Hi all,
I've got this weird situation in work that I can't get my head around.....
We have two WAN routers. One is an MPLS router that is used to receive all routes from our WAN as well as a default, the routes are advertised via iBGP to router R1. To reduce the amount of Internet traffic going across the corporate WAN an ADSL service has been deployed. The ADSL also acts as a backup for the MPLS circuit so that if the MPLS circuit fails an IPSec tunnel is established over the Internet back into the corporate network.
I've attached a diagram that I hope will assist with the issue I am about to discuss.
R2 has a default route pointing to the Internet router. On R2 RIPv2 is configured to advertise this default to R1 (historical reasons) and is configured with an Administrative Distance of 10 so this default is preferred over the BGP advertised default. R1 and R2 are also configured to use OSPF. On R2 the 'default-information originate metric 100' is enabled under the OSPF process. Under R1 the same command is used but with a metric of 50. The idea is that if R1 dies the alternate route to R2 will be used by .115.
The WAN routers patch into a stack of L2 switches, which are then connected to the L3 device. I know this isn't good practice, but this is legacy. An SVI is then used on the L3 switch to talk to the two WAN routers, the SVI is used as other devices (WAN optimisation) exist in the VLAN inline to the WAN routers.
The issue I have is that R1 is advertising a default route to .113 and well as a default route to .114 with the same metric. This is seeing traffic being load balanced by the router which isn't desirable. I was hoping to see that traffic would only route over .113 until this router fails when it should route over .114.
Any idea why R1 is advertising both routes with the same metric?
In the design .115 is the DR, and .144 is the BDR.
