BGP - Route Propagation iBGP eBGP

mattsthe2

I'm need your help guys.

I'm having trouble getting my head around what happens and what doesn't happen.

I get that eBGP peers share and pass back and forth prefix's.
I'm having trouble getting my head around when iBGP is thrown into the mix. Whats the golden rules?

If I've not clearly explained that then let me know

kpjungle

mattsthe2 wrote: »

I'm need your help guys.

I'm having trouble getting my head around what happens and what doesn't happen.

I get that eBGP peers share and pass back and forth prefix's.
I'm having trouble getting my head around when iBGP is thrown into the mix. Whats the golden rules?

If I've not clearly explained that then let me know

iBGP is simply BGP "knowing" that its peering with another BGP within its own AS. iBGP peers also pass routes back and forth. There are some subtle differences, such as what the next hop advertised to an iBGP is (at least default), compared to sending to an eBGP peer. Also remember that the administrative distance for iBGP routes is 200 compared to the 20 for eBGP routes.

There is a synchronization rule stating that a BGP router will not advertise an iBGP-learned route unless that route was learned from an IGP, to prevent black holes. This is afaik off by default in newer versions.
http://www.cisco.com/en/US/docs/internetworking/case/studies/icsbgp4.html#wp19382

A full mesh is normally required for iBGP neighbors, however, an advanced feature such as a route-reflector can be used.

BGP neighbors within same AS = iBGP.
BGP Neighbors with different AS = eBGP.

Not sure if that answers your question?