Static Routing

Lessons Learned:                   

Default Routing

-To a next-hop

--use the layer 2 address of the next hop for all layer 3 destinations.

To a multipoint interface

-all destinations require layer3 to layer2 resolution

-layer 2 mapping table size problems.

To a point-to-point interface

-no layer 3 to layer 2 resolution required

Connected route:

R2#sh ip route 12.0.0.1

Routing entry for 12.0.0.0/24

  Known via "connected", distance 0, metric 0 (connected, via interface) – Shows the route is a connected route

  Routing Descriptor Blocks:

  * directly connected, via FastEthernet0/0 – shows the interface the next hop is connected to.

      Route metric is 0, traffic share count is 1 – Metric is 0

Static route:

R5#sh ip route 12.0.0.0 255.255.255.0

Routing entry for 12.0.0.0/24

  Known via "static", distance 1, metric 0 – show the route type is static

  Routing Descriptor Blocks:

  * 23.0.0.2  -- show’s the next hop router for the specified network.

      Route metric is 0, traffic share count is 1

SH IP ROUTE:

R5#sh ip route

     23.0.0.0/24 is subnetted, 1 subnets

C       23.0.0.0 is directly connected, FastEthernet0/1

     12.0.0.0/24 is subnetted, 1 subnets

S       12.0.0.0 [1/0] via 23.0.0.2 _ notice there’s no metric only administrative distance and the next hop address.

     14.0.0.0/24 is subnetted, 1 subnets

C       14.0.0.0 is directly connected, FastEthernet0/0

R5#

Let’s change the route statement for the 12.0.0.0 / 24 network to point to an interface.

# no ip route 12.0.0.0 255.255.255.0 23.0.0.2

# ip route 12.0.0.0 255.255.255.0 fastEthernet 0/1

Note: this should not affect the connectivity and we should still have reachability

You would not want to do this in normal situations. For this lab the connection still works….

R5#ping 12.0.0.1       

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 12.0.0.1, timeout is 2 seconds:

.!!!!

Once we change this to the interface we will need to send a request for the layer 2 information.

R5#debug arp

ARP packet debugging is on

R5#ping 12.0.0.3

Type escape sequence to abort.

Sending 5, 100-byte ICMP Echos to 12.0.0.3, timeout is 2 seconds:

*Mar  1 01:17:53.283: IP ARP: creating incomplete entry for IP address: 12.0.0.3 interface FastEthernet0/1

*Mar  1 01:17:53.283: IP ARP: sent req src 23.0.0.5 c005.18d4.0001,

                 dst 12.0.0.3 0000.0000.0000 FastEthernet0/1

*Mar  1 01:17:53.363: IP ARP: rcvd rep src 12.0.0.3 c001.21dc.0001, dst 23.0.0.5 FastEthernet0/1.....

Success rate is 0 percent (0/5)

Note: 12.0.0.3 is not a vaild IP address of the 12 subnet. Also notice that the arp cache is not based on the Interface and not necessarily the next hop ip address. The Arp cache now points to the interface.  

Default routing to the interface:

Another issue when setting the next-hop to an interface is that the arp cache can get very large.

An example could be a border router that has a def-route upstream. Unless its peering BGP and has a full route table. The arp cache will be very large for all the upstream requests.

Basically when the router does it’s lookup and for any destination the router doesn’t have a longest match for – send the traffic to the interface. The problem again is that the router doesn’t know who is the layer 2 neighbor on the link. It assumes whatever the destination is; is connected to the interface.