6 messages in net.nether.puck.cisco-nsp[c-nsp] iBGP - Multihoming ideas
FromSent OnAttachments
John GitauJan 26, 2005 3:30 am 
Bruce PinskyJan 26, 2005 3:32 pm 
John GitauJan 27, 2005 12:53 am 
Mark TinkaJan 27, 2005 5:08 am 
John GitauJan 28, 2005 12:58 am 
Mark TinkaJan 28, 2005 2:48 am 
Actions with this message:
Paste this link in email or IM:
Paste this link in email or IM:
Atom feed for this thread
Paste this URL into your reader:
Subject:[c-nsp] iBGP - Multihoming ideasActions...
From:John Gitau (git@wananchi.com)
Date:Jan 27, 2005 12:53:06 am
List:net.nether.puck.cisco-nsp

-> Depends on what your goals are. What do you want to achieve -> using BGP on these routers? Do you want the Branch A router -> to learn all the routes from Branch B and HQ and do path -> selection on its own? Should Branch B simply default route -> to Branch B and HQ and let those routers make the outbound -> transit decision? Do you want to prefer one exit path vs another? -> ...etc...etc...

I want router A to learn everything from both routers and do the path selection. The main concern right now is routes from the Exchange point are sent to Branch A from both branch B and the HQ. Both HQ and Branch B need to send a default route to Branch A because some routes to the larger internet are not on their routing tables.

So far the best way seems to be establishing two route reflectors at Branch B and the HQ this is especially going to be important if I ever need to connect other BGP speaking routers in the future.

**Gitau