VPN and MPLS book error? question

From: Roberts, Larry (Larry.Roberts@xxxxxxxxxxxx)
Date: Thu May 23 2002 - 15:47:25 GMT-3


   
Im reading the Cisco Press MPLS and VPN Architectures book by Pepelnjak and
Guichard.

I'm not going to try and describe the router layout as it would take a 30
page document to get it straight. I'm just going to pose my question to the
group so that those that have the book will hopefully be kind enough to
review and explain.

I say that the traffic would make it through the MPLS network because the
London ASBR would send the packet that is destined for the 146.22.15/24
network to the next hop address of the Santa Clara ASBR, which is reachable
since they have an IBGP mesh between them (Them being the London ASBR and
the SC ASBR)
The book disagrees and say that it wouldn't make it because it would try and
send it directly to the 146.22.14/24 network.

I am assuming ( since on pg 363 its less then clear ) that the SC ASBR has
set the NEXT-HOP-SELF command for the networks that it advertises.
Is this the bad assumption, and if so, what did they mean in paragraph 3 (
or 4 if you count the note ) by the statement that "each of the BGP next-hop
addresses for the ISP-customer external routers will be present in the VRF
for the specific ISP ..."
Are they saying that the Customer external routes are advertised into the
iBGP session, and then a double lookup is being performed so that the next
hop can be determined ?
I don't think this is the case since the diagram clearly shows the NH being
set to the SC ASBR for the 146.22.15/24 update

HELP!

Thanks

Larry



This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:59:06 GMT-3