bgp rib failure problem or not?

From: Alex Steer (alex.steer@eison.co.uk)
Date: Sun Aug 26 2007 - 11:44:18 ART


Hello everyone,

Just got a quick question with regard to BGP RIB failures on internally
known routes installed into BGP via the network command. I've noticed
quite a few scenarios where this seems to happen and not sure if I'm
missing a trick. I was hoping somebody would like to enlighten me
slightly as to if it causes some sort of problem I'm not noticing.

Here is the topology/masterpiece of modern art:-

Network1==R1======R2======R3

AS numbers

R1, R2 = 100

R3 = 300

Network1 is known to R1 and R2 via an IGP (say ospf) and via the network
command on R1.

R1s routing table Network1 is known via connected

R2s routing table Network1 is known via ospf

R3s routing table Network1 is known via eBGP (ad20) next hop R2

Network1 is participating in the OSPF process

Network1 is placed in the BGP table on R1 via "network" command

Great so it all works fine. So my question is, why does BGP on R2
advertise the route to R3 when it doesn't like the route itself? Is
there a problem that the RIB failure is causing and I'm not seeing? It
looks to be working fine, is there another purpose to informing me of
the r?

R2#show ip bgp

Network Next Hop Metric LocPrf Weight
Path

r>i Network1/24 Router1 0 100
0 i

R3#show ip bgp

Network Next Hop Metric LocPrf Weight
Path

*>i Network1/24 Router2 0 100
0 100 i

By the way I made all of the output up as my lab is off so forgive any
slight mistakes. It's only to aid in posing my question.

Many thanks for any info on this.

Kind regards

Alex



This archive was generated by hypermail 2.1.4 : Sat Sep 01 2007 - 11:32:13 ART