Information About BGP Additional Paths
This section includes the following topics:
Overview
BGP routers and route reflectors (RRs) propagate only their best paths over their sessions. The advertisement of a prefix replaces the previous announcement of that prefix (this behavior is known as an implicit withdraw). The implicit withdraw can achieve better scaling, but at the cost of path diversity.
Path hiding can prevent efficient use of BGP multipath, prevent hitless planned maintenance, and lead to multi-exit discriminator (MED) oscillations and suboptimal hot-potato routing. In a next-hop failure, path hiding inhibits fast and local recovery because the network has to wait for BGP control plane convergence to restore traffic. The BGP Additional Paths feature offers path diversity; the best external or best internal features offer path diversity in limited scenarios.
The BGP Additional Paths feature allows multiple paths for the same prefix to be advertised without the new paths implicitly replacing the previous paths. Path diversity is achieved instead of path hiding.
Path-Hiding Scenario
The following figure shows prefix p with paths p1 and p2 advertised from BR1 and BR4 to RR1. RR1 selects the bes tpath of the two and then advertises only p1 to the PE.
Figure 9-1 RR Hiding an Additional Path
In the above figure, prefix x with path x1 is advertised from BR2 to BR3 (which has path x2) with local preference 100. BR3 also has path x2, but due to the routing policy, BR3 advertises to the RRs x1 (not shown) instead of x2, x2 is suppressed. You could turn on the advertisement of best external on BR3 and advertise x2 to the route reflectors (RRs), but, the RRs advertise only the best path.
Suboptimal Hot-Potato Routing Scenario
To minimize internal transport costs, transit Internet service providers try to forward packets to the closest exit point (according to the Interior Gateway Protocol (IGP) cost). This behavior is known as hot-potato routing. In the distributed RR cluster model of the figure below, assume traffic that is coming from LA must go to Mexico. All links have the same IGP cost. If there are two exit points toward Mexico—one toward Austin and one toward Atlanta—the border router will try to send traffic to Austin based on the lower IGP cost from LA toward Austin than toward Atlanta. In a centralized RR model where the central RR resides where RR3 is (and RR1, RR2, RR4, and RR5 do not exist), the closest exit point toward Mexico, as seen from RR3, might be Atlanta. Sending the traffic from LA toward the Atlanta border router (BR) results in suboptimal hot-potato routing and is not desirable.
Figure 9-2 Distributed RR Cluster
Benefits of Additional BGP Paths
BGP routers and route reflectors (RR) propagate only their best path over their sessions. The advertisement of a prefix replaces the previous announcement of that prefix (also known as an implicit withdraw).
While this behavior might achieve better scaling, it can prevent path diversity, which tends to be poor or completely lost. This behavior prevents efficient use of the BGP multipath, prevents hitless planned maintenance, and lead to multi-exit discriminator (MED) oscillations and suboptimal hot-potato routing. It also inhibits fast and local recovery for next-hop failures, because the network has to wait for BGP control plane convergence to restore traffic.
The BGP Additional Paths feature is a BGP extension that allows the advertisement of multiple paths for the same prefix without the new paths implicitly replacing any previous paths. This behavior promotes path diversity and reduces MED oscillations.
BGP Additional Paths Functionality
You can use the BGP Additional Paths feature by adding a path identifier to each path in the Network Layer Reachability Information (NLRI). The path identifier (ID) can be considered as something similar to a route distinguisher (RD) in virtual private networks (VPNs), except that a path ID can apply to any address family. Path IDs are unique to a peering session and are generated for each network. The path identifier is used to prevent a route announcement from implicitly withdrawing the previous one. The Additional Paths feature allows the advertisement of more paths, in addition to the best path and allows the advertisement of multiple paths for the same prefix, without the new paths implicitly replacing any previous paths.
The BGP Additional Paths feature requires you to take general steps:
1. Specify whether the device can send, receive, or send and receive additional paths at the address family level or the neighbor level. During session establishment, two BGP neighbors negotiate the additional path capabilities (whether they can send or receive) between them.
2. Select a set or sets of candidate paths for advertisement by specifying the selection criteria.
3. Advertise for a neighbor a set or sets of additional paths from the candidate paths marked.
To send or receive additional paths, the additional path capability must be negotiated between the neighbors. If no negotiation occurs, even if the selection criteria marks the best path and the neighbor is configured to advertise the marked paths, the selections are useless because only the best path is advertised.
Configuring BGP to send or receive additional paths triggers negotiation of an additional path’s capability with the device's peers. Neighbors that have negotiated the capability are grouped together in an update group (if other update group policies allow), and in a separate update group from those peers that have not negotiated the capability. Therefore, the additional path capability causes the neighbor's update group membership to be recalculated.
Additional Path Selection
Only the best path is advertised to peers unless you configure the set path-selection all advertise command which advertises all BGP paths as additional paths to peers if the receive capability is enabled.
Advertising a Subset of the Paths Selected
Take care when you select a set of paths but want to advertise a different set of paths. If the set of paths you want to advertise is not a subset of the selected paths, you will not advertise the paths that you want advertised.
Configuring BGP Additional Paths
This section includes the following topics:
Configuring BGP Additional Paths for each Address Family
You can specify whether the device can send and receive additional paths to and from all neighbors within an address family.
BEFORE YOU BEGIN
Ensure that you have enabled the BGP feature.
SUMMARY STEPS
1. configure terminal
2. router bgp as-number
3. address family {ipv4 | ipv6} unicast
4. additional-paths receive
5. additional-paths send
6. additional-paths selection route-map map-name
7. end
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example : switch# configure terminal switch(config)# |
Enters global configuration mode. |
Step 2 |
router bgp as-number Example : switch(config)# router bgp 65000 switch(config-router)# |
Enables BGP and assigns the autonomous system number to the local BGP speaker. |
Step 3 |
address family {ipv4 | ipv6} unicast Example : switch(config-router)# address family ipv6 unicast |
Enters address family configuration mode. |
Step 4 |
additional-paths receive Example : switch(config-router-af)# additional-paths receive |
(Optional) Enables BGP additional paths for a prefix to be received from a capable peer.
Note This capability applies to all neighbors under the specified address family unless the capability is explicitly disabled with the neighbor additional-paths receive disable command, which overrides the configuration for the address family.
|
Step 5 |
additional-paths send Example : switch(config-router-af)# additional-paths send |
(Optional) Enables BGP additional paths for a prefix to be sent to a capable peer.
Note This capability applies to all neighbors under the specified address family unless the capability is explicitly disabled with the neighbor additional-paths send disable command, which overrides the configuration for the address family.
|
Step 6 |
additional-paths selection route-map map-name Example : switch(config-router-af) # additional-paths selection route-map rmap |
(Optional) Configures additional paths selection capability for a prefix. |
Step 7 |
end Example: switch(config-router-af)# end |
(Optional) Exits to privileged EXEC mode. |
Configuring BGP Additional Paths for each Neighbor
You can configure whether a particular neighbor can send or receive additional paths.
SUMMARY STEPS
1. configure terminal
2. router bgp as-number
3. neighbor { ipv4-address | ipv4-prefix/length | ipv6-address | ipv6-prefix/length } [ remote-as { as-num } [. as-num ]]
4. address family {ipv4 | ipv6} unicast
5. capability additional-paths receive [disable]
6. capability additional-paths send [disable]
7. end
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example : switch# configure terminal switch(config)# |
Enters global configuration mode. |
Step 2 |
router bgp as-number Example : switch(config)# router bgp 65000 switch(config-router)# |
Enables BGP and assigns the autonomous system number to the local BGP speaker. |
Step 3 |
neighbor {ipv4-address | ipv4-prefix/length | ipv6-address | ipv6-prefix/length} [ remote-as {as-num} [.as-num ]] Example : switch(config-router)# neighbor 2001:DB8::1037 |
Configures a BGP neighbor (router, VRF) and enters neighbor configuration mode. |
Step 4 |
address family {ipv4 | ipv6} unicast Example : switch(config-router)# address family ipv6 unicast |
Enters address family configuration mode. |
Step 5 |
capability additional-paths receive [disable] Example : switch(config-router-af)# capability additional-paths receive |
(Optional) Configures the receive additional paths capability for the specified neighbor.
Note This command overrides any send or receive capability that is configured at the address-family level.
|
Step 6 |
capability additional-paths send [disable] Example : switch(config-router-af)# capability additional-paths send |
(Optional) Configures the send additional paths capability for the specified neighbor.
Note This command overrides any send or receive capability that is configured at the address-family level.
|
Step 7 |
end Example: switch(config-router-af)# end |
(Optional) Exits to privileged EXEC mode. |
Configuring Additional Paths Using a Peer Policy Template
You can send and receive additional paths by using a peer policy template.
SUMMARY STEPS
1. configure terminal
2. router bgp as-number
3. template peer-policy template-name
4. capability additional-paths receive [disable]
5. capability additional-paths send [disable]
6. exit
7. neighbor { ipv4-address | ipv4-prefix/length | ipv6-address | ipv6-prefix/length } [ remote-as { as-num } [. as-num ]]
8. inherit peer-policy template-name sequence-number
9. end
DETAILED STEPS
|
|
|
Step 1 |
configure terminal Example : switch# configure terminal switch(config)# |
Enters global configuration mode. |
Step 2 |
router bgp as-number Example : switch(config)# router bgp 65000 |
Enters BGP mode and assigns the autonomous system number to the local BGP speaker. |
Step 3 |
template peer-policy template-name Example : switch(config-router)# template peer-policy rr-client-ptl # |
Enters policy-template configuration mode and creates a peer policy template. |
Step 4 |
capability additional-paths receive [disable] Example : switch(config-router-af)# capability additional-paths receive |
(Optional) Configures the receive additional paths capability for the specified neighbor.
Note This command overrides any send or receive capability that is configured at the address-family level.
|
Step 5 |
capability additional-paths send [disable] Example : switch(config-router-af)# capability additional-paths send |
(Optional) Configures the send additional paths capability for the specified neighbor.
Note This command overrides any send or receive capability that is configured at the address-family level.
|
Step 6 |
exit Example : switch(config-router-ptmp)# exit |
Exits policy-template configuration mode and returns to router configuration mode. |
Step 7 |
neighbor {ipv4-address | ipv4-prefix/length | ipv6-address | ipv6-prefix/length} [ remote-as {as-num} [.as-num]] Example : switch(config-router)# neighbor 2001:DB8::1037 |
Configures a BGP neighbor (router, VRF) and enters neighbor configuration mode. |
Step 8 |
address-family ipv4 unicast Example : switch(config-router-neighbor)# address-family ipv4 unicast switch(config-router-neighbor-af)# |
(Optional) Configures global address family configuration mode for the specified address family. |
Step 9 |
inherit peer-policy template-name sequence-number Example : switch(config-router-neighbor-af)# inherit peer-policy rr-client-ptl 10 |
Sends a peer policy template to a neighbor so that the neighbor can inherit the configuration. |
Step 10 |
end Example: switch(config-router-af)# end |
(Optional) Exits to privileged EXEC mode. |
Filtering and Setting Actions for Additional Paths
You can optionally use a route map to filter the paths to be advertised by matching on the prefix of additional paths that are candidates to be advertised. (These prefixes are configured with the additional-paths selection command.)
You can also optionally set one or more actions to take for those paths that pass through the route map. This procedure uses the set metric command. Other set commands are available that are not shown in this task.
You would set a metric for paths marked with all (all paths with a unique next-hop) if the neighbor is receiving the same routes from its neighbors. Suppose the neighbor 2001:DB8::1037 is receiving the same route from different neighbors. Routes received from the local device have a metric of 565 and routes from another device have a metric of 700. Routes with metric 565 have precedence over the routes with metric 700.
SUMMARY STEPS
1. configure terminal
2. route-map route-name [ deny | permit ] [ sequence-number ]
3. set path-selection all advertise
4. set metric metric-value
5. end
DETAILED STEPS
:
|
|
|
Step 1 |
configure terminal Example : switch# configure terminal switch(config)# |
Enters global configuration mode. |
Step 2 |
route-map map-name [ deny | permit ] [ sequence-number ] Example : switch(config)# route-map add_path4 permit 10 |
Defines a route map and the conditions for redistributing routes from one routing protocol into another. |
Step 3 |
set path-selection all advertise Example : switch(config-route-map)# set path-selection all advertise |
Advertises all BGP paths as additional paths to peers if the receive capability is enabled. |
Step 4 |
set metric metric-value Example : switch(config-route-map)# set metric 500 |
Sets the metric of the additional paths that pass the match criteria.
- Other set commands can be used to take action on the paths that pass the route map.
|
Step 5 |
end Example: switch(config-router-af)# end |
(Optional) Exits to privileged EXEC mode. |
Configuration Examples for BGP Additional Paths
This section includes the following topics:
BGP Additional Paths Send and Receive Capabilities
R1
In this example, R1's address is 2001:db8::1045; its neighbor R2 has an address of 2001:db8::1037. Updates are sent from R2 to R1 with additional-paths (all paths advertised). Updates are sent from R1 to R2 with only the classic BGP best path advertised because R2 can only send additional paths, not receive additional paths.
route-map add_path4 permit 10
set metric 500
set path-selection all advertise
!!
router bgp 1
address-family ipv6 unicast
additional-paths send
additional-paths receive
additional-paths selection route-map add_path4
neighbor 2001:db8::1037
address-family ipv6 unicast
capability additional-paths send
capability additional-paths receive
R2
route-map add_path4 permit 10
set metric 500
set path-selection all advertise
!!
router bgp 2
address-family ipv6 unicast
additional-paths selection route-map add_path4
neighbor 2001:db8::1045
address-family ipv6 unicast
capability additional-paths send
BGP Additional Paths Using a Peer Policy Template
This example shows that the neighbor with IP address 2001:db8::1037 has the send and receive capability for additional paths enabled through the template named rr-client-pt1:
router bgp 65000
address-family ipv6 unicast
additional-paths send
additional-paths receive
additional-paths selection route-map add_path4
neighbor 2001:db8::1037
address-family ipv6 unicast
inherit peer-policy rr-client-pt1 10
template peer-policy rr-client-pt1
capability additional-paths send
capability additional-paths receive