Route Reflectors
See why an iBGP route stops at an intermediate peer, then enable reflection on R2.
From a full mesh to a reflector
A full mesh of six BGP speakers needs 15 sessions: N(N−1)/2. A single reflector with five clients needs five client sessions, but becomes a control-plane dependency. Redundant reflectors reduce that failure risk. A reflector advertises selected paths; it does not automatically put all learned alternatives into every client’s table.
Three-router experiment
R1–R2 and R2–R3 are ordinary iBGP sessions in AS 65123. OSPF supplies loopback and transit reachability. R1 originates the host route. R3 can resolve its next hop, but initially R2 cannot advertise an iBGP-learned route to another ordinary iBGP neighbor.
| Link | First endpoint | Second endpoint |
|---|---|---|
| R1–R2 | GigabitEthernet1 192.0.2.1/30 | GigabitEthernet1 192.0.2.2/30 |
| R2–R3 | GigabitEthernet2 192.0.2.5/30 | GigabitEthernet1 192.0.2.6/30 |
Complete baseline configurations
R1
hostname R1
ip cef
interface Loopback0
ip address 10.255.0.1 255.255.255.255
exit
interface Loopback10
ip address 198.51.100.1 255.255.255.255
exit
interface GigabitEthernet1
ip address 192.0.2.1 255.255.255.252
no shutdown
exit
router ospf 1
router-id 10.255.0.1
network 10.255.0.1 0.0.0.0 area 0
passive-interface Loopback0
network 192.0.2.1 0.0.0.0 area 0
exit
router bgp 65123
bgp router-id 10.255.0.1
no bgp default ipv4-unicast
neighbor 192.0.2.2 remote-as 65123
address-family ipv4 unicast
no auto-summary
neighbor 192.0.2.2 activate
network 198.51.100.1 mask 255.255.255.255
exit-address-family
endR2
hostname R2
ip cef
interface Loopback0
ip address 10.255.0.2 255.255.255.255
exit
interface GigabitEthernet1
ip address 192.0.2.2 255.255.255.252
no shutdown
exit
interface GigabitEthernet2
ip address 192.0.2.5 255.255.255.252
no shutdown
exit
router ospf 1
router-id 10.255.0.2
network 10.255.0.2 0.0.0.0 area 0
passive-interface Loopback0
network 192.0.2.2 0.0.0.0 area 0
network 192.0.2.5 0.0.0.0 area 0
exit
router bgp 65123
bgp router-id 10.255.0.2
no bgp default ipv4-unicast
neighbor 192.0.2.1 remote-as 65123
neighbor 192.0.2.6 remote-as 65123
address-family ipv4 unicast
no auto-summary
neighbor 192.0.2.1 activate
neighbor 192.0.2.6 activate
exit-address-family
endR3
hostname R3
ip cef
interface Loopback0
ip address 10.255.0.3 255.255.255.255
exit
interface GigabitEthernet1
ip address 192.0.2.6 255.255.255.252
no shutdown
exit
router ospf 1
router-id 10.255.0.3
network 10.255.0.3 0.0.0.0 area 0
passive-interface Loopback0
network 192.0.2.6 0.0.0.0 area 0
exit
router bgp 65123
bgp router-id 10.255.0.3
no bgp default ipv4-unicast
neighbor 192.0.2.5 remote-as 65123
address-family ipv4 unicast
no auto-summary
neighbor 192.0.2.5 activate
exit-address-family
endR3# show ip bgp 198.51.100.1/32
% Network not in tableSeparate clients and non-clients
| Route learned from | Eligible reflection destinations |
|---|---|
| eBGP | Clients and non-clients, subject to policy. |
| iBGP client | Other clients and non-clients. |
| iBGP non-client | Clients; not another non-client. |
The table summarizes iBGP reflection. Ordinary eBGP export rules still apply. Do not reflect a route back to its source as a useful new path; ORIGINATOR_ID and CLUSTER_LIST provide additional loop protection. Reflection normally preserves NEXT_HOP, so IGP reachability remains necessary.
Make R2 the reflector
router bgp 65123
address-family ipv4 unicast
neighbor 192.0.2.1 route-reflector-client
neighbor 192.0.2.6 route-reflector-client
exit-address-familyApply these commands on R2 only. R1 and R3 retain ordinary iBGP configuration. R2’s clients are identified by its directly connected neighbor addresses in this lab.
Inspect the reflected route
R3# show ip bgp 198.51.100.1/32
BGP routing table entry for 198.51.100.1/32
Paths: (1 available, best #1, table default)
Local
192.0.2.1 from 192.0.2.5 (10.255.0.2)
Origin IGP, metric 0, localpref 100, valid, internal, best
Originator: 10.255.0.1, Cluster list: 10.255.0.2The advertising peer is R2, while the preserved next hop and originator point back toward R1. R3’s OSPF route to the R1–R2 link makes that next hop reachable. The cluster list records reflection history; a reflector rejects a route containing its own cluster ID.
Restore ordinary iBGP
router bgp 65123
address-family ipv4 unicast
no neighbor 192.0.2.1 route-reflector-client
no neighbor 192.0.2.6 route-reflector-client
exit-address-familyR3# show ip bgp 198.51.100.1/32
% Network not in tableThe route disappears after the withdrawal, even though the two sessions remain up. A production design needs reflection or the appropriate full mesh, not this intentionally incomplete chain.
Key Takeaways
- Reflection removes the need for clients to form a full mesh.
- Client and non-client advertisement rules differ.
- ORIGINATOR_ID and CLUSTER_LIST prevent reflection loops.
- NEXT_HOP is normally preserved.
- Path visibility and reflector redundancy need deliberate design.
Protocol and command references
Route reflection · BGP specification · Router-symbol attribution.