In the previous post, we learnt about connected and static routes. In today’s post, we would move on to more interesting stuff – Dynamic Routing. For the CCNA Examination, you need to understand RIP, EIGRP (Single AS) and Single Area OSPF. OSPF and EIGRP have already been examined on this site so this post would be focused on RIP.

For the CCNA exam, you need to know RIP configuration, auto-summarization, split horizon and passive interfaces. In this post I would explain all these using practical examples. I encourage you to use whatever simulation tools you have to practice all the concepts so that you can maximize your personal learning experience.

CCNA Training – Resources (Intense)

RIP – Routing Information Protocol is the most basic routing protocol. It is supported on most network equipment (multi-vendor support) and it is quite easy to deploy. Before we look at an example, there are a few things to remember from the IPv4 Addressing IP routing theory post:

  • RIP is a distance vector protocol – it only uses the distance and the direction (vector) to learn about the routes.
  • Metric: The quality of the route is determined by the number of hops (routers) between a router and the destination network. The lower the hop count, the better.

Our Sample Network is shown below:

 

The network has been set up so that directly connected routers can ping each other. Right now, the London Router can ping the F0/0 of the New_York Router and the Toronto Router can ping the F0/1 interface of the New_York Router. No static routing has been set up so there is no reachablility beyond directly connected interfaces.

Your mission, should you choose to accept it, is to ensure complete reachablilty across the entire network without using any static routing. You may begin.

First let’s examine the routing table of the London router:

Tip: If you need to issue an exec command (e.g. show, copy, debug, etc.) from the config mode, you can just prefix the command with ‘do’ – as seen in the ‘do show ip route’ exhibit above. That way, you don’t have to always exit out of the config mode.

From the routing table, the router has its connection to New_York and three other networks (represented by loopbacks in this case).

Also notice the way the routes are represented: 172.16.0.0/24 shows that the router consists of a major class B network (172.16.0.0/16), which has been subnetted into /24 subnets. If you still have issues with subnetting, please read the subnetting post and drop a comment below for any clarifications you may need.

Similarly, New_York and Toronto’s routing tables are shown below:

To configure RIP, you issue the ‘router rip’ command from the config mode. There are 2 versions: version 1 is quite old and is being deprecated, so you should be configuring version 2. By default, Cisco Routers listen to v1 and v2 packets but only send version 1 RIP packets. To configure version 2, you should just issue the version 2 command under the router configuration mode. To advertise the networks into RIP, enter the classful address of the network using the ‘network’ command.

The basic RIP configurations for the London, New_York and Toronto Routers are shown below:

Now we have to wait for a while because RIP updates happen every 30 seconds. So we need to sit down, cross our fingers and wait for the routers to learn about the routes. Always remember this so you do not freak out when working under pressure.

Okay, let us check the routing tables and see the new routes that have been learnt by the routers. London first:


Two routes have been learnt via RIP: 192.168.12.0/24 and 192.168.1.0/24.

The routing table for the New York router is shown below:

The New_York Router has learnt 4 routes (three routes from 172.16.12.2 and one from 192.168.12.2).

Similarly, the Toronto Router has learnt the 172.16.0.0/16 network via RIP.

Notice anything strange?

Take a closer look at the routing tables for New_York and Toronto routers. Although the New York router learns about the entire networks behind the London router (172.16.1.0/24, 172.16.2.0/24 and 172.16.3.0/24), it only sends a Class B summary network to Toronto (172.16.0.0/16).

Also, Toronto only sends a Class C summary (192.168.1.0/24) to London instead of the individual networks in its routing table (192.168.1.0/29, 192.168.1.8/29 and 192.168.1.16/29).

This phenomenon is called Auto-summarization. Auto-summarization is the summarization of individual networks into major classful networks at boundaries. This simply means that whenever a router is sending an update about networks that are members of a classful range (remember Class A, Class B and Class C?) to a router that is not in that classful range, it would send a summary!

Now let’s look at the London router again. The individual networks (172.16.1.0/24, 172.16.2.0/24 and 172.16.3.0/24) are a part of the major class B range – 172.16.0.0/16. When sending an update to 172.16.12.1, it checks if it’s a member of that range (172.16.0.0/16) and since it is, it sends all the routes to New_York, which is why New_York gets the individual routes.

When New_York wants to forward the updates to Toronto, the same check is performed. Since Toronto (192.168.12.2) is not part of the Class B range (172.16.0.0/24), New_York only sends a summary (172.16.0.0/24) to Toronto.

Also, when Toronto wants to send updates about its networks (192.168.1.0/29, 192.168.1.8/29 and 192.168.1.16/29) to New_York, it checks if New_York is part of the major classful network (192.168.1.0/24). Since New_York’s address is 192.168.12.1, it is NOT part of the major Class C network, so it sends a summary to New_York as we saw in the routing table.

Auto-summarization is enabled by default in RIP (both version 1 and 2). To disable auto-summarization, we use the command “no auto-summary” under the router configuration mode as shown below:

After disabling auto-summarization on all routers, the new routing tables would look like this:


The individual routes from Toronto are now being advertised all the way to London.

One other thing to notice is that the administrative distance of these routes is 120 (default for RIP) and the metric is 2. The metric is based on the Hop count, which is the number of routers between London and the networks. The routers between London and the networks are New_York and Toronto, and they are 2 routers.

Before we conclude on RIP and move on to EIGRP, let’s talk about 2 other RIP concepts that are important for the CCNA exam:

Split Horizon

This concept is basic; the rule is to ensure that updates are not sent out the same interface that they are learnt from. So in this case, updates that the New_York router receives from London (F0/0 interface) would not be sent back out through F0/0. This sounds like common sense, right? Wrong!

Think about point to multipoint interfaces (like Frame-relay), where updates received from one interface may need to be sent back out through the same interface. In RIPv2, split horizon is turned on by default. To disable split-horizon, issue the interface command “no ip split-horizon”.

Passive interface: The passive interface is used to prevent RIP from sending updates out of an interface. You should note that this does not prevent RIP from receiving updates from that interface or installing the routes in the routing table. This is useful in cases where you might have RIP servers on the LAN (may be configured on Windows servers or Linux servers) and you do not want to send updates to these servers.

Using our sample network, passive interface should be issued on all the interfaces of the London Router except the F0/0 interface. Similarly, passive interface should be issued on all the interfaces of the Toronto router except the F0/0 interface. Passive interface is configured with the command “passive-interface <interface-name number>” and can be removed with the “no passive-interface command”.

When you have many interfaces on the router, you can use the “passive-interface default command” and use “no passive-interface <interface-name number>” to turn on the interface that should send routing protocol updates. The passive interface configuration for London and Toronto routers are shown below:

It is important to note that passive interface alone is not enough security for RIP. More security features include RIP Authentication, but this is beyond the scope of the CCNA exam.

Finally, you should be familiar with some show commands to verify your configuration.

Some of the important ones include:

Show ip route – Displays your routing table.

Show ip protocols: displays a summary of the routing protocols that are enabled on the network. It gives an indication of the timers, passive interfaces, metrics and other useful information.

So there you have it: Routing Information Protocol for the CCNA exam! This wraps up all the IP routing knowledge that you need to have in order to pass the CCNA exam. If you have any questions on your mind, feel free to use the comments section. See you soon!