From: Keith Kruepke (lister@xxxxxxxxxxx)
Date: Thu Aug 10 2000 - 12:21:56 GMT-3
Sam,
It was a good theory, but...
Someone pointed out that I was close but missing just one fact. The RIP route
is not replaced--it is 4.0.0.0/8, so it coexists with the BGP route in the IP r
outing table. But the BGP route is to 4.4.4.0/24, so it is more specific when
trying to reach 4.4.4.4. I probably should have seen the 4.0.0.0/8 route, sinc
e it was immediately below the 4.4.4.0/24 in the routing table...
Because it was not a problem with the AD, as I had suggested, unfortunately, th
is solution would not quite solve the problem. (If I had been right, then this
would definitely be a good solution.) I think the best suggestion was to just
leave the 4.4.4.0/24 network out of BGP completely...
Keith
----- Original Message -----
From: "Sam Munzani" <sam@chinet.com>
To: "Keith Kruepke" <lister@kruepke.com>
Cc: "Scott" <trig@usa.net>; <ccielab@groupstudy.com>
Sent: Thursday, August 10, 2000 9:54 AM
Subject: Re: BGP routing loop?
Keith,
I totally agree with your theory. What you think about following as a
solution. I have not tested it but just had an idea.
On each router user BGP attribute backdoor command so it will treat bgp
4.4.4.0 routes with 200 metric instead of 20. This will solve his
problem. In real life static routes is the best option. However, I heard
in lab they strictly say not to use any static routes.
Sam
On Wed, 9 Aug 2000, Keith Kruepke wrote:
> Scott,
>
> I think this line from your routing table may be part of the problem:
>
> B 4.4.4.0/24 [20/0] via 4.4.4.4, 00:00:36
>
> The route to 4.4.4.0/24 goes through 4.4.4.4--that doesn't seem right to me.
>
> I'm guessing that 4.4.4.0 is initially learned via RIP, so that you can get t
o 4.4.4.4 and start an EBGP session. Then the EBGP neighbor advertises 4.4.4.0
to router2, and that route would take precedence in the IP routing table (AD o
f 20 is much better than RIP). Then the RIP route is gone, so connectivity is
lost, and the EBGP route disappears. Then the RIP route returns, and connectiv
ity is reestablished, and the cycle continues.
>
> Based on the description, this is my theory, but maybe someone has another id
ea? I think this is part of the reason you would normally want to use the dire
ctly connected interface IPs as the neighbor addresses for EBGP under most circ
umstances. If you still want to use the loopbacks, I would put in static route
s to those networks, rather than using RIP.
>
> Keith
>
> ----- Original Message -----
> From: "Scott" <trig@usa.net>
> To: <ccielab@groupstudy.com>
> Sent: Wednesday, August 09, 2000 9:49 PM
> Subject: BGP routing loop?
>
>
> Hello
>
> I was trying to connect two AS with a EBGP multihop command between two
> loopbacks and it looks like I am getting a some type of routing loop. The BGP
> routes are put in the routing table and removed every 60 seconds (see
> attached). I have IP connectivity via RIP and the connection is showing as
> established. Any suggestions? Thanks
>
> Scotty
>
>
> router2#show ip route
> Codes: C - connected, S - static, I - IGRP, R - RIP, M - mobile, B - BGP
> D - EIGRP, EX - EIGRP external, O - OSPF, IA - OSPF inter area
> N1 - OSPF NSSA external type 1, N2 - OSPF NSSA external type 2
> E1 - OSPF external type 1, E2 - OSPF external type 2, E - EGP
> i - IS-IS, L1 - IS-IS level-1, L2 - IS-IS level-2, * - candidate
> default
> U - per-user static route, o - ODR
>
> Gateway of last resort is not set
>
> 138.6.0.0/24 is subnetted, 2 subnets
> B 138.6.5.0 [200/0] via 172.16.35.5, 00:09:52
> B 138.6.1.0 [200/0] via 172.16.35.5, 00:09:52
> 4.0.0.0/8 is variably subnetted, 2 subnets, 2 masks
> B 4.4.4.0/24 [20/0] via 4.4.4.4, 00:00:36
> R 4.0.0.0/8 [120/11] via 172.16.24.4, 00:00:27, Serial1
> 172.16.0.0/24 is subnetted, 5 subnets
> O IA 172.16.35.0 [110/74] via 172.16.23.3, 00:10:44, Ethernet0
> C 172.16.24.0 is directly connected, Serial1
> C 172.16.23.0 is directly connected, Ethernet0
> C 172.16.12.0 is directly connected, Serial0
> C 172.16.2.0 is directly connected, Loopback0
> router2#ping 4.4.4.4
>
> Type escape sequence to abort.
> Sending 5, 100-byte ICMP Echos to 4.4.4.4, timeout is 2 seconds:
> .....
> Success rate is 0 percent (0/5)
> router2#
> 05:08:12: BGP: no valid path for 4.4.4.0/24
> 05:08:12: BGP: nettable_walker 4.4.4.0/24 no best path
> 05:08:12: BGP: 4.4.4.4 computing updates, neighbor version 15, table version
> 16,
> starting at 0.0.0.0
> 05:08:12: BGP: 4.4.4.4 update run completed, ran for 0ms, neighbor version 15
,
> s
> tart version 16, throttled to 16, check point net 0.0.0.0
> 05:08:12: BGP: 172.16.23.1 computing updates, neighbor version 15, table
> version
> 16, starting at 0.0.0.0
> 05:08:12: BGP: 172.16.23.1 send UPDATE 4.4.4.0/24 -- unreachable
> 05:08:12: BGP: 172.16.23.1 1 updates enqueued (average=27, maximum=27)
> 05:08:12: BGP: 172.16.23.1 update run completed, ran for 8ms, neighbor versio
n
> 1
> 5, start version 16, throttled to 16, check point net 0.0.0.0
> 05:08:12: BGP: 172.16.23.3 computing updates, neighbor version 15, table
> version
>
> !
> interface Loopback0
> ip address 172.16.2.2 255.255.255.0
> no ip directed-broadcast
> !
> !
> interface Serial1
> ip address 172.16.24.2 255.255.255.0
> no ip directed-broadcast
> clockrate 64000
> !
> interface BRI0
> no ip address
> no ip directed-broadcast
> shutdown
> !
> router ospf 3
> network 172.16.23.0 0.0.0.255 area 3
> !
> router rip
> version 2
> network 172.16.0.0
> default-information originate
> !
> router bgp 3
> no synchronization
> network 172.16.2.0 mask 255.255.255.0
> neighbor 4.4.4.4 remote-as 4
> neighbor 4.4.4.4 ebgp-multihop 5
> neighbor 4.4.4.4 update-source Loopback0
> neighbor 4.4.4.4 distribute-list 2 out
> neighbor 172.16.23.1 remote-as 3
> neighbor 172.16.23.3 remote-as 3
> !
>
>
> Router 1
> !
> interface Loopback0
> ip address 4.4.4.4 255.255.255.0
> no ip directed-broadcast
> !
> interface Loopback1
> ip address 215.6.10.1 255.255.255.0
> no ip directed-broadcast
> !
> interface Loopback2
> ip address 215.6.11.1 255.255.255.0
> no ip directed-broadcast
> !
> interface Loopback5
> no ip address
> no ip directed-broadcast
> !
> interface Ethernet0
> no ip address
> no ip directed-broadcast
> appletalk cable-range 600-600 600.5
> appletalk zone biteme
> appletalk protocol eigrp
> bridge-group 1
> !
> interface Serial0
> ip address 172.16.24.4 255.255.255.0
> no ip directed-broadcast
> no ip mroute-cache
> no fair-queue
> !
> interface Serial1
> ip address 172.16.10.10 255.255.255.252
> no ip directed-broadcast
> bridge-group 1
> !
> interface BRI0
> no ip address
> no ip directed-broadcast
> shutdown
> !
> router rip
> version 2
> offset-list 1 out 10
> network 4.0.0.0
> network 172.16.0.0
> !
> router bgp 4
> no synchronization
> network 4.4.4.0 mask 255.255.255.0
> neighbor 172.16.2.2 remote-as 3
> neighbor 172.16.2.2 ebgp-multihop 5
> neighbor 172.16.2.2 update-source Loopback0
>
>
This archive was generated by hypermail 2.1.4 : Thu Jun 13 2002 - 08:24:23 GMT-3