From: Victor Cappuccio (vcappuccio@gmail.com)
Date: Thu Oct 16 2008 - 10:27:33 ART
Hi Geerts
I got interested in your problem, these is what I came out, I hope I
understood your issue, and that with these you can solve the problem
OSPF Routes R1 -- BGP R2
R2(config)#do show ip bgp summ | b Neigh
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down
State/PfxRcd
200.12.12.1
R2(config-if)#do show run | in ip route
ip route 10.20.1.1 255.255.255.255 Null0
ip route 10.20.2.2 255.255.255.255 Null0
ip route 10.20.3.3 255.255.255.255 Null0
R2(config-if)#no ip route 10.20.1.1 255.255.255.255 Null0
R2(config)#
*Oct 16 15:21:09.863: RT: del 10.20.1.1/32 via 0.0.0.0, static metric [1/0]
*Oct 16 15:21:09.863: RT: delete subnet route to 10.20.1.1/32
*Oct 16 15:21:09.867: RT: NET-RED 10.20.1.1/32
*Oct 16 15:21:10.023: RT: del 10.0.0.0/8 via 200.12.12.1, bgp metric [20/0]
*Oct 16 15:21:10.023: RT: delete subnet route to 10.0.0.0/8
*Oct 16 15:21:10.023: RT: NET-RED 10.0.0.0/8
R1(config)#do show ip route ospf
10.0.0.0/8 is variably subnetted, 11 subnets, 4 masks
O 10.10.5.5/32 [110/2] via 10.1.13.3, 00:07:08, FastEthernet1/0
O 10.10.4.4/32 [110/2] via 10.1.13.3, 00:07:08, FastEthernet1/0
O 10.10.3.3/32 [110/2] via 10.1.13.3, 00:07:08, FastEthernet1/0
O 10.10.2.2/32 [110/2] via 10.1.13.3, 00:07:08, FastEthernet1/0
O 10.6.0.0/16 [110/2] via 10.1.13.3, 00:07:08, FastEthernet1/0
R1(config)#
router bgp 100
no synchronization
bgp log-neighbor-changes
aggregate-address 10.0.0.0 255.0.0.0 advertise-map X
redistribute ospf 1 match internal external 1 external 2
neighbor 200.1.12.2 remote-as 200
neighbor 200.1.12.2 ebgp-multihop 255
neighbor 200.1.12.2 update-source FastEthernet1/1
no auto-summary
route-map X permit 10
match ip address prefix-list II
ip prefix-list II seq 5 permit 10.20.1.1/32
Verification
R1(config)#do show ip bgp summ | b Neigh
Neighbor V AS MsgRcvd MsgSent TblVer InQ OutQ Up/Down
State/PfxRcd
200.1.12.2 4 200 18 21 16 0 0 00:05:27 3
R1(config)#do show ip bgp neigh 200.1.12.2 ad | in 10.0.0
*> 10.0.0.0 0.0.0.0 32768 i
R1(config)#do show ip route 10.20.1.1
Routing entry for 10.20.1.1/32
Known via "bgp 100", distance 20, metric 0
Tag 200, type external
Last update from 200.1.12.2 00:01:48 ago
Routing Descriptor Blocks:
* 200.1.12.2, from 200.1.12.2, 00:01:48 ago
Route metric is 0, traffic share count is 1
AS Hops 1
Route tag 200
R2(config)#no ip route 10.20.1.1 255.255.255.255 Null0
R2(config)#
*Oct 16 15:25:37.939: RT: del 10.20.1.1/32 via 0.0.0.0, static metric [1/0]
*Oct 16 15:25:37.943: RT: delete subnet route to 10.20.1.1/32
*Oct 16 15:25:37.943: RT: NET-RED 10.20.1.1/32
*Oct 16 15:25:38.275: RT: del 10.0.0.0/8 via 200.12.12.1, bgp metric [20/0]
*Oct 16 15:25:38.275: RT: delete subnet route to 10.0.0.0/8
*Oct 16 15:25:38.279: RT: NET-RED 10.0.0.0/8
R1(config)#do show ip route 10.20.1.1
% Subnet not in table
R1(config)#do show ip bgp neigh 200.1.12.2 ad | in 10.0.0
R1(config)#
Victor.-
On Thu, Oct 16, 2008 at 2:03 PM, Geert Nijs <Geert.Nijs@simac.be> wrote:
> Hi group,
>
> ----OSPF----|ROUTER|----BGP
>
> I receive both from OSPF and from BGP routes in the 10.0.0.0/8 range.
>
> Lets say: 10.10.0.0/16 (and smaller) from OSPF and 10.20.0.0/16 (and
> smaller) from BGP
>
> OSPF is redistributed into BGP.
> So in the BGP table, both ranges are present.
>
> I would like to generate an aggregate route in BGP to summarize all
> 10.0.0.0/8 networks.
>
> However, this aggregate can only be based on the BGP received routes. The
> OSPF 10.10.0.0/16 range should not count (or be considered) in generating
> the aggregate.
> The goal is that if the BGP session should fail, the aggregate should
> disappear (even if there are still OSPF routes present in the range of the
> aggregate).
>
> I thought about something like:
>
> -----------------------------------------------
> router bgp xxxx
> aggregate-address 10.0.0.0 255.0.0.0 suppress-map SUPROUTES
>
> ip prefix-list OSPFROUTES seq 10 permit 10.10.0.0/16 le 32
>
> route-map SUPROUTES permit 10
> match ip address prefix-list OSPFROUTES
> --------------------------------------------------
>
> However, this doesn't work. The suppress-map -only- supresses the
> advertisement of the more specific routes, the routes
> are still taken into account for generating the aggregate.
> If i shut the BGP session, the aggregate is still installed.
>
> Is there a way to just -completely ignore- 'some' routes when generating an
> aggregate ?
>
> regards,
> Geert
>
>
>
> ________________________________
> disclaimer : http://webservices.simac.be/disclaimer.htm
>
>
> Blogs and organic groups at http://www.ccie.net
>
> _______________________________________________________________________
> Subscription information may be found at:
> http://www.groupstudy.com/list/CCIELab.html
>
>
>
>
>
>
>
>
-- Victor Cappuccio CCIE R/S# 20657 CCSI# 30452 www.anetworkerblog.comBlogs and organic groups at http://www.ccie.net
This archive was generated by hypermail 2.1.4 : Sat Nov 01 2008 - 15:35:21 ARST