From: Joe Jia (ellenjjl@xxxxxxxxxx)
Date: Fri Mar 08 2002 - 19:26:59 GMT-3
Thx, Jim,
Actually I want to verify the BGP Best Path Selection Algorithm. According
to Jeff Doyle TCP/IP Vol II, Page 291-292,
it says:
All else being equal, the BGP decision process choose normal EBGP routes
over confederation EBGP routes and confederation EBGP routes over IBGP
routes.
But according to Cisco CCO DOC:
http://www.cisco.com/warp/customer/459/25.shtml:
1.. Prefer external (eBGP) over internal (iBGP) paths. Note: Paths
containing AS_CONFED_SEQUENCE are local to the confederation, and therefore
treated as internal paths. There is no distinction between Confederation
External and Confederation Internal.
2.. Prefer the path with the lowest IGP metric to the BGP next hop.
So, which one is right?
When I did the lab, I simulate the 172.17.11.0 route one from the iBGP peer,
and one comes from the eBGP peer, so I think it should chooses the eBGP
route as the best path. But the result is not. I was confused. Note:(65535)
is a confederation peer.
-----------
Could someone please help me to explain the following, why the bgp chooses
the iBGP as the best one?
Why doesn't it choose the external one as the best one?
P1R3>show ip bgp
Network Next Hop Metric
LocalProf Weight Path
* >i 172.17.11.0/24 10.33.255.2 0
100 0 (65535) i
* 10.34.255.1 0
100 0 (65535) i
* 172.17.255.1 0
0 1000 i
P1R3> show ip bgp 172.17.11.0
BGP routing table entry for 172.17.11.0/24, version 29
Paths: (3 available, best#1, table Default-IP-Routing-Table)
Advertised to non peer-group peers:
10.34.255.1 172.17.255.1
(65535)
10.33.255.2 (metric 65) from 10.33.255.2 (10.33.255.2)
Origin IGP , metric 0, localpref 100, valid,
internal, best
(65535)
10.34.255.1 (metric 65) from 10.34.255.1 (172.17.11.1)
Origin IGP , metric 0, localpref 100, valid,
confed-external
1000
172.17.255.1 (metric 65) from 172.17.255.1 (172.17.255.1)
Origin IGP , metric 0, localpref 100, valid,
external
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 10:56:57 GMT-3