I got an issue with reference to attach design.
Traffic from Loc_A designated to Loc_C passes Loc_B, but if link between Loc_B and Loc_A is down then it goes directly.
How do I influence traffic from Loc_A designated to Loc_C to go direct.
Loc_A is connected to Loc_B via Metro Ethernet ( OSPF is the routing protocol )
Loc_A is connected to Loc_C via IPSEC_GRE Tunnel ( EIGRP for tunnel interface )
Loc_B is connected to Loc_C via IPSEC_GRE Tunnel ( Eigrp for Tunnel Interface )
On Loc_A / Loc_B / Loc_C there is mutual redistribution between EIGRP and OSPF
As a result of the mutual redistribution the subnets in question look better over the OSPF Metro E with an admin distance of 110. Networks redistributed into EIGRP have an admin distance of 170.
Why not run a single routing protocol for one homogonous routing domain?
You could use then ‘distance’ command under EIGRP or OSPF to make specific routes in question look more attractive over the tunnel network. You could also use the ‘distance eigrp’ command to change the admin distance for EIGRP globally. These two options don’t generally scale well but they are effective.
Here’s a good admin distance doc:
thanks for replying, with distance command it doesnt work either. Initial config of Loc_A / Loc_B / Loc_C are attached.
In near future we plan to have more internet links with unequal bandwidth, so I selected EIGRP.
I also have plan to apply route-map to restrict the flow of subnet between GRE Tunel
Distance command not working as expected.
Loc_A receives routes for Loc_C from Loc_B and Loc_C but prefers Loc_B.
Can I get help on influence the traffic flow.
#sh ip route eigrp
D EX 192.168.88.0/24 [170/10496000] via 172.20.1.2, 00:06:51, Tunnel4
show ip route 192.168.88.0
Routing entry for 192.168.88.0/24
Known via "ospf 1", distance 110, metric 20, type extern 2, forward metric 3
Redistributing via eigrp 2
Advertised by eigrp 2
Last update from 172.16.1.1 on FastEthernet0/1, 00:07:23 ago
Routing Descriptor Blocks:
* 172.16.1.1, from 172.20.0.9, 00:07:23 ago, via FastEthernet0/1
Route metric is 20, traffic share count is 1
Try this config on the WAN routers where mutual redistribution occurs.
route-map eigrp2ospf deny 10
match tag 111
route-map eigrp2ospf permit 20
set tag 222
route-map ospf2eigrp deny 10
match tag 222
route-map ospf2eigrp permit 20
set tag 111
router ospf 1
redistribute eigrp 2 subnets type 1 route-map eigrp2ospf
router eigrp 2
redistribute ospf 1 route-map ospf2eigrp
This prevents routes from back feeding into the originating protocol. It will also set OSPF external routes to type 1. The metric of a type 1 route will increase as it’s propagated throughout the network. This should be helpful in this topology. If you still have sub-optimal pathing after this change adjusting the OSPF metrics should fix it.