[toc:faq]
Secondary IP
You can configure more than one IP under FE0/0 (WAN) using the following command:
ip address x.x.x.x y.y.y.y secondary
where x.x.x.x is the IP address and y.y.y.y is the mask.
Dual WAN Interface
You can also create a secondary WAN IP interface following a procedure similar to this one:
https://www.myciscocommunity.com/message/7933#7933
In the example above, things like NAT, DHCP client on the interface, etc. would have to be modified accordingly.
VLAN Trunk
Alternatively, you can create a do1q trunk between the WAN interface and a switch, and have two different IP addresses on the router side (one for each subinterface). Then somehow connect the switch to your ISP(s).
SLA Probe using ICMP
As far as configuring redundancy, consider this:
UC500-Switch-ADSL1-----ISP Router 1
|
ADSL2-----ISP Router 2
You can implement a configuration with IP SLA probes that monitor a particular IP using ICMP echo replies (ping). You can then associate a static route to that IP SLA condition, so if the IP becomes unavailable, so does the static route.
Now, what IP do we monitor? If the customer has static IP's on both WAN links, ideally we would like to monitor ISP Router 1' s IP and ISP Router 2's IP. If they don't, we can monitor ADSL 1's and 2's respectively, but this will only work if the actual ADSL is a router (not bridging) and it's down (which is less probable than the actual WAN link going down).
So, if your customer has (or is willing to have) static IP's on the WAN, propose the following configuration:
!
ip sla 1
icmp-echo **IP of ISP Router 1**
timeout 10
threshold 2
frequency 10
ip sla schedule 1 life forever start-time now
!
ip sla 2
icmp-echo **IP of ISP Router 2**
timeout 10
threshold 2
frequency 10
ip sla schedule 2 life forever start-time now
!
track ip sla 1 reachability
!
track ip sla 2 reachability
!
ip route 0.0.0.0 0.0.0.0 **IP of ISP Router 1** track 1
!
ip route 0.0.0.0 0.0.0.0 **IP of ISP Router 2** track 2
!
end
For these last two commands, you can also play with the administrative distance, in case you want link redundancy instead of load balancing.