- BGP is designed to perform next-hop AS packet forwarding.
- BGP path selection is based upon policy-routing.
- A BGP speaker must never advertise a prefix it does not know to get to.
The advertising EBGP speaker sets the next hop attribute and prepends the as path list.
By default the advertising IBGP speaker does not set the next hop attribute or prepend the as path list. You can make the IBGP speaker set the next hop attribute by configuring the ” neightbor x.x.x.x next-hop-self” command.
Some useful commands:
- debug bgp all events
- debug bgp ipv4 unicast
- debug bgp ipv4 unicast keepalives
- debug ip error
- debug ip bgp updates
EBGP – not directly connected:
- nei x.x.x.x ebgp-multihop <hop>
or
- nei x.x.x.x ttl-security hop <hop>
- do I need the nei x.x.x.x update-source Y{z} command?
neigh x.x.x.x ttl-security hops 3 < — external neighbor maybe up to 3 hops away. (255-3 = max 252 hops away.
on the neighbor: nei x.x.x.x ebgp-multihop 255 – don’t forget ttl decrements. Use 255 for neighors of ttl sec
To originate a prefix in BGP:
network x.x.x.x mask y.y.y.y
By default, advertising EBGP speaking routers set the BGP next-hop attribute.
By default, advertising IBGP speaking routers DO NOT set the BGP next-hop attribute.
Route reflector:
RR servers propagate routes inside the AS based on the following rules:
- If a route is received from nonclient peer, reflect to clients only.
- If a route is received from a client peer, reflect to all nonclient peers and also to client peers, except the originator of the route.
- If a route is received from an EBGP peer, reflect to all client and nonclient peers.
Setting the cluster-id with the IOS.
- use the BGP router-id x.x.x.x command.
- use the BGP cluster-id x.x.x.x command.
* when both commands are used the BGP cluster-id x.x.x.x takes precedence.