Cisco's Routing Exam: A Tale of Three Protocols
Love tedious details? Can't wait to learn everything you can about BGP, EIGRP and OSPF? Then you'll probably enjoy studying for Cisco's Routing exam.
by Richard Shanks - courtesy of TCPMag.com
11/14/2001 --
|
|
|
Exam |
|
|
|
Routing (640-5037) |
|
Certification, Vendor |
|
|
|
CCNP or CCDP, Cisco |
|
Status |
|
|
|
Live |
|
Reviewer's Rating |
|
|
|
"A tedious but fair exam contrasting three routing protocols." |
|
Test Information |
|
|
|
62 questions, 17 minutes, need 690 on a scale of 300 to 1,000 to pass. Cost: $125 (U.S.) |
|
Who Should Take This Exam? |
|
|
|
Those pursuing Cisco's CCNP or CCDP certifications. |
|
What Classes Prepare You? |
|
|
|
Building Scalable Cisco Networks. |
|
Test Objectives |
|
|
|
Click here |
|
|
|
|
|
|
|
|
|
|
|
|
|
The Routing exam is one of four required to attain Cisco's intermediate networking certification, the Cisco Certified Network Professional (CCNP). It also counts toward Cisco's Certified Design Professional (CCDP) program.
As the name implies, the Routing exam (based on the Building Scalable Cisco Networks course) covers the routing issues that a CCNP/CCDP candidate should be familiar with - which, not surprisingly, is quite a lot.
Before I even began studying for the Routing exam, I knew that it was going to be short on theory and long on details and command syntax. After browsing the Building Scalable Cisco Networks table of contents, it was obvious that the main focus of the course was how to configure OSPF, EIGRP and BGP. And if the CCNA exam was any indication of how this exam would approach testing that knowledge, I knew preparing was going to be a tedious process.
Below I've broken down the major areas of study I undertook. You should also view the complete objectives linked to in the above sidebar.
First, the Basics
If you don't have much experience with TCP/IP and routing basics, or if you haven't kept those skills honed, you're going to have a tough time understanding the more advanced routing protocols tested on this exam. Therefore, make sure you have your TCP/IP fundamentals down before moving on to anything else.
Remember that by default, IP addresses are broken down into five classes:
Address Class |
Range of IP Addresses |
A |
1.0.0.0 - 126.255.255.255 |
B |
128.0.0.0 - 191.255.255.255 |
C |
192.0.0.0 - 223.255.255.255 |
D |
224.0.0.0 - 239.255.255.255 |
E |
40.0.0.0 - 255.255.255.255 |
The 0.0.0.0 network is reserved as a local network indicator, and the 127.0.0.0 is reserved as a local loopback address. In addition, several networks are reserved for private use, meaning that they can be used by any machine that isn't directly connected to the Internet. These addresses are 10.0.0.0 to 10.255.255.255, 172.16.0.0 to 172.31.255.255 and 192.168.0.0 to192.168.255.255. Because these addresses cannot be routed across the Internet, the use of network address translation(NAT) is required to provide connectivity.
When it comes to routing basics, it's important to remember that there's a big difference between routed protocols and routing protocols. A routed protocol is a network protocol, such as IP and IPX, that can be routed across networks as opposed to one that cannot be routed, such as NetBEUI. A routing protocol is the protocol that actually handles the exchange of routing information from one router to another, such as RIP, IGRP or OSPF.
Routing protocols can be divided into various categories, such as classful or classless, and distance-vector or link-state. A classful routing protocol is one that exchanges routing information based on the classful IP boundaries. A router running a classful routing protocol such as RIPv1 or IGRP that receives an update for a network to which it is not physically connected will summarize that network at the default classful boundary. If the router received a route to network 10.1.2.0, for instance, it would summarize the route to 10.0.0.0 in its routing table, because an 8-bit (255.0.0.0) subnet mask is the default classful boundary for Class A addresses. A classless routing protocol, such as RIPv2, OSPF or EIGRP, exchanges the subnet mask with the subnet information. So if the aforementioned router was running OSPF, it would receive a route to network 10.1.2.0 with a 24-bit (255.255.255.0) subnet mask included, and would not summarize the route to 10.0.0.0.
Routing protocols can be further categorized based on the way they handle routing updates. RIPv1, RIPv2 and IGRP are distance vector protocols, meaning they send a periodic broadcast (or in some cases multicast) that contains their full routing table. Link-state protocols, on the other hand, detect changes in network topology and generate updates based on that knowledge. Because of this behavior, link-state protocols offer faster convergence and less overhead than distance vector protocols. OSPF and IS-IS are examples of link-state routing protocols. EIGRP, although technically an advanced distance vector protocol, shares many features of link-state protocols.
The Protocol Troika
Once you have a firm grip on the basics, it's time to move on to the three routing protocols: Open Shortest-Path First (OSPF), EIGRP and Border Gateway Protocol (BGP). The vast majority of this exam focuses on how these three protocols work and how to configure them on Cisco devices. Cisco emphasizes each equally, so you'll need to know all of them fairly well.
For most Cisco exams, knowledge of the Cisco IOS and command syntax is pretty crucial. However, it's more vital for this exam than for any other I've taken. No matter what you know about these protocols, if you don't know how to configure them on a Cisco router, you're toast.
Tip: Physical access to a Cisco router will go a long way toward helping you remember the various commands each of these protocols use.
OSPF
OSPF is designed for large networks that make routing by distance vector protocols inefficient or impossible. In addition to supporting variable-length subnet masks (because OSPF is a link-state protocol), it provides faster convergence and more efficient use of bandwidth than RIP or IGRP. Whereas RIP supports only 15 hops in a route, OSPF has no such limitations, allowing the network to grow beyond what RIP could support. OSPF also has an advanced algorithm (the SPF algorithm) that bases path selection on cost rather than hop count, like RIP does. These advantages have made OSPF a very popular routing protocol for larger networks.
In order to add scalability to very large networks, OSPF supports the segregation of the network into separate areas. The separation of one large network into smaller areas (i.e., autonomous systems) is known as hierarchical routing. Hierarchical routing allows routing to continue between the various areas (inter-area routing), but confines many of the routing processes to those individual areas. This prevents routers in other areas from having to recalculate their routing tables unless a route in their own area changes. Summarization is used between areas to reduce the amount of routes that must be advertised from one area to another. The complicated process of design and configuration represents the major drawback of OSPF.
Tip: The routers used on an OSPF network may have various terms associated with them based on their location in the topology. An internal router is a router with all interfaces in a single area. A backbone router (BBR) is a router with at least one interface in area 0, which interconnects the other areas. An area border router (ABR) is a router with interfaces in more than one area inside the same autonomous system. An autonomous system border router (ASBR) is a router with at least one interface outside the autonomous system, or OSPF network.
EIGRP
Cisco's proprietary routing protocol, EIGRP, is described as an advanced distance vector routing protocol because it combines the simple configuration of traditional distance vector protocols with some of the advantages of link-state protocols, such as neighbor discovery and incremental updates. EIGRP also supports multiple protocols, such as IP, IPX and AppleTalk. EIGRP maintains three separate tables for each protocol that it's routing: a neighbor table, a topology table and a routing table, for a total of up to nine separate tables! EIGRP employs the same DUAL algorithm used in IGRP, which calculates the best path to a destination based on bandwidth and delay by default. The primary route, called a successor, is stored in the routing table along with any backup routes, called feasible successors. The inclusion of the backup routes in the routing table makes convergence very fast in cases of topology changes where a route is lost.
Tip: By default, EIGRP has support for variable-length subnet masks disabled. To enable it, you must use the 'no auto summary' command in router configuration mode.
BGP
BGP is a routing protocol designed to scale to massive networks, such as the Internet. It is typically used at the ISP level to exchange information with other ISPs, to which multiple links exist. If TCP/IP is the language of the Internet, BGP serves as its punctuation, organizing the Internet into coherent groups. BGP calls these groups autonomous systems (AS) and uses the AS numbers to direct information from one point to any destination around the world. A BGP autonomous system appears to other autonomous systems as a single entity, responsible for (and capable of) successful routing to any address it contains. Because it routes between multiple autonomous systems, BGP is an example of an exterior gateway protocol. All of the other routing protocols we've mentioned so far, RIP, OSPF, IGRP and EIGRP, are considered interior gateway protocols because they operate inside a single, autonomous system.
Tip: BGP running between routers inside a single autonomous system is referred to as IBGP, whereas BGP running between routers in different autonomous systems is referred to as EBGP.
BGP runs over TCP port 179, and uses several attributes to relay route and metric information to its peers, or neighboring BGP routers. In order to route between multiple autonomous systems, BGP uses an autonomous system path attribute, which is simply a list of autonomous systems that must be traversed in order to reach the intended destination. A next-hop attribute specifies, as one might guess, the IP address of the next hop in the path to the destination. BGP administration can be simplified by using peer groups or communities to avoid having to modify each router individually. The overhead associated with BGP updates can be reduced by using route reflectors, which are simply routers configured to relay BGP updates within an autonomous system. Due to the vast differences between BGP and interior gateway protocols, BGP may warrant some extra study time.
Don't Stop There…
Though you may be a master of routing protocols, you're not done studying yet. In order to make things work, you must first learn how to redistribute routes from one protocol to another (or from one autonomous system to another). You will also need to learn how to control routing updates with route filters to prevent unneeded updates from hogging that precious bandwidth. If that's not enough for you, you can work on policy-based routing, too. The sky, or rather your tolerance for tedious details, is the limit when it comes to studying for the Routing exam.
How Should I Prepare?
The Cisco Press book, Building Scalable Cisco Networks, was pretty much my sole resource in studying for this exam. Why? This book does a better job of covering the material on the exam than most. Nearly everything is there, in great (but sometimes agonizing) detail. Besides, by the time I finished reading this book, the notion of starting another book on routing was too painful to consider very seriously. The subject material was a bit on the dreary side, but I came away from the book with a wealth of knowledge that has already come in handy on the job.
My hands-on experience at work and in the labs also proved very useful in this exam, which as I mentioned earlier, stresses commands and syntaxes more than any other I've ever sat. The authors of Building Scalable Cisco Networks, Catherine Paquet and Diane Teare, have included a handy command list at the end of each chapter which is worth looking over at least once, in hopes that the syntaxes will be burned into your memory when it comes to testing time. Seriously, though, it's not that bad. If you have practical experience and do your homework, the test is fair.
On the Horizon
Now that I've knocked out the Switching and Routing exams, I've begun studying for the Remote Access exam, which I've put off as long as I can. Once I've managed to muster enough resolve to drudge through the book and sit the exam, I think I'll be 90 percent ready for the Troubleshooting exam. After that… well, after that I think I'll take a vacation.
This article reprinted with permission from TCPMag.com, November 2001.
Have you taken this exam? Rate it below!
Richard Shanks () is a freelance writer and network administrator with the Houston Advanced Research Center.
|