BGP as-override vs allow-as-in

These two functions are pretty similar, just with subtle differences. They can be used in an environment where a customer is using one AS number for many sites that are connected to an ISP. This is shown in the example below.

You can see that AS 65001 connects to the ISP at two locations. So when R2 receives the prefix 99.99.99.99/32, he will see that the AS path is via 1, 65001. Because of the loop prevention mechanism, R2 will have to reject this prefix because it can see its own AS in the AS_PATH attribute. I’ll demonstrate that now. But first, I’m going to post the full configurations because this can be pretty confusing to configure.

PE1#
ip vrf google
 rd 100:1
 route-target export 100:1
 route-target import 100:1
!
interface Loopback0
 ip address 1.1.1.1 255.255.255.255
!
interface FastEthernet0/0
 ip vrf forwarding google
 ip address 10.10.10.10 255.255.255.0
!
interface FastEthernet0/1
 ip address 15.15.15.1 255.255.255.0
 mpls ip
!
router ospf 1
 network 1.1.1.1 0.0.0.0 area 0
 network 15.15.15.0 0.0.0.255 area 0
!
router bgp 1
 no synchronization
 bgp log-neighbor-changes
 neighbor 2.2.2.2 remote-as 1
 neighbor 2.2.2.2 update-source Loopback0
 neighbor 2.2.2.2 next-hop-self
 no auto-summary
 !
 address-family vpnv4
  neighbor 2.2.2.2 activate
  neighbor 2.2.2.2 send-community extended
 exit-address-family
 !
 address-family ipv4 vrf google
  neighbor 10.10.10.11 remote-as 65001
  neighbor 10.10.10.11 activate
  no synchronization
 exit-address-family

PE2#
ip vrf google
 rd 100:1
 route-target export 100:1
 route-target import 100:1
!
interface Loopback0
 ip address 2.2.2.2 255.255.255.255
!
interface FastEthernet0/0
 ip vrf forwarding google
 ip address 20.20.20.20 255.255.255.0
!
interface FastEthernet0/1
 ip address 25.25.25.2 255.255.255.0
 mpls ip
!
router ospf 1
 network 2.2.2.2 0.0.0.0 area 0
 network 25.25.25.0 0.0.0.255 area 0
!
router bgp 1
 no synchronization
 bgp log-neighbor-changes
 neighbor 1.1.1.1 remote-as 1
 neighbor 1.1.1.1 update-source Loopback0
 neighbor 1.1.1.1 next-hop-self
 no auto-summary
 !
 address-family vpnv4
  neighbor 1.1.1.1 activate
  neighbor 1.1.1.1 send-community extended
 exit-address-family
 !
 address-family ipv4 vrf google
  neighbor 20.20.20.21 remote-as 65001
  neighbor 20.20.20.21 activate
  no synchronization
 exit-address-family

R1#
interface Loopback0
 ip address 99.99.99.99 255.255.255.255
!
interface FastEthernet0/0
 ip address 10.10.10.11 255.255.255.0
!
router bgp 65001
 no synchronization
 bgp log-neighbor-changes
 network 10.10.10.0 mask 255.255.255.0
 network 99.99.99.99 mask 255.255.255.255
 neighbor 10.10.10.10 remote-as 1
 no auto-summary

R2#
interface FastEthernet0/0
 ip address 20.20.20.21 255.255.255.0
!
router bgp 65001
 no synchronization
 bgp log-neighbor-changes
 network 20.20.20.0 mask 255.255.255.0
 neighbor 20.20.20.20 remote-as 1
 no auto-summary

P1#
interface FastEthernet0/0
 ip address 15.15.15.2 255.255.255.0
 mpls ip
!
interface FastEthernet0/1
 ip address 25.25.25.1 255.255.255.0
 mpls ip
!
router ospf 1
 network 15.15.15.0 0.0.0.255 area 0
 network 25.25.25.0 0.0.0.255 area 0
 network 35.35.35.0 0.0.0.255 area 0

So currently, on R2, he is not accepting any prefixes from R1 in the other site; as shown below (we would expect the 10.10.10.0/24 and 99.99.99.99/32 networks to be in the BGP table).

R2#sh ip bgp | b Network
   Network          Next Hop            Metric LocPrf Weight Path
*> 20.20.20.0/24    0.0.0.0                  0         32768 i

The debug below shows why he’s not accepting these prefixes.

*Mar  1 02:28:03.515: %BGP-5-ADJCHANGE: neighbor 20.20.20.20 Up
*Mar  1 02:28:03.559: BGP(0): 20.20.20.20 rcv UPDATE w/ attr: nexthop 20.20.20.20, origin i, originator 0.0.0.0, path 1 65001, community , extended community
*Mar  1 02:28:03.559: BGP(0): 20.20.20.20 rcv UPDATE about 10.10.10.0/24 -- DENIED due to: AS-PATH contains our own AS;
*Mar  1 02:28:03.559: BGP(0): 20.20.20.20 rcv UPDATE about 99.99.99.99/32 -- DENIED due to: AS-PATH contains our own AS;

One way to fix this is to use the allow-as-in command. This allows R2 to override the loop prevention mechanism by allowing an instance of AS 65001 to be in the AS_PATH. Let’s do that now.

R2(config-router)#neighbor 20.20.20.20 allowas-in 1
R2(config-router)#
*Mar  1 02:34:34.927: BGP: 20.20.20.20 sending REFRESH_REQ(5) for afi/safi: 1/1
*Mar  1 02:34:34.927: BGP: 20.20.20.20 send message type 5, length (incl. header) 23
*Mar  1 02:34:35.015: BGP(0): 20.20.20.20 rcvd UPDATE w/ attr: nexthop 20.20.20.20, origin i, path 1 65001
*Mar  1 02:34:35.015: BGP(0): 20.20.20.20 rcvd 10.10.10.0/24
*Mar  1 02:34:35.015: BGP(0): 20.20.20.20 rcvd 99.99.99.99/32
*Mar  1 02:34:35.019: BGP(0): Revise route installing 1 of 1 routes for 10.10.10.0/24 -> 20.20.20.20(main) to main IP table
*Mar  1 02:34:35.019: BGP(0): Revise route installing 1 of 1 routes for 99.99.99.99/32 -> 20.20.20.20(main) to main IP tableR2(config-router)#do sh ip bgp
BGP table version is 4, local router ID is 20.20.20.21
Status codes: s suppressed, d damped, h history, * valid, > best, i - internal,
              r RIB-failure, S Stale
Origin codes: i - IGP, e - EGP, ? - incomplete

   Network          Next Hop            Metric LocPrf Weight Path
*> 10.10.10.0/24    20.20.20.20                            0 1 65001 i
*> 20.20.20.0/24    0.0.0.0                  0         32768 i
*> 99.99.99.99/32   20.20.20.20                            0 1 65001 i

So you can see (on R2) that the AS_PATH is 1, 65001 for these prefixes. It keeps all the AS_PATH information and simply just allows 1 occurance of 65001 to be in the AS_PATH; thus overriding the loop prevention mechanism. We would obviously need to do this on R1 in order for R1 to have reachability to the 20.20.20.0/24 prefix (sitting between PE2 and R2) so that he can have a route back to R2.

R1(config)#router bgp 65001
R1(config-router)#neighbor 10.10.10.10 allowas-in 1

R2#ping 99.99.99.99

Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 99.99.99.99, timeout is 2 seconds:
!!!!!
Success rate is 100 percent (5/5), round-trip min/avg/max = 68/83/116 ms

The other way you can complete this task is by getting PE1 & PE2 to just strip AS 65001 from the BGP UPDATE before sending it to the customer edge routers. Let’s do that now.

R1(config-router)#no neighbor 10.10.10.10 allowas-in 1
R2(config-router)#no neighbor 20.20.20.20 allowas-in 1

pe2(config)#router bgp 1
pe2(config-router)#address-family ipv4 unicast vrf google
pe2(config-router-af)#neighbor 20.20.20.21 as-overridepe1(config)#router bgp 1
pe1(config-router)#address-family ipv4 unicast vrf google
pe1(config-router-af)#neighbor 10.10.10.11 as-override

By configuring this command it actually resets the peer, so there’s no need to clear any neighbors. The result of this is shown on R2 below.

R2#sh ip bgp | b Network

   Network          Next Hop            Metric LocPrf Weight Path
*> 10.10.10.0/24    20.20.20.20                            0 1 1 i
*> 20.20.20.0/24    0.0.0.0                  0         32768 i
*> 99.99.99.99/32   20.20.20.20                            0 1 1 i

So the AS_PATH has been overridden by the PE routers to their AS number instead. This is the key difference between the two commands. Allow-as-in allowed the loop prevention to be ignored for the configured amount of instances, and the as-override caused the PE routers to modify the AS_PATH.