RIP exercises

In this section, we will examine the operation of RIP.

Exercise - Set up RIP

The virtual routers in our experiment are running quagga, a software router that includes a shell interface similar to the Cisco IOS interface. Open the shell interface on each router with:

export VTYSH_PAGER=more
sudo vtysh

(If you get an error message sudo: vtysh: command not found or Exiting: failed to connect to any daemons at this point, the configuration script hasn’t finished running! Wait a few minutes for it to finish.)

In the quagga shell, run

show ip route

to see what routes are currently configured on each router. Save this output.

Using this shell interface, we can configure the routers to use the RIP protocol.

First, enter Global Configuration mode in each router:

configure terminal

The prompt should change (to include a (config) at the end), indicating that you are now in configuration mode.

Then, type

router rip

to enable RIP. Finally, you need to associate one or more networks to the RIP routing process. Run

network 10.10.0.0/16

so that all addresses from 10.10.0.0-10.10.255.255 will be enabled for RIP. (Note: this syntax is slightly different in quagga then in Cisco IOS.) Then run exit twice, until you are back in the regular quagga shell (not in config mode).

RIPv2 multicasts its routing table every 30 seconds to the multicast IP address 224.0.0.9. On the workstations in each network segment (romeo, hamlet, othello, petruchio), run

sudo tcpdump -en -i eth1 -w $(hostname -s)-rip.pcap

for about a minute or two, to capture these messages.

Then, run

show ip route

in the quagga shell on each router, to see the new routes, and save the output.

Also see RIP-specific information in the quagga shell on each router with

show ip rip

Note that the “Metric” column here shows the hop count to each destination network. Save all outputs.

After a few minutes, you can stop the tcpdump processes on the workstations with Ctrl+C. Transfer these to your laptop with scp, or play them back with tcpdump using:

sudo tcpdump -r $(hostname -s)-rip.pcap -env

Lab report: Show the RIP messages received by router-4. Using these RIP messages, draw the distance table and the routing table at router-4, assuming that number of hops is used as the metric. Compare to the output of show ip rip and show ip route at router-4.

Exercise - RIP response to link failure

In this exercise, we will examine how RIP reponds to link failures. We will bring down the interface on router-1 that connects to the LAN with router-2 and hamlet, and we will observe how the routing tables adapt to the change in the topology. Then, we will bring this interface back up, and observe the changes again.

First, on any router, run

show ip rip status

in the quagga shell, and save the output. Make a note of two important timer values: how often each router sends updates, and after how long without an update a route is removed from the routing table (the timeout value).

Start tcpdump on each of the four workstations (romeo, hamlet, othello, petruchio) with

sudo tcpdump -i eth1 -w $(hostname -s)-rip-failure.pcap

Let these run during this exercise.

On each of the routers, in the quagga shell, run

show ip rip

to see the current RIP database. Save the output.

On router-1, idenfity the name of the interface that has the address 10.10.62.1 (e.g. eth1 or eth2). (You can refer to your previous ifconfig output, or you can use the show ip route output in the quagga shell, and look for the name of the interface that is directly connected to the 10.10.62.0/24 subnet.) This is the interface that connects Router 1 to the network segment that Router 2 is on. You will use this interface name in the following commands.

Then, on Router 1, use the quagga shell to bring down this interface. Run

configure terminal
interface eth1
shutdown

(or substitute eth2 if that is the name of the interface with address 10.10.62.1). Then, run exit twice to return to the regular quagga shell.

Run

show ip rip

again in the quagga shell on each router. You may see some transient routing rules with a metric of 16; in RIPv2, the maximum hop count is 15, and 16 is considered an “infinite” hop count, i.e. an unreachable network.

Keep running

show ip rip

on each router. For any routers that still have a route via 10.10.62.1, note the “Time” column in the output. You should see the timer slowly running out, until eventually the route is removed.

Once there are no more routes via 10.10.62.1 on any of the routers, get the final routing tables on all four routers with

show ip rip

and save the output. Then, in the quagga shell on Router 1, run

configure terminal
interface eth1
no shutdown

(or substitute eth2 if that is the name of the interface with address 10.10.62.1) to bring back up the disabled interface. Also, run exit twice until you return to the regular quagga shell.

Again, run

show ip rip

on all four routers. Keep checking until the original routes have been restored, and then save the final

show ip rip

output.

Wait at least one more minute. Then, use Ctrl+C to stop the tcpdump processes, and retrieve them from the workstations with scp. You can also play back these packet captures with

tcpdump -r $(hostname -s)-rip-failure.pcap -env

Lab report: From the output of show ip rip status, how often do the routers send updates? After how long without updates will a route be removed from the routing table (the timeout value)?

Lab report: Show the RIP tables and RIP messages you captured at each stage of this exercise. Explain how the routing tables changed when the most direct path between Router 1 and Router 2 was disabled, and then changed again when the direct path was re-established.