BGP - CCIE Notes

1. Within an AS, bgp peers do not need to be directly connected.
2. For routers that run ebgp, neighbors are usually directly connected.
3. ALL bgp speakers within an AS MUST establish a peer relationship unless you use Route
4. reflectors or confederations.
5. When a bgp speaker receives an update from other bgp speakers in its own AS, (via ibgp) the
6. receiving bgp speaker uses ebgp to forward the update to ebgp speakers only.
7. The BGP synchronization rule states that if an AS provides transit service to another AS, BGP
8. should not advertise a route until all of the routers within the AS have learned the route via an
9. IGP.
10. You can disable synchronization if one of the following is true:
• 1. Your AS does not pass traffic from one AS to another.
• 2. ALL the transit routers in your AS run BGP
11. The only difference between advertising a static and a default route, is that when you redistribute a
12. static, BGP sets the origin attribute of updates to incomplete.
13. Redistributing a static route is the best way to advertise a supernet because it stops the route from
14. flapping.
15. To ensure a loop free inter-domain topology, BGP does not accept updates that originated from its own AS.
16. Origin attribute- will be “i” when injected with network command in router configuration mode, “e” when learned through EGP, “?” incomplete when a route is redistributed into bgp.
17. BGP specifies that the next hop of EBGP learned routes remain unchanged into and through IBGP.

BGP Attributes

The weight attribute is a special CISCO attribute that is used in the path selection when there is more than one route to the destination. The weight attribute is local to the router on which it is assigned and is NOT propagated in routing updates. (higher more preferred), there are 3 ways to set weight:

• Access-list
• Route-map
• Neighbor weight command

The local preference attribute indicates the preferred path when there is multiple paths. (higher=better). Unlike the weight attribute, the local preference is carried with route updates and exchanged with routers in the same AS. 2 ways to set local preference:

• use the bgp default local-preference command
• route-maps

The MED attribute is a hint to EBGP peers about the preferred path into an AS when there are multiple. (lower=better). Unlike local preference, the MED is exchanged between AS’s, but a MED that comes into

an AS does not leave the AS.

The community attribute provides a way of grouping destinations to which routing decisions can be

applied. To send the attribute you MUST use the neighbor send-community router config command.

Other topics:

BGP Route Reflectors- eliminates full mesh requirement.

BGP Confederations- makes “mini- AS’s” inside of an AS.

BGP Peer groups – a group of neighbors that share the same update policies.

Brief, BGP Path Selection Process:

1. Is the Next-Hop present for the Route.
2. Prefer Largest Weight, if Cisco.
3. Prefer largest local preference, after Weight.
4. Internally Generated Routes have high preference.
5. Prefer Shortest AS Path
6. Prefer Incomplete Origin over IGP and IGP over EGP.
7. Prefer the lowest MED ( Metric )
8. Prefer closest route learned through an IGP
9. If still all stuff is same, make decisions on BGP Router IDs, lowest is always preferred.

Important BGP Commands:

Aggregate-address	Configure bgp aggregate entries
Auto-summary
Default-metric	Set metric of redistributed routes
Distance	Define admin distance
Distribute-list	Filter Networks in routing updates
Maximum-paths	Forward Packets on multiple paths
Synchronization	Perform IGP Synchronization (IBGP)
Timers	Adjust BGP Update Timers
Traffic-share	Algorithm for computing traffic share over alternate routes
Neighbor advertise-map	Specific route-map for conditional adverstisements
Neighbor advertisement-interval	Min. Interval b/w EBGP Routing Updates
Neighbor distribute-list	Filter Routes specific to neighbor
Neighbor Ebgp-multihop	Allow EBGP Neighbors not on directly connected networks
Neighbor Filter-list	Enable BGP Filters
Neighbor maximum-prefix	Max. Limit of Routes the neighbor could learn.
Neighbor next-hop-self	Disable Next-Hop Calculation for neighbor and advertise itself as the neighbor
Neighbor peer-group	Assign a Peer Group to the neighbor
Neighbor prefix-list	Filter updates from this neighbor
Neighbor remote-as	Define the AS
Neighbor remove-private-as	No Private AS #s in outbound updates
Neighbor route-map	Apply a route-map
Neighbor route-reflector-client	Enable Route-Reflection on this Router
Neighbor send-community	Send the Community Attribute to this neighbor
Neighbor shutdown	Administratively disable peering with the neighbor
Neighbor timers	BGP Neighbor specific timers
Neighbor unsuppress-map	Route-Map to selectively allow suppressed routes to that specific neighbor
Neighbor update-source	Define Source interface for the neighbor
Neighbor weight	Specify neighbor specific weight (Cisco Only)