From: Sameh El Tawil (eltawil@free.fr)
Date: Mon Oct 11 2004 - 17:28:11 GMT-3
Hi Group,
I have seen in one of the examples in the Karl Solie book volume-2 that he was
using the route-reflector-client command for all neighbors in an AS with three
fully-meshed routers. The result is to have every route appearing twice in the
BGP table; one time from the direct session with the originating neighbor, and
the second time reflected from the other router.
When I configured this example in the lab, I was surprised to see that the
route coming from the originator directly is always preferred over the
reflected one. I am wondering which step in the best path selection algorithm
tells the router to make this choice. Unless I am missing something, there is
no rule to choose the route learned from the route's next hop over the same
information coming from the route reflector.
Any ideas?
Here is some output to illustrate:
- 10.2.2.0 is originated by the neighbor (by the network command on the
neighbor)
- 62.121.0.0 is originated by neighbor (learned from an EBGP session on
that neighbor)
2612#sh ip bgp 10.2.2.0
BGP routing table entry for 10.2.2.0/24, version 27
Paths: (2 available, best #1, table Default-IP-Routing-Table)
Advertised to non peer-group peers:
108.201.14.1
Local
192.168.1.2 (metric 74) from 192.168.1.2 (192.168.253.4)
Origin IGP, metric 0, localpref 100, valid, internal, best
Local
192.168.1.2 (metric 74) from 10.1.1.2 (192.168.254.1)
Origin IGP, metric 0, localpref 100, valid, internal
Originator: 192.168.253.4, Cluster list: 192.168.254.1
2612#sh ip bgp 62.121.0.0
BGP routing table entry for 62.121.0.0/16, version 25
Paths: (2 available, best #2, table Default-IP-Routing-Table)
Advertised to non peer-group peers:
108.201.14.1
203
10.1.1.2 from 192.168.1.2 (192.168.253.4)
Origin IGP, metric 0, localpref 100, valid, internal
Originator: 192.168.254.1, Cluster list: 10.2.5.1
203
10.1.1.2 from 10.1.1.2 (192.168.254.1)
Origin IGP, metric 0, localpref 100, valid, internal, best
2612#
Cheers,
Sameh
This archive was generated by hypermail 2.1.4 : Sat Nov 06 2004 - 17:11:46 GMT-3