Networking

Talking Shop: Studying for the Network + exam

Passing the Network + exam requires a solid knowledge of TCP/IP


The Network+ exam provides an excellent opportunity for IT professionals to add networking expertise to their resumes. But don’t believe that you can earn CompTIA’s networking accreditation just because you passed Microsoft’s Networking Essentials exam or have been working in the industry for years. It’s a tough test of your skills.

Last week, I described the particulars of the exam itself. I also reviewed topics you should study to pass the networking practices portion of the exam. This week, I’ll review those networking technology subjects that are likely to compose three-fourths of your Net+ test.

Knowledge of networking technology topics
In a flashback to IT 101, you’re going to find many questions that test your understanding of the fundamentals. It’s best to brush up on all networking technologies, especially those you haven’t worked with for some time.

Network topologies and structure
Ensure that you’re familiar with different network topologies. Know the differences between star, bus, mesh, and ring topologies. Understand the strengths and weaknesses (including fault tolerance) of each. Be sure you understand the unique needs of each topology too.

Also, know which networking elements to check first when failures occur. For example, remember that coaxial cables require terminators to prevent disruptive signal bounces from occurring.

Memorize the difference between a network backbone and a segment. The backbone is the portion of the network to which all segments and servers connect, while a segment refers to any network section that branches out of the backbone.

Operating systems
Since this is a vendor-neutral exam, CompTIA will test your familiarity with Microsoft, Novell, and UNIX operating systems. While you’re not expected to prove complex configurations for each platform, you should understand basic administration procedures and interoperability capabilities for these operating systems. You should also be familiar with the directory services that each OS uses.

I wouldn’t worry about mastering Novell’s bindery or NDS or memorizing how to replicate Microsoft’s Active Directory across multiple domains. However, I do recommend that you learn what tools and processes are used to administer different directory services. You should also pay attention to key differences. Consider a Novell example. You should know that the bindery is a flat database, whereas NDS is a hierarchical database that stores objects. NDS is network-centric, not server-centric like the bindery. These are the kinds of details you should learn about directory services.

Protocols
Study TCP/IP but also spend some time reviewing IPX and NetBEUI functionality. Know how to use the appropriate troubleshooting tools for each. CompTIA lists the following on its Net+ Web page:
  • ARP
  • Telnet
  • NBTSTAT
  • TRACERT
  • NETSTAT
  • IPCONFIG
  • WINIPCFG
  • FTP
  • PING

Refresh your subnetting skills too. That’s a pretty tall order, I know. The protocol information by itself pretty much formed an entire Windows NT 4.0 MCP exam. TechRepublic has published numerous articles on these topics, which will give you more information:

Data management
You’ll also be tested on disk mirroring, striping, and volume management. Be sure you’re familiar with backup strategies.

While these questions may prove simple on the exam, don’t let them trip you up. Remember the following:
  • RAID 0 refers to disk striping without parity. In other words, performance is excellent but there’s no fault tolerance; lose one disk and you lose all your data. RAID 0 implementations are common.
  • RAID 1 refers to disk mirroring, in which one disk is a mirror of the other. Thus, it’s not particularly fast, but it’s fault tolerant. RAID 1 implementations are also pretty common.
  • RAID 2 is not commonly used, but you should know that it requires at least three disks. One disk does nothing but store parity information, so RAID 2 does provide fault tolerance.
  • RAID 3 is somewhat common. It, too, uses a parity drive. The difference is that RAID 2 data is striped in bits, whereas RAID 3 stripes data in bytes. As a result, more data is read and written per operation, thereby enhancing performance.
  • RAID 4 is RAID 0 with the added benefit of a parity drive. It's similar to RAID 2 and 3, but RAID 4 striping occurs in blocks. RAID 4 implementations are less common.
  • RAID 5 is a more popular implementation in which data is striped across multiple disks. The difference between it and RAID 0 is that RAID 5 also stripes parity across multiple disks. Performance is enhanced vs. traditional disk read and write operations, while fault tolerance is also provided.

You’ll want to be familiar with data backups. The most important thing to remember about incremental backups is that they flip a file’s archive bit. Differential backups do not perform that action. As a result, incremental backups are easy to perform. However, restoring differential backups requires even less effort.

Cabling
You’ll definitely receive questions testing your knowledge of cables. Know how far cables can be stretched before signal loss occurs. Study up on the specifications for twisted-pair, fiber-optic, ThickNet, and ThinNet cables, paying particular attention to the maximum speeds that each supports.

Know the limitations and strengths of each cable type. Remember that twisted-pair cables, of course, use RJ-45 connectors, while ThickNet (10Base5) uses vampire tap (DB-15 or DIX) connectors. ThickNet sometimes uses N-series connectors too. ThinNet (10Base2), meanwhile, uses BNC connectors.

The OSI model
Don’t try to take this exam without reviewing the OSI model. The importance of theory is often underestimated in troubleshooting network issues. Know which TCP/IP application protocols perform what actions at which layers of the OSI model.

You can read up on the OSI model in these TechRepublic articles:

Dial-up connections
If you’re like me, you haven’t configured a dial-up connection for ages. You spend most of your time setting up either DSL, cable, or LAN links. However, before I take the Net+ exam, I’m going to reinstall my U.S. Robotics modem—if I can find it in my computer parts closet—because the Net+ exam requires knowledge of dial-up networking configuration.

Along those lines, you’ll want to keep in mind which IRQ a system’s COM ports use, along with the corresponding I/O address. Use Table A to keep these settings straight.

Table A
Port IRQ I/O Address
COM1 4 3F8
COM2 3 2F8
COM3 4 3E8
COM4 3 2E8
Modem port, IRQ, and I/O address specifications

Remote connectivity
Be prepared for remote access connection and protocol questions. Commit the following remote connectivity protocols, and their characteristics, to memory:
  • Serial Line Internet Protocol (SLIP) is used often with UNIX networks, but it does not support encryption. Instead, it passes passwords as clear text.
  • Point-to-Point Protocol (PPP) performs error correction, which SLIP does not.
  • Point-to-Point Tunneling Protocol (PPTP) works with PPP to create a secure connection from a remote system to a server or to a LAN.

PPTP, while not compatible with all server platforms, offers an inexpensive method for linking remote systems to a LAN or a WAN over the Internet. Because a secure tunnel is created, long-distance dial-up connections or expensive data circuits aren’t required. Connections can be created through the Internet using simple, local dial-up connections.

You should also review the strengths and weaknesses of Integrated Services Digital Network (ISDN) lines vs. using the Public Switched Telephone Network (PSTN). Be sure that you can keep T-1s and T-3s straight too. T-1s offer 1.544-Mbps digital connections, while T-3s offer 44.763-Mbps digital links.

Security
I expect that as the Network+ exam is updated, increased emphasis will be placed upon security. Right now, only some 6 percent of the exam focuses on security.

What security topics should you master? Familiarize yourself with recommended password practices. Enable automatic lockouts. Administrator accounts should be renamed, and unused accounts should be disabled. Passwords shouldn’t be easy to guess. So passwords shouldn’t be users’ pets' names, spouses’ names, or otherwise easily identifiable passwords. For added protection, passwords should be alphanumeric.

While you needn’t be a veteran firewall administrator to earn Network+ accreditation, you should understand access control lists, packet-level filtering, application-level filtering, and the functioning of proxy servers.

You should also familiarize yourself with common hacker practices. Even if you don’t plan to take the Net+ exam, you should know how IP spoofing works and the characteristics of different denial of service attacks.

Get up to speed on encryption and the use of private keys. Know how to conduct a security audit and implement a security policy too.

Eckel's take
CompTIA’s Network+ exam covers a vast territory. In some ways, it combines Microsoft’s old Networking Essentials and TCP/IP exams into one test. While you’ll have to set time aside to prepare for the test, this is one accreditation that I believe most support professionals and network administrators should earn. It’s vendor neutral, it tests the basics that all IT professionals should know, and it doesn’t expire.

The number of topics you’ll need to review is, of course, substantial. But it’s reasonable. Rather than set out on a wild goose chase for appropriate study materials, wait until next week. I’ll examine several Network+ study guides and let you know which ones that I think offer you the best shot of earning the certification.

How are you preparing for the Network+ exam?
We look forward to getting your input and hearing about your experiences regarding this topic. Join the discussion below or send the editor an e-mail.

 

Editor's Picks