[j-nsp] bgp to ospf

[Richard Zheng]

On Thu, Jun 16, 2011 at 7:55 AM, Jeff Wheeler <jsw at inconcepts.biz> wrote:

> On Thu, Jun 16, 2011 at 12:48 PM, Payam Chychi <pchychi at gmail.com> wrote:
> > Were you able to figure this out?
>
> As I mentioned, take my post with a grain of salt.  I may be incorrect here
> about the actual functioning, as I have never, ever had reason to utilize
> this in a practical network.  But if you do some reading, the intended
> purpose of Fwd Adr is expressed above.  The original poster should not be
> using it for what he wants to do (I don't think it will work), and should
> instead utilize BGP or change his topology.
>
>
I came up with the same conclusion. I figured that OSPF can only set forward
address under very strict conditions. Using another router's IP is not
supported by most implementation, not sure supported by RFC though.

My summary is posted here,
http://mailman.nanog.org/pipermail/nanog/2011-June/037479.html.

I did come across another interesting issue when implementing IBGP between
RTR A and RTR B. RTR A sent the customer route to border router B, then B
redistributed from IBGP to OSPF. In the whole AS, the route was bounced on
and off from routing table every few seconds.

JUNOS uses different administrative distance from Cisco. In JUNOS, OSPF is
150 and BGP is 170. When the route is learned from EBGP, it is 170 and sent
to IBGP, then redistribute to OSPF. Then the route is propagated through
OSPF to RTR A, the same route from OSPF has a distance of 150, it overrides
the EBGP route and then withdraws via IBGP to RTRB. RTR B stops
redistribution from IBGP to OSPF. Once it disappears from OSPF, the EBGP
route comes back. It goes through the same cycle again.

The solution I can think about is to change administrative distance of BGP
to a lower number, say 120. So the OSPF route won't take over in the routing
table. Should be ok, but not quite sure.

Richard