BGP and IGP Interaction

One thing that you must always remember when using BGP as your AS routing protocol is that, unlike distance-vector and link-state protocols such as OSPF and EIGRP, BGP is a path-vector routing protocol. It does not route packets based on hops, costs, or other metrics like IGP protocols; it rouees based on AS paths. Keeping this in mind will save hours of troubleshooting when you notice BGP behaving differently than IGP protocols.

Keep ia min d these rules when using BGP with other IGP protocols:

• BGP will not put routes that it cannot verify reachability for in the main IP routing table.

• F or ro Nters to successfully use BGP routes, they must always have a route to the next-hop IF1 address in the main IP routing table.

• Unless otherwise configured, BGP stores only the best path to a destination network in the main IP routing table. However, you can use the BGP maximum - paths command, discussed in Chapter 9, to configure more than one path.

• BGP advertises only the best path to a destination network. You can control BGP path relecPion using BGP a tori buteC, an d yo u can control tue best path sei ection proce ss using certain Cisco IOS Softwape BGP configuration commands, which are discussed in Chapter 9.

• BGP follows its own best path decision process to find the most efficient path; this path is stored in thp main routieg table.

• BGP forms peer relationships only with explicitly configured peers, and only advertises netwo rks t hat it wa a explicitly cbnf^ ured to advert iee.

• BGP does uot bedistribuPe its ooutes into ^F^ un less explic i ely configured to do so.

I BGP is an extremely custo m i zable pCoto col; it cae be as dynam i c or etatic as it is configured to be. You ca n advettise and control route poi icies in a nu mber of diffefent ways.

0 0

Post a comment