising global prefixes and NO_EXPORT prefixes and providing actions for
2312 customer routes based on community values. Extensive use is made of route-maps
2313 and the 'call' feature to support selective advertising of prefixes. This
2314 example is intended as guidance only, it has NOT been tested and almost
2315 certainly contains silly mistakes, if not serious flaws.
2320 bgp router-id 10.236.87.1
2321 neighbor upstream capability dynamic
2322 neighbor cust capability dynamic
2323 neighbor peer capability dynamic
2324 neighbor 10.1.1.1 remote-as 64515
2325 neighbor 10.1.1.1 peer-group upstream
2326 neighbor 10.2.1.1 remote-as 64516
2327 neighbor 10.2.1.1 peer-group upstream
2328 neighbor 10.3.1.1 remote-as 64517
2329 neighbor 10.3.1.1 peer-group cust-default
2330 neighbor 10.3.1.1 description customer1
2331 neighbor 10.4.1.1 remote-as 64518
2332 neighbor 10.4.1.1 peer-group cust
2333 neighbor 10.4.1.1 description customer2
2334 neighbor 10.5.1.1 remote-as 64519
2335 neighbor 10.5.1.1 peer-group peer
2336 neighbor 10.5.1.1 description peer AS 1
2337 neighbor 10.6.1.1 remote-as 64520
2338 neighbor 10.6.1.1 peer-group peer
2339 neighbor 10.6.1.1 description peer AS 2
2341 address-family ipv4 unicast
2342 network 10.123.456.0/24
2343 network 10.123.456.128/25 route-map rm-no-export
2344 neighbor upstream route-map rm-upstream-out out
2345 neighbor cust route-map rm-cust-in in
2346 neighbor cust route-map rm-cust-out out
2347 neighbor cust send-community both
2348 neighbor peer route-map rm-peer-in in
2349 neighbor peer route-map rm-peer-out out
2350 neighbor peer send-community both
2351 neighbor 10.3.1.1 prefix-list pl-cust1-network in
2352 neighbor 10.4.1.1 prefix-list pl-cust2-network in
2353 neighbor 10.5.1.1 prefix-list pl-peer1-network in
2354 neighbor 10.6.1.1 prefix-list pl-peer2-network in
2357 ip prefix-list pl-default permit 0.0.0.0/0
2359 ip prefix-list pl-upstream-peers permit 10.1.1.1/32
2360 ip prefix-list pl-upstream-peers permit 10.2.1.1/32
2362 ip prefix-list pl-cust1-network permit 10.3.1.0/24
2363 ip prefix-list pl-cust1-network permit 10.3.2.0/24
2365 ip prefix-list pl-cust2-network permit 10.4.1.0/24
2367 ip prefix-list pl-peer1-network permit 10.5.1.0/24
2368 ip prefix-list pl-peer1-network permit 10.5.2.0/24
2369 ip prefix-list pl-peer1-network permit 192.168.0.0/24
2371 ip prefix-list pl-peer2-network permit 10.6.1.0/24
2372 ip prefix-list pl-peer2-network permit 10.6.2.0/24
2373 ip prefix-list pl-peer2-network permit 192.168.1.0/24
2374 ip prefix-list pl-peer2-network permit 192.168.2.0/24
2375 ip prefix-list pl-peer2-network permit 172.16.1/24
2377 ip as-path access-list asp-own-as permit ^$
2378 ip as-path access-list asp-own-as permit _64512_
2380 ! #################################################################
2381 ! Match communities we provide actions for, on routes receives from
2382 ! customers. Communities values of <our-ASN>:X, with X, have actions:
2384 ! 100 - blackhole the prefix
2385 ! 200 - set no_export
2386 ! 300 - advertise only to other customers
2387 ! 400 - advertise only to upstreams
2388 ! 500 - set no_export when advertising to upstreams
2389 ! 2X00 - set local_preference to X00
2391 ! blackhole the prefix of the route
2392 ip community-list standard cm-blackhole permit 64512:100
2394 ! set no-export community before advertising
2395 ip community-list standard cm-set-no-export permit 64512:200
2397 ! advertise only to other customers
2398 ip community-list standard cm-cust-only permit 64512:300
2400 ! advertise only to upstreams
2401 ip community-list standard cm-upstream-only permit 64512:400
2403 ! advertise to upstreams with no-export
2404 ip community-list standard cm-upstream-noexport permit 64512:500
2406 ! set local-pref to least significant 3 digits of the community
2407 ip community-list standard cm-prefmod-100 permit 64512:2100
2408 ip community-list standard cm-prefmod-200 permit 64512:2200
2409 ip community-list standard cm-prefmod-300 permit 64512:2300
2410 ip community-list standard cm-prefmod-400 permit 64512:2400
2411 ip community-list expanded cme-prefmod-range permit 64512:2...
2413 ! Informational communities
2415 ! 3000 - learned from upstream
2416 ! 3100 - learned from customer
2417 ! 3200 - learned from peer
2419 ip community-list standard cm-learnt-upstream permit 64512:3000
2420 ip community-list standard cm-learnt-cust permit 64512:3100
2421 ip community-list standard cm-learnt-peer permit 64512:3200
2423 ! ###################################################################
2424 ! Utility route-maps
2426 ! These utility route-maps generally should not used to permit/deny
2427 ! routes, i.e. they do not have meaning as filters, and hence probably
2428 ! should be used with 'on-match next'. These all finish with an empty
2429 ! permit entry so as not interfere with processing in the caller.
2431 route-map rm-no-export permit 10
2432 set community additive no-export
2433 route-map rm-no-export permit 20
2435 route-map rm-blackhole permit 10
2436 description blackhole, up-pref and ensure it cant escape this AS
2437 set ip next-hop 127.0.0.1
2438 set local-preference 10
2439 set community additive no-export
2440 route-map rm-blackhole permit 20
2442 ! Set local-pref as requested
2443 route-map rm-prefmod permit 10
2444 match community cm-prefmod-100
2445 set local-preference 100
2446 route-map rm-prefmod permit 20
2447 match community cm-prefmod-200
2448 set local-preference 200
2449 route-map rm-prefmod permit 30
2450 match community cm-prefmod-300
2451 set local-preference 300
2452 route-map rm-prefmod permit 40
2453 match community cm-prefmod-400
2454 set local-preference 400
2455 route-map rm-prefmod permit 50
2457 ! Community actions to take on receipt of route.
2458 route-map rm-community-in permit 10
2459 description check for blackholing, no point continuing if it matches.
2460 match community cm-blackhole
2462 route-map rm-community-in permit 20
2463 match community cm-set-no-export
2466 route-map rm-community-in permit 30
2467 match community cme-prefmod-range
2469 route-map rm-community-in permit 40
2471 ! #####################################################################
2472 ! Community actions to take when advertising a route.
2473 ! These are filtering route-maps,
2475 ! Deny customer routes to upstream with cust-only set.
2476 route-map rm-community-filt-to-upstream deny 10
2477 match community cm-learnt-cust
2478 match community cm-cust-only
2479 route-map rm-community-filt-to-upstream permit 20
2481 ! Deny customer routes to other customers with upstream-only set.
2482 route-map rm-community-filt-to-cust deny 10
2483 match community cm-learnt-cust
2484 match community cm-upstream-only
2485 route-map rm-community-filt-to-cust permit 20
2487 ! ###################################################################
2488 ! The top-level route-maps applied to sessions. Further entries could
2489 ! be added obviously..
2492 route-map rm-cust-in permit 10
2493 call rm-community-in
2495 route-map rm-cust-in permit 20
2496 set community additive 64512:3100
2497 route-map rm-cust-in permit 30
2499 route-map rm-cust-out permit 10
2500 call rm-community-filt-to-cust
2502 route-map rm-cust-out permit 20
2504 ! Upstream transit ASes
2505 route-map rm-upstream-out permit 10
2506 description filter customer prefixes which are marked cust-only
2507 call rm-community-filt-to-upstream
2509 route-map rm-upstream-out permit 20
2510 description only customer routes are provided to upstreams/peers
2511 match community cm-learnt-cust
2514 ! outbound policy is same as for upstream
2515 route-map rm-peer-out permit 10
2516 call rm-upstream-out
2518 route-map rm-peer-in permit 10
2519 set community additive 64512:3200
2522 Example of how to set up a 6-Bone connection.
2526 ! bgpd configuration
2527 ! ==================
2529 ! MP-BGP configuration
2532 bgp router-id 10.0.0.1
2533 neighbor 3ffe:1cfa:0:2:2a0:c9ff:fe9e:f56 remote-as `as-number`
2536 network 3ffe:506::/32
2537 neighbor 3ffe:1cfa:0:2:2a0:c9ff:fe9e:f56 activate
2538 neighbor 3ffe:1cfa:0:2:2a0:c9ff:fe9e:f56 route-map set-nexthop out
2539 neighbor 3ffe:1cfa:0:2:2c0:4fff:fe68:a231 remote-as `as-number`
2540 neighbor 3ffe:1cfa:0:2:2c0:4fff:fe68:a231 route-map set-nexthop out
2543 ipv6 access-list all permit any
2545 ! Set output nexthop address.
2547 route-map set-nexthop permit 10
2548 match ipv6 address all
2549 set ipv6 nexthop global 3ffe:1cfa:0:2:2c0:4fff:fe68:a225
2550 set ipv6 nexthop local fe80::2c0:4fff:fe68:a225
2556 .. include:: routeserver.rst
2558 .. include:: rpki.rst
2560 .. include:: flowspec.rst
2562 .. [#med-transitivity-rant] For some set of objects to have an order, there *must* be some binary ordering relation that is defined for *every* combination of those objects, and that relation *must* be transitive. I.e.:, if the relation operator is <, and if a < b and b < c then that relation must carry over and it *must* be that a < c for the objects to have an order. The ordering relation may allow for equality, i.e. a < b and b < a may both be true and imply that a and b are equal in the order and not distinguished by it, in which case the set has a partial order. Otherwise, if there is an order, all the objects have a distinct place in the order and the set has a total order)
2563 .. [bgp-route-osci-cond] McPherson, D. and Gill, V. and Walton, D., "Border Gateway Protocol (BGP) Persistent Route Oscillation Condition", IETF RFC3345
2564 .. [stable-flexible-ibgp] Flavel, A. and M. Roughan, "Stable and flexible iBGP", ACM SIGCOMM 2009
2565 .. [ibgp-correctness] Griffin, T. and G. Wilfong, "On the correctness of IBGP configuration", ACM SIGCOMM 2002