cancel
Showing results for 
Search instead for 
Did you mean: 
cancel
333
Views
5
Helpful
3
Replies

BGP path selection

Following is the scenario:

There are 4 routers connected in square formation, D1, D2, C1, C2

D1 is physically connected to C1 and D2

D2 is physically connected to C2 and D1

C1 is physically connected to C2

C1,C2 are in AS-1

D1-D2 are in AS-2

C1-D1 has ebgp peering

C2-D2 has ebgp peering

C1-C2 has IBGP peering

D1-D2 has IBGP peering

Our objective is; all the incoming traffic for prefix 10.0.0.0/8 should come towards D1 under normal conditions:

This ia how I am trying to achieve it;

D1 advertises 10.0.0.0/8 using aggregate address command to C1

D2 advertised 10.0.0.0/8 {2,2} using aggregate address command and a route map for as-path prepend to C2

C1's bgp table:

10.0.0.0/8 pointing -->D1

10.0.0.0/8 {2,2}pointing --> C2

C1's routing table:

10.0.0.0/8 --> D1 ( this is as desired OK)

C2's bgp table:

10.0.0.0/8 -->C1

10.0.0.0/8 {2,2} --> D2

C2's routing table:

10.0.0.0/8 {2,2} --D2 ( this is problematic for me, I want C2 to put the other route that it has learned from C1 in its routing table)

According to CISCO's BGP best path selection, Shortest AS_PATH is preferred first as it is rule # 4. EBGP over IBGP is rule# 7

http://www.cisco.com/en/US/tech/tk365/tk80/technologies_tech_note09186a0080094431.shtml

What's happening at C2 is that C2 is preferring EBGP over IBGP (rule7 and not taking into consideration the as-path length rule 4. There is no specific configuration in bgp of C2 for ignoring AS-Path while making selection. What could be the possible reason or a solution to this problem

I am running 12.1E on 6500s

1 Accepted Solution

Accepted Solutions

The reason is most probably that the path on C2 that you receive from C1 is either "not synchronized", in which case you might want to turn off the BGP synchronization (no synch under the BGP process) or that its BGP nexthop is not reachable, in which case you can use the "nei next-hop-self" command on C1 to insure that the BGP next hop will be reachable.

Hope this helps,

Harold Ritter
Sr Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México
Paseo de la Reforma 222
Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México

View solution in original post

3 Replies 3

Harold Ritter
Cisco Employee
Cisco Employee

If the AS path is indeed prepended, this should work as expected. Could you include a "sh ip bgp 10.0.0.0 255.0.0.0" from C1 and C2.

Thanks,

Harold Ritter
Sr Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México
Paseo de la Reforma 222
Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México

The reason is most probably that the path on C2 that you receive from C1 is either "not synchronized", in which case you might want to turn off the BGP synchronization (no synch under the BGP process) or that its BGP nexthop is not reachable, in which case you can use the "nei next-hop-self" command on C1 to insure that the BGP next hop will be reachable.

Hope this helps,

Harold Ritter
Sr Technical Leader
CCIE 4168 (R&S, SP)
harold@cisco.com
México móvil: +52 1 55 8312 4915
Cisco México
Paseo de la Reforma 222
Piso 19
Cuauhtémoc, Juárez
Ciudad de México, 06600
México

The problem is fixed by configuring no-sync.Thanks IQ

Getting Started

Find answers to your questions by entering keywords or phrases in the Search bar above. New here? Use these resources to familiarize yourself with the community: