Microsoft TCP/IP 4.0 Practice Test #3
70-059

 

1. Carl administers a TCP/IP network with four subnets: Subnet1, Subnet2, Subnet3 and Subnet4. He wants to enable the users of every Windows-based computer to browse every other Windows-based computer on all four subnets. The primary domain controller (PDC) is on Subnet1. A backup domain controller (BDC) is on Subnet3. How should Carl set up the network?

a. by installing one BDC on Subnet2 and one BDC on Subnet4

b. by creating a HOSTS file on every Windows-based computer

c. by setting up a DHCP server and enabling DHCP on each Windows-based computer

d. by setting up a WINS server and enabling WINS on each Windows-based computer

 

Choice d is correct. Windows-based computers must know the names of the other Windows-based computers on their network in order to browse and perform peer-to-peer connections. Either an LMHOSTS file or a WINS server can be used to resolve NetBIOS names to their IP addresses. Hostnames require HOSTS files or DNS for resolution. The Dynamic Host Configuration Protocol (DHCP) is used to assign IP addresses to Windows-based computers. DHCP cannot resolve names to IP addresses. Simply installing a domain controller on each segment would require the routers to be configured to forward NetBIOS broadcasts (b-node). This is not recommended because NetBIOS traffic would propagate through the entire network consuming additional bandwidth. RKNG, pp. 149-150, Chp. 3, Browse Service Across a WAN with TCP/IP; TPWIN2, pp. 328-331, Chp. 11, Installing the WINS Server Service; SPT, pp. 187-188, Chp. 9, Lesson 1: WINS Overview.

 

 

2. Richard administers a large network with several thousand users. The network uses DHCP servers, WINS servers and LMHOSTS files for its Windows-based clients and DNS servers and HOSTS files for its UNIX clients and servers. While working at his Windows NT Server computer, Richard receives an error message: "Event 4320, NetBT Error." What is the first step Richard should take in troubleshooting this problem?

a. Purge the LMHOSTS file on his server.

b. Reboot the WINS server.

c. Run nbtstat -n on his server.

d. Run nslookup on the DNS server.

 

Choice c is correct. "Event 4320, NetBT Error" is a name release message sent to your computer indicating that a duplicate name on the network may have been detected. The nbtstat -n command will display the name of the computer with the duplicate IP address. Running the nbtstat -A command against the IP address will reveal the sender's computer name. Only when the offending computer's name is known can further steps be taken to resolve the conflict. RKNG, p. 831, App. G, NetBIOS Names; TPWIN2, pp. 444-445, Chp. 14, Troubleshooting Utilities; "Diagnosing Event ID 4320" on TechNet.

 

 

3. Beth is designing a TCP/IP network with two subnets: RED and WHITE. She configures a Windows NT Server computer as a static router between the two subnets. She installs a second Windows NT Server computer as a DHCP server on subnet RED and a third Windows NT Server computer on subnet WHITE. Beth wants client computers on subnet WHITE to obtain their IP configurations from the DHCP server on RED. Which service must Beth install on the third Windows NT Server computer on subnet WHITE?

a. the LLC sublayer

b. the Proxy service

c. the RAS service

d. the DHCP Relay Agent service

e. the RIP for IP service

 

Choice d is correct. DHCP services are normally restricted to the physical segment (subnet) on which the DHCP server is installed. For DHCP to provide services to other subnets on a network, the DHCP Relay Agent service must be installed on each of those subnets. Otherwise, the DHCP services must be installed on each subnet. RKNG, pp. 420-421, Chp. 7, Planning for DHCP Implementation; TPWIN2, pp. 274-275, Chp. 10, DHCP Concepts and Operation.

 

 

4. The InterNIC has assigned a class C IP address to Stan's company. Looking at the physical requirements of his TCP/IP network, Stan realizes he must create five subnets. If each subnet will need up to 25 host IDs, then which subnet mask should Stan choose for his network?

a. 255.255.255.224

b. 255.255.255.240

c. 255.255.255.248

d. 255.255.255.254

 

Choice a is correct. The subnet mask 255.255.255.224 creates six subnets with 30 host IDs each. To understand subnet masking, you must convert an IP address to its binary equivalent. For a class C address, the right-most octet (often called the z octet) determines how many subnets and host IDs per subnet can be created. In the case of 255.255.255.224, the number 224 equals 11100000 in binary. The three leftmost (highest ordered) digits reserve the subnet ID. The remaining five digits are left for the host ID. A three-digit subnet ID offers eight combinations: 000, 001, 010, 011, 100, 101, 110, and 111. The combinations 000 and 111 cannot be used, since they are reserved. This leaves six valid subnets beginning at 00100001 (33). All zeros and all ones are illegal in the host ID range segment as well, so each valid 5-digit host ID range must run between 00001 and 11110.

This leaves the six subnets of 33-62, 65-94, 97-126, 129-158, 161-190 and 193-222, for a total of 180 possible host IDs. RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, pp. 90-106, Chp. 4, IP Addressing; LTM, Subnets and Subnet Masking.

 

 

5. Which utility allows you to view a list of documents sent to a UNIX-based TCP/IP print server?

a. ftp

b. lpq

c. lpr

d. telnet

 

Choice b is correct. Assuming the Microsoft TCP/IP Printing service (LPD) is installed on your network, you can check the status of a UNIX-based print server's queue by issuing the lpq command utility at the command prompt. The lpq command has three parameters: -S specifies the host to which the UNIX-based print server is attached, -P specifies the printer name, and the optional -l parameter generates a detailed status of the queue. RKNG, p. 687, App. A, lpq; SPT, pp. 317-318, Chp. 14, Using LPR and LPQ.

 

 

6. You administer a large but stable network using LMHOSTS files instead of a WINS server. When a user complains about the length of time it takes to connect to remote hosts, you examine her LMHOSTS file and notice a large number of #PRE and #DOM entries in the file. What step can you take to speed up the reading of her LMHOSTS file?

a. delete the #DOM tag from every #PRE entry

b. delete the #PRE tag from every #DOM entry

c. move all of the #PRE and #DOM entries to the bottom of the file

d. move all of the #PRE and #DOM entries to the top of the file

 

Choice c is correct. During IP initialization, all LMHOSTS file entries with the #PRE statement are loaded into the name cache. When the host needs to resolve a name, it first examines the name cache, then it sends out a b-node broadcast, and finally it searches the LMHOSTS file. For this reason, it is always best to put #PRE entries at the end of the LMHOSTS file, since they load into the NetBIOS name cache at startup. If LMHOSTS files do grow to such a size that network performance is affected, then using a WINS server should be considered. RKNG, p. 515, Chp. 10, Adding Remote System Names by Using #PRE; TPWIN2, pp. 357-358, Chp. 11, Managing LMHOSTS Files; SPT, pp. 179-180, Chp. 8, Lesson 3: Using the LMHOSTS File; "How to Troubleshoot TCP/IP Connectivity with Windows NT" on Technet.

 

 

7. You manage a large network of 900 Windows-based computers on four subnets. The sales staff of 300 users has been issued laptops running Windows 95. When in the office, these users may connect to the network from any one of the subnets. You want them to be automatically assigned an IP address when they log on. Which service will do this?

a. DHCP

b. DNS

c. LPC

d. RIP

e. WINS

 

Choice a is correct. The Dynamic Host Configuration Protocol (DHCP) goes a long way in easing the administrative burden of assigning IP addresses to TCP/IP network hosts. Through the DHCP Manager, you can automatically and dynamically assign IP addresses to each network adapter card installed on a network. Valid IP addresses can be grouped by subnet into scopes, which can be further defined by a lease duration for the IP address of each scope. DHCP can be centrally managed across a network comprised of multiple subnets so long as BOOTP routing is enabled. RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, pp. 274-278, Chp. 10, DHCP Concepts and Operations.

 

 

8. Which file does a DNS root server use to connect to the Internet?

a. cache.dns

b. domain.dns

c. HOSTS

d. place.dom

e. reverse-netid.in-addr.arpa.dns

 

Choice a is correct. The cache.dns file holds the IP addresses of the nine domain name servers that provide root-level name resolution in the United States. Located in the \<systemroot>\System32\DNS directory, the cache.dns file is automatically installed on a Microsoft DNS server. If you do not plan to use your Microsoft DNS server to connect to the Internet, then the cache.dns file should be deleted and replaced with a file containing the NS and A records used by your intranet root name servers. RKNG, p. 483, Chp. 9, Microsoft DNS Server Database; TPWIN2, pp. 380-383, Chp. 12, Managing Microsoft DNS Server; SPT, p. 266, Chp. 12, The Cache File.

 

 

9. You administer a small Windows-based network. You set up a RAS server as a router to connect to an Internet Service Provider (ISP) using a dial-up connection. How should you set the IP settings on the RAS server?

a. Install RIP.

b. Leave the default gateway address blank.

c. Use the IP address provided by the ISP.

d. Use the subnet mask provided by the ISP.

 

Choice b is correct. When using an ISP to connect to the Internet, the default gateway address on the RAS server should be left blank since it is provided by the ISP. The IP address and other IP settings, such as the subnet mask on the RAS server, should conform to the configuration requirements of the local network. Because network clients connect to the ISP through the RAS server, it is important that the RAS server does not send its own IP address as the data packet source. To prevent this, you must add the path \HKEY_LOCAL_MACHINE\System \CurrentControlSet\Services \RasArp\Parameters to the server's registry and set it to 0. NTNS, p. 67, Chp. 4, IP Routing; "Configuring Dial-up Networking for ISP DNS" on TechNet.

 

 

10. A Windows NT Server computer named BROWNS resides on a remote subnet. From his Windows NT Workstation computer CUBS, Peter cannot ping BROWNS using its IP address. He can ping all other computers on his subnet as well as computers on subnets other than the BROWNS computer's subnet. The subnet on which CUBS is located has only one router. What are possible causes of the problem?

a. BROWNS is set up with an incorrect default gateway address.

b. BROWNS is set up with an incorrect subnet mask.

c. CUBS is set up with an incorrect default gateway address.

d. CUBS is set up with an incorrect WINS server address.

 

Only choices a and b are correct. In this example, the remote server BROWNS may itself have an improperly configured gateway address or subnet mask. Trying to ping BROWNS from another computer is the quickest way of eliminating that possibility. If the problem is not with BROWNS and other users are not reporting similar problems, then CUBS is the most likely candidate. Since CUBS can ping other remote hosts successfully, its default gateway address is set up correctly. Pinging an IP address does not utilize WINS. WINS is used to resolve a NetBIOS name to an IP address. RKNG, pp. 347-349, Chp. 6, Address Resolution Protocol; TPWIN2, pp. 110-112, Chp. 4, IP Routing; "TCP/IP Name Resolution" on TechNet.

 

 

11. Kathy wants to manually add an entry to the WINS database for a Windows NT Workstation computer that has three network adapters. The workstation is configured to use WINS. What type of entry needs to be added?

a. Domain

b. Group

c. Internet group

d. Multihomed

e. Unique

 

Choice d is correct. Multihomed entries are reserved for computers that have more than one IP address. These addresses can represent multiple network adapter cards, each with its own IP address, or a single network adapter card with multiple IP addresses. You can register up to 25 IP addresses to each multihomed group. A multihomed computer can be a WINS client on multiple networks, with each network adapter card on a separate network registered to a different WINS server. However, if a multihomed computer is a host on more than one network, you should not configure the computer as a WINS server. RKNG, pp. 464-465, Chp. 8, Managing Static NetBIOS Name-to-IP-Address Mappings; TPWIN2, p. 340, Chp. 11, Managing WINS Servers; SPT, pp. 198-200, Chp. 9, Configuring Static Entries for Non-WINS Clients; "MS Windows NT Server 4.0 - WINS Architecture and Capacity Planning" on TechNet.

 

 

12. Karen has five multihomed Windows NT Server computers functioning as routers on her network. She wants to configure the routing tables on these servers with the least amount of administrative effort possible. How should Karen proceed?

a. by installing DHCP Relay Agent

b. by installing RIP for IP

c. by running netstat.exe

d. by running route.exe

 

Choice b is correct. The Multi-Protocol Router (MPR) for Windows NT 4.0 contains the Routing Information Protocol (RIP) for TCP/IP, RIP for IPX and the Boot Protocol (BOOTP) Relay Agent for DHCP. Once installed, you only need to enable IP Routing on the Advanced TCP/IP tab. Typing ROUTE PRINT at the command prompt displays the current routing table. You stop and start the RIP for IP service using Control Panel. RKNG, pp. 351-352, Chp. 6, Internet Protocol; TPWIN2, p. 118, Chp. 4, IP Routing; SPT, pp. 122-123, Chp. 6, RIP; "Multi-Protocol Router Installation and Configuration" on TechNet.

 

 

13. After installing Windows NT Workstation on his computer, Jake wants to try making a connection to a remote Windows NT Server computer using TCP/IP. Which command should Jake use?

a. arp

b. nbtstat

c. netstat

d. ping

e. route

 

Choice d is correct. The ping command is the first command you should use when testing TCP/IP connections. To test whether TCP/IP has been properly installed and configured on your computer, ping the loopback address 127.0.0.1. To test TCP/IP connectivity on your network, try pinging to progressively remote connections. For example, ping another computer on your subnet, then your default gateway, and finally an address on a remote subnet. If pinging works in every case, then any future problems you encounter will most likely involve name resolution. RKNG, pp. 548-550, Chp. 12, Test Connection to the TCP/IP Network by Using Ping, pp. 696-697, App. A, ping; TPWIN2, pp. 235-237, Chp. 8, Testing the TCP/IP Configuration.

 

 

14. A Windows 95 user complains that he cannot connect to a computer on a remote subnet by using its computer name. Other users on his subnet do not report having any trouble. The network has a WINS server, but not a DHCP server. The computer does not use an LMHOSTS file. Seated at his computer, you ping 127.0.0.1 and receive a reply. You successfully ping the server on the remote subnet. What is the most probable cause of the problem?

a. TCP/IP has not been installed.

b. The default gateway address on the Windows 95 computer is incorrect.

c. The subnet mask on the Windows 95 computer is incorrect.

d. The Windows 95 computer is not WINS-enabled.

 

Choice d is correct. Unless you use DHCP on a network, you must manually set up each client for WINS. The ability to ping an address but not connect to it using its computer name indicates a problem with computer name resolution. Since an LMHOSTS file is not used, the computer must not be correctly configured for WINS. RKNG, p. 442, Chp. 8, Microsoft WINS Clients; TPWIN2, p. 324, Chp. 11, Installing the WINS Server Service.

 

 

15. You are assigned the task of setting up a TCP/IP network. The network will be comprised of eight subnets. You want to place all of the Windows NT Server computers on one subnet. Two thousand Windows NT Workstation computers will be distributed across the other seven subnets. You want users to be able to open Network Neighborhood on their workstations and browse resources on the network servers. Which of the following services must you install?

a. a DHCP server

b. a WINS server

c. six BOOTP Relay Agents

d. six WINS proxies

 

Choice b is correct. The Windows Internet Name Service (WINS) provides a dynamic, distributed database that resolves NetBIOS names to IP addresses over LANs or WANs. Whether to manage WINS from a single server or multiple servers on a network will be determined by the number of network hosts and subnets, existing network traffic, and available bandwidth. Where multiple WINS servers are installed, you can configure replication to occur at given intervals or based on the number of changes to the WINS database. Where WINS spans multiple subnets, WINS proxies must be installed on each subnet that contain non-WINS-enabled computers and that lack a local WINS server. A WINS proxy listens for name query broadcasts (b-node) sent by non-WINS-enabled hosts. These broadcasts are then passed on to the WINS server for resolution. Upon resolving the name, the WINS server sends a response back to the proxy which sends the response back to the requesting host.

RKNG, pp. 438-439, Chp. 8, Introduction to Windows Internet Name Service, and pp. 448-449, Microsoft WINS Proxy; TPWIN2, pp. 311-313, Chp. 11, Architecture of the Windows Internet Name Service, and pp. 324-325, Installing the WINS Server Service.

 

16. A DHCP server provides the IP addresses of your network's two WINS servers to all client computers. You want these client computers to use WINS for name resolution before they use broadcast mode. Which mode needs to be specified?

a. b-node

b. h-node

c. m-node

d. p-node

 

Choice b is correct. The default mode of WINS clients is h-node. First, h-node directs a p-node request to the WINS server for name resolution. If this fails, it makes a b-node broadcast on the local network to achieve name resolution. The reverse of h-node is m-node. It first tries b-node name resolution, then p-node. P-node is used by the WINS client to communicate directly with the WINS server. If this method fails, then WINS name resolution fails. RKNG, pp. 376-377, Chp. 6, NetBIOS Interface; TPWIN2 pp. 308-311, Chp. 11, Resolving Names on Microsoft Networks; SPT, pp. 176-177, Chp. 8, NetBIOS over TCP/IP Name Resolution Nodes.

 

 

17. Which of the following protocols does Point-to-Point Tunneling Protocol (PPTP) support?

a. AppleTalk

b. IP

c. IPX

d. NetBEUI

 

Only choices b, c and d are correct. PPTP enables you to create virtual, secure (encrypted) networks over TCP/IP-based LANs, WANs, dial-up lines and the Internet. However, PPTP is not restricted to just TCP/IP. You can also encapsulate IPX and NetBEUI in the IP protocol. RKNG, pp. 398-399, Chp. 6, Microsoft Remote Access Service; TPWIN2, p. 520, Chp. 15, Using the Point-to-Point Tunneling Protocol.

 

 

18. Michelle moves a Windows 95 computer originally configured as a DHCP client from one subnet to another on a DHCP network. Although the computer can connect to the local Windows NT Server computer on its new subnet, it fails to connect to the remote Windows NT Server computer on its old subnet. What is a possible cause of the problem?

a. The default gateway address on the Windows 95 computer has been manually set.

b. The default gateway address on the Windows 95 computer is set as a local-level option on the DHCP server.

c. The IP address on the Windows 95 computer has been manually set.

d. The IP address on the Windows 95 computer is set as a scope-level option on the DHCP server.

 

Choice a is correct. Options set at client level override those set at the scope and global levels. Options set at the scope level override those set at the global level. Client level options are only available to reserved DHCP clients. You can also change a default DHCP option manually at the client. If the computer had not been able to connect to a computer on the local subnet, then its IP address might also have been a cause of the problem. RKNG, pp. 408-413, Chp. 7, Microsoft DHCP Server; TPWIN2, pp. 294-298, Chp. 10, DHCP Configuration Options.

 

 

19. Which Windows NT service automatically updates its IP routing tables using the routing information from other network routers?

a. DHCP

b. DHCP Relay Agent

c. RIP for IP

d. RIP for IPX

 

Choice c is correct. Although both the Routing Information Protocol (RIP) for IP and RIP for IPX can automatically update their routing tables based on routing information gathered from connecting routers, RIP for IP is used on a TCP/IP network; RIP for IPX is used on a Novell IPX/SPX network. The DHCP Relay Agent, which is included in the Multi-Protocol Router (MPR) for Windows NT 4.0, allows DHCP servers to manage remote subnets. RKNG, pp. 351-352, Chp. 6, Internet Protocol; TPWIN2, p. 118, Chp. 4, IP Routing; "Multi-Protocol Router Installation and Configuration" on TechNet.

 

 

20. You want to minimize the number of static records on your network's DNS server. How can you make the non-Microsoft TCP/IP clients use WINS to resolve the hostname portion of each Fully Qualified Domain Name (FQDN)?

a. Enable the WINS server to resolve Fully Qualified Domain Names.

b. Enable the DNS server to use the WINS server for name resolution.

c. Under name resolution on the client computer, list DNS first, then WINS.

d. Under name resolution on the client computer, list WINS first, then DNS.

 

Choice b is correct. You can closely integrate Microsoft DNS and WINS servers to reduce the number of static records on the DNS server. This can be accomplished by checking the Use WINS Resolution box on the WINS Lookup tab of the Zone Properties page of the DNS server and by adding the IP address(es) of the appropriate WINS server(s). Working in tandem, the DNS server can resolve the FQDN down to the hostname, which it then passes on to the WINS server. Assuming the hostname is the same as the NetBIOS name, WINS resolves the NetBIOS name to its IP address and returns the address to the DNS server. RKNG, pp. 490-492, Chp. 9, Planning for Microsoft DNS and Microsoft WINS Integration; TPWIN2, pp. 398-399, Chp. 12, Managing the DNS Server.

 

 

21. What should Jason use to see a list of all current TCP/IP connections to his Windows NT Server computer?

a. arp

b. tracert

c. netstat

d. route

 

Choice c is correct. The netstat -r command displays both the routing table and current connections. Netstat may also be used to report protocol statistics. RKNG, p. 552, Chp. 12, Troubleshooting NetBIOS Name Resolution by Using NBTStat, p. 688, App. A, nbtstat, and p. 690, netstat; TPWIN2, p. 444, Chp. 14, Troubleshooting Utilities.

 

 

22. Andy administers a TCP/IP network with two subnets, as shown in the Exhibit. The subnet mask for the network is 255.255.240.0. A user on the Windows NT Workstation computer BONDS on subnet GREEN complains that he cannot connect to the Windows NT Server computer STOCKS on subnet YELLOW. No other user on subnet GREEN has a problem connecting to STOCKS. Andy runs the ipconfig utility on BONDS and receives the following readout:

IP address. . . : 154.44.35.10

Subnet mask . . : 255.255.240.0

Default gateway : 154.44.50.2

Why does the workstation BONDS fail to connect to the server on YELLOW?

a. The default gateway address on STOCKS is incorrect.

b. The default gateway address on BONDS is incorrect.

c. The IP address on STOCKS is incorrect.

d. The IP address on BONDS is incorrect.

e. The subnet mask on BONDS is incorrect.

 

Choice d is correct. A host's IP address must fall within the range allowed on the subnet by the subnet mask. The mask 255.255.240.0 on a class B address creates 14 subnets in blocks of 16. Thus, the IP address for BONDS is not in the same subnet as the default gateway. One of the IP addresses is incorrect. However, because the other hosts on the same subnet have no problem connecting to the remote server and their IP addresses fall into the same subnet as that of the default gateway, it must be the IP address of BONDS that is incorrect. RKNG, pp. 547-549, Chp. 12, Identify the TCP/IP Configuration by Using IPConfig; TPWIN2, pp. 246-247, Chp. 9, Rules of Routing.

 

 

23. Betty successfully administers a small, stable TCP/IP network of Windows-based computers on three subnets using LMHOSTS files. As her company grows, her network is scheduled to expand to over 1,800 users on 12 subnets. Many users have laptop computers that they want to connect to the network from any subnet. Computer name resolution using LMHOSTS files is becoming unmanageable. What service should Betty install to provide for name resolution?

a. DHCP

b. DNS

c. HOSTS

d. WINS

 

Choice d is correct. WINS is the logical step up from LMHOSTS files. WINS resolves NetBIOS names to IP addresses; DNS resolves hostnames to IP addresses. WINS name registration is dynamic; DNS name registration is static. WINS supports DHCP, DNS does not. For all of these reasons, Windows-based networks should implement WINS when LMHOSTS files become unmanageable. For Internet connections, a Microsoft DNS server can be seamlessly integrated into a WINS-enabled network. RKNG, pp. 437-439, Chp. 8, Managing Microsoft WINS Servers; TPWIN2, pp. 307-308, Chp. 11, Managing WINS.

 

 

24. Wilma installs the SNMP service on a Windows NT Server computer. She wants to prevent any unauthorized SNMP management consoles from managing this server. What should Wilma do?

a. Assign a password to the SNMP community name.

b. Designate the server to be a trap destination.

c. Enable CHAP encryption on the MIB file packets.

d. Enable the Only Accept SNMP Packets From These Hosts option on the server.

 

Choice d is correct. In SNMP, the community name that the SNMP management console shares with its SNMP hosts functions as a kind of password. For any interaction to occur between SNMP hosts, they must all belong to the same community. However, the community name cannot be protected with a password. A second level of security is provided with the Only Accept SNMP Packets From These Hosts option on the Security tab of the Microsoft SNMP Properties page on the server. Here you can enter specific host IP or IPX addresses (in this example, that of the SNMP management console) from which SNMP packets will be accepted. All other packets will be rejected. Because SNMP management programs must initiate data gathering operations with an SNMP packet, enabling the Only Accept SNMP Packets From These Hosts option will preclude access by unauthorized SNMP management consoles.

RKNG, pp. 531-535, Chp. 11, Windows NT-based Implementation of SNMP; TPWIN2, pp. 445-448, Chp. 14, Activating SNMP on Microsoft TCP/IP Hosts; SPT, pp. 333-340, Chp. 15, Lesson 3: Installing and Configuring the SNMP Service.

 

 

25. When manually adding static mapping entries to the WINS database, which type of computer should you identify as Unique?

a. a multihomed Windows NT Server computer

b. a Network Monitor agent

c. a print server

d. a UNIX computer

 

Choice d is correct. A Unique entry maps a computer name to one IP address. Marking a UNIX-based computer as a Unique static entry ensures that the computer's IP address will be available to WINS-enabled network clients. WINS has five static entry types: Domain Name, Group, Internet group, Multihomed and Unique. Multihomed entries are reserved for computers that have more than one IP address. The Internet group is any user-defined resource group, such as printers, RAS servers, and Network Monitor agents. RKNG, pp. 464-465, Chp. 8, Managing Static NetBIOS Name-to-IP-Address Mappings; TPWIN2, p. 340, Chp. 11, Managing WINS Servers; "MS Windows NT Server 4.0 - WINS Architecture and Capacity Planning" on TechNet.

 

 

26. Fran uses a Windows NT Workstation computer on a TCP/IP network with several subnets. Running the command ftp 144.104.3.4, she connects to a Windows NT Server computer on a remote subnet. However, when Fran specifies the same server's UNC path in Windows NT Explorer, she cannot connect. What is the most likely cause of the problem?

a. The server is not set up to use a WINS server.

b. The server's default gateway address is incorrect.

c. The workstation is denied access to the FTP server's directories.

d. The workstation is not set up to use a DNS server.

e. The workstation's default gateway address is incorrect.

 

Choice a is correct. The workstation can successfully connect to the remote server. This indicates that the default gateway addresses on both the server and the workstation are correct. It also demonstrates successful access to the FTP server's directories. The Windows browsing service uses LMHOSTS files or WINS to resolve NetBIOS names to their IP addresses. DNS is used to resolve hostnames to their IP addresses. RKNG, pp. 149-150, Chp. 3, Browse Service Across a WAN with TCP/IP; TPWIN2, pp. 328-331, Chp. 11, Installing the WINS Server Service.

 

 

27. Suppose the following situation exists:

Amanda's company is headquartered in Miami with a second office in Chicago. Both offices use TCP/IP as the networking protocol. The Miami office has ten Windows NT Server computers and 1,000 Windows NT Workstation computer clients. The Chicago office has two Windows NT Server computers and 225 Windows NT Workstation computer clients. The WINS server in Miami is called MIA; the WINS server in Chicago is called CHI. Amanda wants to set up WINS database replication between the two WINS servers.

Required result:

Optional desired results:

Proposed solution:

Which results does the proposed solution produce?

a. The proposed solution produces the required result and both of the optional desired results.

b. The proposed solution produces the required result and one of the optional desired results.

c. The proposed solution produces the required result and none of the optional desired results.

d. The proposed solution does not produce the required result.

 

Choice b is correct. The proposed solution produces the required result, but only one of the optional desired results. As a push partner, Miami will notify its pull partner, Chicago, that 20 changes have occurred to Miami's database. Chicago will dutifully send a replication request. Miami will then replicate all 20 of its changes to Chicago. Chicago will also replicate Miami's database changes once every 24 hours regardless if updates have occurred to Miami's WINS database or not. The proposed solution does not satisfy the optional desired result that the Chicago WINS database is replicated to Miami. The Replicate Only With Partners field in the Advanced tab of the WINS Server Configuration dialog box allows the administrator to replicate the database (either as push or pull) to a WINS server partner not listed in the dialog box.

The default is for the option to be checked. RKNG, pp. 221-226, Chp. 5, Installing and Configuring WINS, and pp. 452-458, Chp. 8, Planning for Microsoft WINS Server Implementation; TPWIN2, pp. 341-345, Chp. 11, Managing WINS Servers; SPT, pp. 208-213, Chp. 9, Lesson 4: Database Replication Between WINS Servers.

 

 

28. Suppose the following situation exists:

Amanda's company is headquartered in Miami with a second office in Chicago. Both offices use TCP/IP as the networking protocol. The Miami office has ten Windows NT Server computers and 1,000 Windows NT Workstation computer clients. The Chicago office has two Windows NT Server computers and 225 Windows NT Workstation computer clients. The WINS server in Miami is called MIA; the WINS server in Chicago is called CHI. Amanda wants to set up WINS database replication between the two WINS servers.

Required result:

Optional desired results:

Proposed solution:

Which results does the proposed solution produce?

a. The proposed solution produces the required result and both of the optional desired results.

b. The proposed solution produces the required result and one of the optional desired results.

c. The proposed solution produces the required result and none of the optional desired results.

d. The proposed solution does not produce the required result.

 

Choice a is correct. The proposed solution ensures that two-way replication between Miami and Chicago will occur every 24 hours. In this configuration, each site is both a push and a pull partner. As mutual pull partners, each site will request updates from the other site's WINS database every 24 hours. As mutual push partners, each site will notify the other site when its update count (1000 for Chicago, 200 for Miami) has been reached and its database needs to be replicated. Notice that in this scenario the update count roughly reflects the number of WINS clients at each site. In a real situation, only real data will indicate what the proper update count should be. On one hand, it should not be so low that the data link would be inundated with low-priority data traffic. This is particularly true if the data link is slow and WINS-based name resolution is seldom required between the sites.

On the other hand, the update count should not be set so high as to be meaningless. This is particularly true when the sites are also configured as pull partners. RKNG, pp. 221-226, Chp. 5, Installing and Configuring WINS, and pp. 452-458, Chp. 8, Planning for Microsoft WINS Server Implementation; TPWIN2, pp. 341-345, Chp. 11, Managing WINS Servers.

 

 

29. . Suppose the following situation exists:

Amanda's company is headquartered in Miami with a second office in Chicago. Both offices use TCP/IP as the networking protocol. The Miami office has ten Windows NT Server computers and 1,000 Windows NT Workstation computer clients. The Chicago office has two Windows NT Server computers and 225 Windows NT Workstation computer clients. The WINS server in Miami is called MIA; the WINS server in Chicago is called CHI. Amanda wants to set up WINS database replication between the two WINS servers.

Required result:

Optional desired results:

Proposed solution:

Which results does the proposed solution produce?

a. The proposed solution produces the required result and both of the optional desired results.

b. The proposed solution produces the required result and one of the optional desired results.

c. The proposed solution produces the required result and none of the optional desired results.

d. The proposed solution does not produce the required result.

 

Choice a is correct. The proposed solution allows for two-way WINS database replication between Miami and Chicago within a 24-hour period. The solution does not provide for either site to be a push partner to the other. Often on a WAN, where immediate database updates might not be necessary, push partnerships between sites may not be advisable. The Replicate Only With Partners field in the Advanced tab of the WINS Server Configuration dialog box allows the administrator to replicate the database (either as push or pull) to a WINS server partner not listed in the dialog box. The default is for the option to be checked. RKNG, pp. 221-226, Chp. 5, Installing and Configuring WINS, and pp. 452-458, Chp. 8, Planning for Microsoft WINS Server Implementation; TPWIN2, pp. 341-345, Chp. 11, Managing WINS Servers.

 

 

30. Windows NT Workstation computer users need to send documents to a print device connected to a UNIX computer. You want all documents to be sent to and managed by a Windows NT Server computer. You install the LPD service on the UNIX computer. What other steps must you take?

a. Create a printer share on the server.

b. Install the TCP/IP Printing service on the server.

c. Map a logical printer port to the UNIX printer on the Windows NT Server.

d. Install the LPR utility on the workstations.

 

Only choices a, b and c are correct. The Microsoft TCP/IP Printing service (LPD) on a Windows NT computer allows clients to print to any printer attached directly to the network or to a UNIX server. The clients do not need to run TCP/IP. The clients must share only a common protocol with the computer running the LPD software. The LPD printer must be shared on the network like any other network printer. When configuring the LPR clients, use the actual printer name as shown in the Printers folder, not the printer share name as shown on the Sharing tab of the printer's Properties page. An LPR port that points to the printer on the UNIX machine needs to be added using the Add LPR Printer dialog box on the Windows NT Server computer. NTCP, pp. 174-175, Chp. 5, Configuring TCP/IP and UNIX Printers.

 

 

31. Tom wants to gather TCP/IP statistics from his Windows NT Server computer and save them to a log file for later analysis in a spreadsheet. What should Tom use?

a. nbtstat

b. netstat

c. Network Monitor

d. Performance Monitor

e. SNMP

 

Choice d is correct. You should use Performance Monitor to gather TCP/IP statistics that you intend to export to other applications. Once you have added the desired counters to the chart, you can export it as a file by using the Export Chart... option under the File pull-down menu. One file format option is the comma-separated variable (CSV) format, which most spreadsheets can easily read. RKNG, p. 559, Chp. 12, Using Performance Monitor; TPWIN2, pp. 456-457, Chp. 14, Monitoring TCP/IP with Performance Monitor; "Chapter 5 Keeping Connected" on TechNet.

 

 

32. You manage a network with six subnets. The network has two DHCP servers and three WINS servers. All client computers on the network are Windows-based computers and use DHCP. One user reports receiving a message that she has a duplicate IP address. What is the cause of the problem?

a. DHCP lookup is not enabled on one of the WINS servers.

b. The user's computer is registered with more that one WINS server.

c. The DHCP servers have overlapping scopes.

d. WINS lookup is not enabled on one of the DHCP servers.

 

Choice c is correct. DHCP scopes generally correspond to physical subnets. In order to provide for redundancy should one DHCP server fail, you should create secondary (reserve) scopes for neighboring subnets. This assumes that BOOTP routing is enabled. A normal rule of thumb is to allocate 70% of the available IP addresses to the primary scope and 30% to the secondary scope. However, because no DHCP server can know what addresses have been assigned by another DHCP server, great care must be exercised to ensure that there is no overlapping of IP addresses among scopes. Where overlapping does occur, the assigning of duplicate IP addresses will inevitably result. Note that beginning with Service Pack 2, DHCP for Windows NT 4.0 contains a new "Conflict Detection Attempts" listbox that helps to detect duplicate IP addresses before they are assigned.

RKNG, p. 428, Chp. 7, Setting Local Policies; TPWIN2 p. 290, Chp. 10, Managing Multiple DHCP Servers; "DHCP: Detecting and Flagging Duplicate IP Addresses" on TechNet.

 

 

33. Your company has the network ID 144.104.0.0. Each subnet must provide at least 700 hosts IDs. Which subnet mask will provide the minimum number of host IDs with the greatest number of subnets?

a. 255.255.248.0

b. 255.255.252.0

c. 255.255.254.0

d. 255.255.255.0

 

Choice b is correct. The subnet mask 255.255.252.0 will create 62 subnets with 1,022 hosts each. The next largest subnet mask 255.255.254.0 would create 126 subnets, but only 510 hosts. Seen in binary format, the subnet mask 255.255.252.0 reads as 11111111.11111111.11111100.00000000. Examining the third octet, we see that the first six digits are reserved, i.e., covered up by ones. It is within these six digits that the subnets are created. We eliminate 000000 and 111111, since all zeros and all ones are reserved. This leaves the six-digit combinations 00001 through 111110 or, in decimal, 1 through 62. We calculate the number of hosts by raising two to the tenth power, then subtracting two. There are ten remaining digits in the third and fourth octets (00.00000000). We subtract two because all zeros and all ones are reserved. The number of possible hosts IDs will run from 0000000001 through 1111111110.

Converted to decimal, this equals 1 through 1,022. RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, pp. 90-106, Chp. 4, IP Addressing; LTM, Subnets and Subnet Masking.

 

 

34. Sherry wants to add an #INCLUDE statement in an LMHOSTS file for several remote servers. What additional steps must she take for the information in the #INCLUDE statement to be properly loaded into the cache?

a. Enclose the #INCLUDE statement with #BEGIN_ALTERNATE and #END_ALTERNATE statements.

b. Place a #DOM tag before the #INCLUDE statement.

c. Place a #PRE tag before the #INCLUDE statement.

d. Use the Universal Naming Convention (UNC) to specify the remote servers.

 

Only choices c and d are correct. The LMHOSTS file locates remote computers for file, printer, and remote access services, as well as for domain services such as logon, browsing and replication. The #INCLUDE statement lists remote hosts according to the Universal Naming Convention (UNC). When TCP/IP initializes, it first loads into the cache the entries containing the #PRE tag. It then parses the entries in the #INCLUDE statement as if they were local. This can best be illustrated with the following example:

119.33.84.98 remotesrv #PRE

119.33.84.97 bostonsrv #PRE #DOM:NewEngland #NEHQ

119.33.84.123 distribsrv #PRE #distribution server

#INCLUDE \\remotesrv\public\lmhosts #LMHOSTS from this server

#INCLUDE \\bostonsrv\public\lmhosts #LMHOSTS from this server.

At startup, the IP addresses for remotesrv, bostonsrv and distribsrv are loaded into the cache because each carries the #PRE tag.

The #INCLUDE statements are then read for remotesrv and bostonsrv. Based on the paths given in the UNC entries and the IP addresses taken from the #PRE listing, the LMHOSTS files at remotesrv and bostonsrv are parsed for their #PRE entries. In this way, it is possible to maintain one central LMHOSTS file for an entire network. The local LMHOSTS file only requires a #PRE tag for the server that the central file resides on and an #INCLUDE entry for the path to the file. The #DOM tag is only used to identify domain servers. The #BEGIN_ALTERNATE and #END_ALTERNATE statements are used to group #INCLUDE statements. This is often used to provide LMHOSTS file redundancy in case a server should fail. Each #INCLUDE statement in the group is searched until one loads and the search is stopped. RKNG, p. 518, Chp. 10, Defining a Central LMHOSTS File by Using #INCLUDE; TPWIN2 pp. 356-358, Chp. 11, Managing LMHOSTS files.

 

 

35. Rose oversees a network with four subnets: SubnetA, SubnetB, SubnetC and SubnetD. A Windows NT Workstation computer user on SubnetB complains that he cannot connect to a Windows NT Server computer on SubnetA, but he can connect to local servers. Other users on SubnetB have no trouble connecting to the SubnetA server. Rose runs IPCONFIG on the workstation and notes the following output:

Ethernet adapter EE161:

IP Address. . . . . : 144.104.84.21

Subnet Mask . . . . : 255.255.224.0

Default Gateway . . : 144.104.32.1

What is the reason the workstation cannot connect to the remote server?

a. The default gateway is incorrect.

b. The Ethernet adapter cannot run in 32-bit mode.

c. The IP address is incorrect.

d. The subnet mask is incorrect.

 

Choice a is correct. A default gateway's IP address must be in the same subnet as the host's IP address. An address beginning with 144 indicates a class B address. On a network with four subnets, the most efficient class B subnet mask is 255.255.224.0, which creates six subnets. These subnets are comprised of the following ranges:

SubnetA: 144.104.32.1 through 144.104.63.254

SubnetB: 144.104.64.1 through 144.104.95.254

SubnetC: 144.104.96.1 through 144.104.127.254

SubnetD: 144.104.128.1 through 144.104.159.254

SubnetE: 144.104.160.1 through 144.104.191.254

SubnetF: 144.104.192.1 through 144.104.223.254

Note that the 255.255.224.0 subnet mask makes the address ranges 144.104.0.1 through 144.104.31.254 and 144.104.224.1 through 144.104.255.254 illegal. Looking at the IPCONFIG output, we see that the host IP address 144.104.84.21 makes it a member of SubnetB. The default gateway IP address 144.104.32.1 makes it the very first member of SubnetA.

For this reason, the host cannot send remote IP addresses to its default gateway for resolution, causing the connection to the remote computer to fail. One could argue that the host IP, not the default gateway, is incorrect. But in such a case, the host would not be able to connect to other hosts on the subnet. RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, pp. 90-106, Chp. 4, IP Addressing; LTM, Subnets and Subnet Masking.

 

 

36. You have been hired as a consultant to plan a TCP/IP network for a company with the network address 216.176.255.0. You determine that the network will require nine subnets. Which subnet mask will provide the highest number of host IDs per subnet?

a. 255.255.255.192

b. 255.255.255.224

c. 255.255.255.240

d. 255.255.255.248

 

Choice c is correct. The subnet mask 255.255.255.240 will create 14 subnets with 14 hosts each. The leftmost octet identifies the class of a network address. Class C addresses run from 192 through 223. Applying subnet masks to class C addresses requires carefully weighing the possible combinations of subnets versus host IDs. Only five combinations are available:

Mask / Subnets / Hosts

255.255.255.192 / 2 / 62

255.255.255.224 / 6 / 30

255.255.255.240 / 14 / 14

255.255.255.248 / 30 / 6

255.255.255.252 / 62 / 2

The last octet in the subnet mask 255.255.255.240 translates to binary as 11110000. Each 4-bit segment, expressed as 1111 for the subnet ID and 0000 for the host ID, creates separate addressing ranges of 0001 through 1110, or 1 through 14. In neither segment are all ones or all zeros allowed. Of course, a total of 14 host IDs only holds true for class C addresses.

One and two additional octets are available for class B and class A host addresses respectively, as shown in the following table:

Class A mask: 11111111.11110000.00000000.00000000 (255.240.0.0)

Class B mask: 11111111.11111111.11110000.00000000 (255.255.240.0)

For a class A address, the number of host IDs is 2 raised to the power of 20 minus two or 1,048,574. For a class B address, the number of host IDs is 2 raised to the power of 12 minus 2 or 4,094. In all cases, the number of subnets remains the same at 14. RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, pp. 90-106, Chp. 4, IP Addressing; LTM, Subnets and Subnet Masking.

 

 

37. Jerry's network includes five Windows NT Server computers configured as static routers. What should Jerry use to identify the path that a data packet takes as it passes through the routers?

a. ipconfig

b. netstat

c. nslookup

d. tracert

 

Choice d is correct. The tracert utility is a simple but powerful tool for determining the path that a data packet takes in reaching its destination. Tracert sends an Internet Control Message Protocol (ICMP) packet with an incrementing Time-To-Live (TTL) value. TTL values are designed to prevent data packets from looping endlessly on a network. The default TTL value on a packet is 30. As the packet passes through each router, the router subtracts one from the TTL value. When the TTL value reaches zero, the router drops the packet and returns an ICMP Time Exceeded message to the source. Tracert begins by sending an ICMP packet to the destination address with a TTL value of one. Subtracting one from the TTL value, the first router sends back an ICMP Time Exceeded message. Tracert notes the router's address and sends out a second ICMP packet. In this way, tracert charts the course of a packet through all the routers to the destination.

RKNG, p. 556, Chp. 12, Examine the Route Between Network Connections by Using Tracert, and p. 706, App. A, tracert; TPWIN2, pp. 268-269, Chp. 9, Testing Routing with Tracert; "Using TRACERT to Troubleshoot TCP/IP Problems in Windows NT" on TechNet.

 

 

38. Gene has a Windows NT Workstation computer that uses a HOSTS file, not a DNS server, for hostname resolution. His local HOSTS file contains the following entries:

176.3.22.19 market #market telnet server

133.3.18.25 dataserv #central data server

176.3.66.87 NT_research #NT researchers server

176.3.66.122 researcher #researchers file server

When Gene runs the command ftp dataserv, he connects to a UNIX server on a remote subnet. However, when Gene runs the command ftp 176.3.18.25, he cannot connect to the UNIX server. What is the cause of the problem?

a. Gene does not have FTP access permissions on the UNIX server.

b. The IP address of the UNIX server is not 176.3.18.25.

c. The name of the UNIX server is not dataserv.

d. The UNIX server cannot resolve the workstation's NetBIOS name.

 

Choice b is correct. Since Gene can successfully use the "ftp dataserv" command, the IP address of the UNIX server must be correct as listed in Gene's HOSTS file. The IP address of dataserv is 133.3.18.25 according to the HOSTS file, not 176.3.18.25. With both HOSTS and LMHOSTS files, users must also be certain that the IP address is correct. If Gene had been denied FTP access permissions on the UNIX server, he would not have been able to successfully connect to the server using the ftp dataserv command. TPWIN2, pp. 150-151, Chp. 6, Naming Hosts on the Internet.

 

 

39. You are setting up a DHCP Relay Agent on a Windows NT Server computer. What information do you need?

a. the DHCP node type

b. the DHCP Relay Agent lease duration

c. the IP address of the DHCP server

d. the NetBIOS name of the DHCP server

 

Choice c is correct. DHCP Relay Agents are located on subnets that do not have their own DHCP servers. Any Windows NT Server or Workstation computer configured with the DHCP Relay Agent service can forward DHCP requests for IP addresses across the network to the DHCP Server. You install the DHCP Relay Agent service on the Services tab of the Network Dialog box in Control Panel. On the DHCP Relay tab of the TCP/IP Properties page, add the IP address of the DHCP Server. RKNG, pp. 212-214, Chp. 5, Configuring DHCP; SPT, p. 155, Chp. 7, Lesson 3: Enabling a DHCP Relay Agent.

 

 

40. What is the most likely reason you would change the NetBIOS scope value on the WINS Address tab?

a. to exclude a NetBIOS computer name in the DHCP database

b. to exclude a NetBIOS computer name in the WINS database

c. to implement the DNS service

d. to isolate a group of computers on the network

 

Choice d is correct. The NetBIOS scope allows you to isolate a group of computers on a network. By default, the NetBIOS scope value is NULL. The NetBIOS scope, sometimes called the TCP/IP scope, appends a second name to the NetBIOS computer name. Only computers with the same NetBIOS scope can communicate with other computers. For this reason, the NetBIOS scope option should be used with extreme caution. It should not be used in conjunction with the DNS service. It can also break trust relationships between domains, since pass-through authentication cannot occur on domains with different NetBIOS scopes. RKNG, pp. 374-375, Chp. 6, Network Application Interfaces.

 

 

41. From his Windows NT Workstation computer, Phillip wants to connect to a Windows NT Server computer on a remote subnet. When he runs the command ftp remoteserv.sspcorp.com, he makes the connection. However, when he tries to connect to the same server by specifying its computer name in Windows NT Explorer, he fails to connect. What is a possible cause of the problem?

a. The workstation is not set up to use the DNS service.

b. The server is not set up to use the WINS service.

c. The subnet mask on the workstation is incorrect.

d. The subnet mask on the server is incorrect.

 

Choice b is correct. Since communication between the workstation and server is possible using FTP, the subnet mask on both computers must be correct. The FTP service uses the hostname to make the connection. To resolve hostnames to their IP addresses, the DNS service is required. The Windows browsing service used by Windows NT Explorer requires LMHOSTS files or WINS to resolve NetBIOS names to their IP addresses. Therefore, a possible cause of the problem would be that the server (or workstation) is not set up to use the WINS service. RKNG, pp. 149-150, Chp. 3, Browse Service Across a WAN with TCP/IP; TPWIN2, pp. 328-331, Chp. 11, Installing the WINS Server Service.

 

 

42. Harry administers the TCP/IP network shown in the Exhibit. He wants to add one static routing entry to ST_RTR_B and one static routing entry to ST_RTR_C so that all three subnets can communicate with one another.

Which pair of static routing entries should Harry add?

a. on host ST_RTR_B: route add 166.15.90.0 mask 255.255.255.0 166.15.75.1

on host ST_RTR_C: route add 166.15.40.0 mask 255.255.255.0 166.15.75.2

b. on host ST_RTR_C: route add 166.15.90.0 mask 255.255.255.0 166.15.75.1

on host ST_RTR_B: route add 166.15.40.0 mask 255.255.255.0 166.15.75.2

c. on host ST_RTR_B: route add 166.15.75.1 mask 255.255.224.0 166.15.240.0

on host ST_RTR_C: route add 166.15.75.2 mask 255.255.240.0 166.15.224.0

d. on host ST_RTR_B: route add 166.15.40.0 166.15.75.2

on host ST_RTR_C: route add 166.15.90.0 166.15.75.1

 

Choice a is correct. The correct syntax is ROUTE ADD [destination net ID] MASK [netmask] [gateway address]. The mask is only needed when subnet masking is used. The rule in creating static routing is to bind the outer subnet (the destination net ID) through the IP address on the far side of the middle subnet (the gateway address). In the above example, ST_RTR_B binds the outer subnet 166.15.90.0, i.e., the subnet it cannot reach directly, through the gateway address 166.15.75.1. ST_RTR_B can connect to 166.15.75.1 because one of its network adapter cards is on the subnet 166.15.75.0. The gateway address 166.15.75.1 can connect to the network 166.15.90.0 because both share the multihomed computer ST_RTR_C. Thus, the subnet 166.15.40.0 can "see" its way to subnet 166.15.90.0. RKNG, pp. 554-556, Chp. 12, The Route Table; "TCP/IP Routing Basics for Windows NT" on TechNet.

 

 

43. Tim wants to use Performance Monitor on his Windows NT Workstation computer to compile the TCP/IP network statistics generated by his network's Windows NT Server computers. What does Tim need to do?

a. Install the SNMP service on each Windows NT Server computer.

b. Install the SNMP service on his Windows NT Workstation computer.

c. Set up each server to use the WINS service.

d. Set up each Windows NT Server computer as an SNMP trap.

e. Set up his Windows NT Workstation computer as an SNMP trap.

 

Choice a is correct. Capturing TCP/IP network statistics with Performance Monitor requires that the SNMP service be installed on those Windows NT-based computers you want to monitor. Performance Monitor counters can measure statistics for NIC, IP, ICMP, UDP, TCP and NetBT. When both services are running, you can also measure FTP server and WINS server performance objects. NTCP, p. 278, Chp. 8, Monitoring Network Activity; TPWIN2, pp. 456-457, Chp. 14, Monitoring TCP/IP with Performance Monitor.

 

 

44. A Windows 95 user complains that she cannot connect to any computers on her local subnet even when using their IP addresses. Seated at her computer, you ping 127.0.0.1 and receive a reply. Other users on her subnet do not report having any trouble. All client computers on the network are WINS-enabled. What is the most probable cause of the problem?

a. All other computers on the user's subnet are non-Windows-based computers.

b. The default gateway address on the Windows 95 computer is incorrect.

c. The subnet mask on the Windows 95 computer is incorrect.

d. The WINS server for her subnet is down.

 

Choice c is correct. An improperly entered subnet mask may cause IP to view local addresses as remote addresses or vice-versa. In such cases, IP will send the data packet to the default gateway for remote resolution. Of course, the default gateway will not know where to forward the packet since its destination is really local. Address resolution fails. Pinging 127.0.0.1 is always a good idea when troubleshooting TCP/IP problems. Successfully pinging 127.0.0.1 indicates that TCP/IP is installed on the machine. However, it says nothing about local configuration variables such as the correct default gateway or the subnet mask. RKNG, pp. 347-349, Chp. 6, Address Resolution Protocol; TPWIN2, pp. 110-112, Chp. 4, IP Routing; "TCP/IP Name Resolution" on TechNet.

 

 

45. Sharon's boss has asked her to determine the number of host IDs their company's new TCP/IP network will require. How should Sharon calculate the total number of host IDs required?

a. She should estimate one host ID for each user profile.

b. She should estimate one host ID for each PC network adapter card.

c. She should estimate one host ID for each subnet.

d. She should estimate one host ID for each router interface.

 

Only choices b and d are correct. Together, a host ID and a network ID form an IP address. The network ID identifies the network. The host ID identifies each device on the network. For this reason, every device attached to a TCP/IP network, such as a network adapter card, a router, or a network-attached printer, requires a host ID that is unique to that network. A network that is not attached to the Internet can have its own system of IP addressing as long as that system conforms to TCP/IP networking rules. If a network is to be attached to the Internet, however, then the network must obtain an IP address from the Internet Network Information Center (InterNIC). RKNG, pp. 404-406, Chp. 7, Introduction to Dynamic Host Configuration Protocol; TPWIN2, p. 90, Chp. 4, IP Addressing; LTM, Subnets and Subnet Masking.

 

 

46. Tom administers 180 Windows-based hosts on a network with six subnets. Two DHCP servers dynamically assign IP addresses to the host computers. If one of the DHCP servers fails, how can Tom provide for DHCP addressing redundancy on his network?

a. by implementing one global DHCP scope on each DHCP server

b. by implementing two global DHCP scopes on each DHCP server

c. by implementing three DHCP scopes on each DHCP server

d. by implementing six DHCP scopes on each DHCP server

 

Choice d is correct. Currently, Microsoft DHCP servers do not share or replicate their databases, so any kind of recovery or backup server strategy is not possible. However, you can provide for uninterrupted DHCP service should one DHCP server fail by creating secondary address scopes on a second DHCP server. Because neither DHCP server can know what addresses the other server has assigned, there can be no overlapping of IP addresses between primary and secondary scopes. RKNG, p. 428, Chp. 7, Setting Local Policies; TPWIN2 p. 290, Chp. 10, Managing Multiple DHCP Servers.

 

 

47. Mark has a number of files he wishes to copy from a remote UNIX server to his Windows NT Server computer. Which utility should Mark use?

a. ftp

b. lpq

c. lpr

d. xcopy

 

Choice a is correct. The File Transfer Protocol (FTP) provides a simple method of transferring files to and from UNIX computers. You can run FTP either at the command prompt or through Internet Explorer. However, Internet Explorer only allows you to download, not upload, files using FTP. When FTP is used at the command prompt, the get command downloads files, the put command uploads files. Remember, however, that FTP is an inherently unsecure protocol when used on the Internet. If security is a concern, use FTP only anonymously or choose another protocol. RKNG, pp. 306-309, Chp. 5, UNIX Administration, and pp. 683-685, App. A, FTP; TPWIN2, pp. 185-189, Chp. 6, TCP/IP Applications.

 

 

48. On her Windows NT Workstation computer, Debbie can connect to a remote Windows NT Server computer by specifying the server's UNC path in Windows NT Explorer. However, when she tries to connect via FTP using the server's IP address 199.144.5.5, Debbie cannot connect. What is one possible cause of the problem?

a. The default gateway address of the Windows NT Workstation computer is incorrect.

b. The default gateway address of the Windows NT Server computer is incorrect.

c. The Windows NT Server computer is not set up to use the WINS service.

d. The Windows NT Workstation computer is denied access to the FTP server's directories.

e. The Windows NT Workstation computer is not set up to use the DNS service.

 

Choice d is correct. In this scenario, the most likely cause of the problem is the FTP access permission. The default gateway addresses on both the workstation and server must be correct since a UNC connection is possible. If either were wrong, connections of any type would be impossible. Neither the WINS service nor the DNS service plays a role, since neither NetBIOS name nor hostname resolution is involved when using an IP address. On the other hand, using the FTP utility invokes FTP logon security on the remote server. Thus, it would be possible to connect to the remote server by using Windows NT Explorer, HTTP or Telnet, and still fail to connect using FTP. By default, Microsoft's FTP service is set to allow anonymous connections, but username/password security can be implemented. NTCP, p. 108, Chp. 3, System Policy; RKNG, pp. 306-309, Chp. 5, UNIX Administration; TPWIN2, pp. 177-181, Chp. 6, TCP/IP Applications.

 

 

49. Earl wants all of the Windows-based client computers on his network to use a DNS server to resolve any names not found on the WINS server. What is the best way for Earl to proceed?

a. He should enable WINS lookup for name resolution on the DNS server.

b. He should enable DNS for name resolution on each client computer.

c. He should list DNS first then WINS under name resolution on each client computer.

d. He should list WINS first then DNS under name resolution on each client computer.

 

Choice b is correct. When the Enable DNS For Windows Resolution check box on the WINS Address tab of the Microsoft TCP/IP Properties page is selected, the WINS service will pass on to the DNS service any hostnames it cannot resolve. Name queries longer than 15 characters are sent directly to the DNS service for resolution because NetBIOS names cannot exceed 15 characters. Name queries of 15 characters or less are sent to the WINS service. If the WINS service cannot resolve the name, then it passes the name on to the DNS service for resolution. It is possible in the client computer's registry to select which service will first attempt to resolve a name. However, this is not recommended. RKNG, pp. 490-492, Chp. 9, Planning for Microsoft DNS and Microsoft WINS Integration.

 

 

50. Cindy has just installed the DNS service on a Windows NT Server computer. She needs to add a resource record for her domain's mail server. Which resource record must she add?

a. CNAME

b. MX

c. PTR

d. WKS

 

Choice b is correct. Resource records are contained in DNS database files that map hostnames to the IP addresses of important TCP/IP network computers. The MX record is the mail exchange resource record that either processes or forwards mail for the DNS domain name. The CNAME (canonical name) resource record allows you to create an alias for a given hostname. The PTR (IP Address to Name Pointer) resource record, the opposite of the A resource record, maps IP addresses to hostnames in a DNS reverse-lookup zone. The WKS (well-known service) resource record describes which services a particular protocol will provide on a given interface. RKNG, pp. 479-482, Chp. 9, Microsoft DNS Server Database; TPWIN2, pp. 371-373, Chp. 12, Managing Microsoft DNS Server.

 

 

51. Working at a Windows NT Server computer, Mark wants to view a chart of TCP/IP protocol statistics for his server. What should he use?

a. arp

b. nbtstat

c. netstat

d. Network Monitor

e. Performance Monitor

 

Choice e is correct. The chart mode is the default mode for Performance Monitor. The TCP/IP utilities arp, nbtstat and netstat give only snapshots of a Windows NT computer's TCP/IP configuration. Network Monitor captures data that must be viewed using Network Monitor's display filters. Performance Monitor offers seven TCP/IP objects for monitoring: FTP Server (19 counters), ICMP (27 counters), IP (17 counters), Network Interface (8 counters), TCP (9 counters), UDP (5 counters) and WINS Server (15 counters). However, before using Performance Monitor for TCP/IP, you must install SNMP. NTCP, pp. 310-311, Chp. 10, Network Monitor Overview; RKNG, p. 559, Chp. 12, Using Performance Monitor; TPWIN2, pp. 456-457, Chp. 14, Monitoring TCP/IP with Performance Monitor.

 

 

52. Rachel fails repeatedly to access a local Windows NT Server computer on her subnet. Using Network Monitor to troubleshoot the problem, she finds that every time she tries to connect to the server, her workstation broadcasts an ARP request for the default gateway. No other users on the TCP/IP network have trouble accessing the server. What is the cause of Rachel's problem?

a. The workstation is not set up to use the DNS service.

b. The workstation is not set up to use the WINS service.

c. The workstation is set up with a duplicate IP address.

d. The workstation is set up with an incorrect subnet mask.

 

Choice d is correct. On a local subnet, IP must resolve the destination IP address to its media access control (MAC) address. Neither HOSTS, DNS nor WINS is required. First, however, IP must determine that the destination address is local. It does this through a process called ANDing. IP compares its own host address against the host subnet mask, compares the destination address against the host subnet mask, and compares the results. If the results are the same, then the address is local. ARP broadcasts on the local network to find the computer with that IP address. The computer with the matching IP address answers the broadcast and returns its MAC address. If the results are not the same, then IP sends the request to the default gateway for remote resolution through whatever means are available. For this reason, an incorrect subnet mask will produce false ANDing results. A local address will appear to be a remote address, and a remote address could appear to be local.

In cases of duplicate IP addresses, Windows NT immediately notifies the second user of the address error when he or she logs on. RKNG, pp. 347-349, Chp. 6, Address Resolution Protocol; TPWIN2, pp. 110-112, Chp. 4, IP Routing; "TCP/IP Name Resolution" on TechNet.

 

 

53. Dustin's network has 2 DNS servers. Dustin plans to set up a third DNS server. Dustin plans to use this third DNS server to process DNS queries for Internet resources. How should this third DNS server be set up?

a. as a zone server

b. as a forwarder

c. as a primary server

d. as a secondary server

 

Choice b is correct. If a DNS server cannot resolve a DNS request within its own zone, the request can be sent to a forwarder. A forwarder is a DNS server configured to resolve out-of-zone requests, usually by contacting other DNS servers on the Internet. A forwarder can also help to restrict direct intranet communication with the Internet. Placed outside a firewall, a forwarder can provide external DNS services to the Internet, thereby shielding internal resources from hackers. RKNG, pp. 487-490, Chp. 9, Planning for Microsoft DNS Server Implementation; "DNS and MS Windows NT 4.0" on TechNet.

 

 

54. You have a RAS server connected to the Internet. Remote users use dial-up networking to connect to the RAS server through a local Internet Service Provider (ISP). You need to provide security so that remote users can access your network using only Point-to-Point Tunneling Protocol (PPTP). How should you set up the RAS server?

a. by activating Microsoft multi-protocol routing

b. by activating multi-link

c. by activating PPTP advanced security

d. by activating PPTP filtering

e. by activating RAS security

 

Choice d is correct. With the advent of Point-to-Point Tunneling Protocol (PPTP) in Windows NT 4.0, Internet security increased dramatically. PPTP employs both Password Authentication Protocol (PAP) and the Challenge Handshake Authentication Protocol (CHAP). In addition, PPTP filtering can further enhance security by disabling the RAS server for all connections other than PPTP. You must define PPTP filtering for each network adapter that's connected to the Internet. RKNG, pp. 337-338, Chp. 5, Protecting a RAS Server from Internet Attacks; TPWIN2, p. 523, Chp. 15, Using the Point-to-Point Tunneling Protocol.

 

 

55. Suppose the following situation exists:

Brandy administers a TCP/IP network of Microsoft-based computers on four subnets. She wants to install two DHCP servers, each on a separate subnet, that will automatically assign IP addresses to the host computers.

Required result:

Optional desired results:

Proposed solution:

Which results does the proposed solution produce?

a. The proposed solution produces the required result and all of the optional desired results.

b. The proposed solution produces the required result and one of the optional desired results.

c. The proposed solution produces the required result, but does not produce any of the optional desired results.

d. The proposed solution does not produce the required result.

 

Choice d is correct. The proposed solution comes close to, but does not produce, the required result. On a routed network, each subnet must have either a DHCP server or relay agent in order to provide cross-network DHCP connectivity. However, for one DHCP server to properly act as a backup server to another DHCP server, there can be no overlapping of IP addresses between respective scopes. For example, assume that DHCP ServerA is responsible for assigning IP addresses to SubnetA. SubnetA's scope is comprised of 14 addresses (w.x.y.17 through w.x.y.30). If DHCP ServerB is to act as a backup to ServerA, then ServerA must give some of its IP addresses to ServerB. ServerA could restrict its scope to w.x.y.17 through w.x.y.25 and give ServerB the IP addresses w.x.y.26 through w.x.y.30.

In this way, you avoid the danger of assigning duplicate IP addresses using DHCP. The first optional desired result, to provide the same unique IP address to each Windows NT Server computer, is produced by creating a client reservation for each Windows NT Server computer. The second optional desired result, to assign the IP addresses of the WINS and DNS servers to all DHCP clients, is not mentioned in the proposed solution. RKNG, p. 428, Chp. 7, Setting Local Policies; TPWIN2 p. 290, Chp. 10, Managing Multiple DHCP Servers.

 

 

56. Suppose the following situation exists:

Tom administers a TCP/IP network of Microsoft-based computers on six subnets. He wants to install two DHCP servers, each on a separate subnet, that will automatically assign IP addresses to the host computers.

Required result:

Optional desired results:

Proposed solution:

Which results does the proposed solution produce?

a. The proposed solution produces the required result and all of the optional desired results.

b. The proposed solution produces the required result and one of the optional desired results.

c. The proposed solution produces the required result, but does not produce any of the optional desired results.

d. The proposed solution does not produce the required result.

 

Choice b is correct. The proposed solution produces the required result and one optional desired result. DHCP servers cannot replicate amongst themselves the IP addresses that they have assigned. For this reason, each DHCP server must have a unique range of IP addresses for each scope. These ranges must never overlap. How you allocate IP addresses among DHCP servers depends upon many factors: the number of hosts and subnets, the amount of network traffic, and the reliability of the network and its servers. As a general rule, Microsoft recommends a 70/30 percent split between primary and secondary (backup) scopes. The first optional desired result is produced by creating a client reservation on the DHCP server for each Windows NT Server computer. This ensures that the same unique IP address will be issued to that Windows NT Server computer upon initialization.

The second optional desired result, assigning the IP addresses of the WINS and DNS servers to all DHCP clients, is not mentioned in the proposed solution. RKNG, p. 428, Chp. 7, Setting Local Policies; TPWIN2 p. 290, Chp. 10, Managing Multiple DHCP Servers.

 

 

57. Camille administers a TCP/IP network with four subnets: SubnetA, SubnetB, SubnetC and SubnetD. The network ID is 211.40.88.0 with a subnet mask of 255.255.255.224. The IP address range on each subnet is as follows:

SubnetA: 211.40.88.33 - 211.40.88.62

SubnetB: 211.40.88.65 - 211.40.88.94

SubnetC: 211.40.88.97 - 211.40.88.126

SubnetD: 211.40.88.129 - 211.40.88.158

The network adapter cards that connect the network's three routers to the four subnets have the following IP addresses:

RouterAB: 211.40.88.62 and 211.40.88.65

RouterBC: 211.40.88.94 and 211.40.88.97

RouterCD: 211.40.88.126 and 211.40.88.129

A Windows 95 user on SubnetB (IP address 211.40.88.85) complains to Camille that he cannot connect to a Windows NT Server computer (IP address 211.40.88.101) on SubnetC. He reports no problems connecting to a Windows NT Server computer on SubnetA (IP address 211.40.88.50) using Windows Explorer.

Camille pings 211.40.88.101 from the Windows 95 computer and receives a Request Timed Out message. In testing the user's network connections, which IP address should Camille ping next?

a. 127.0.0.1

b. 211.40.88.65

c. 211.40.88.85

d. 211.40.88.94

e. 211.40.88.97

 

Choice d is correct. A problem in pinging another IP address, especially a remote one, can occur at any node along the path. Pinging the host address (or 127.0.0.1, the loopback address) only ensures that TCP/IP has been properly installed on the local host. In this case, this is unnecessary because the user's ability to connect to SubnetA proves that TCP/IP is properly installed. Therefore, the next step is to ping the near side of the router, the port 211.40.88.94, to determine that the router connection works. If you can ping the router's near side, then ping the router's far side, the port 211.40.88.97, to determine that the internal connection 211.40.88.94-211.40.88.97 works. If it does, then the SubnetB host can find its way to SubnetC. At this point, the most likely cause of the problem lies with the destination host, e.g., it has been disconnected from the network or turned off.

RKNG, pp. 548-550, Chp. 12, Test Connection to the TCP/IP Network by Using Ping; TPWIN2, pp. 235-237, Chp. 8, Testing the TCP/IP Configuration.

 

 

58. You manage a single Windows NT domain on a TCP/IP network divided into six subnets. One subnet contains the primary domain controller (PDC). Each of the other five subnets has a backup domain controller (BDC). WINS is not enabled on your network. You want each domain controller to be able to validate client logon attempts from any subnet. You also want the PDC to replicate any changes made to its user accounts. How can you best do this?

a. Create an LMHOSTS file with entries for the other domain controllers on each BDC.

b. Create an LMHOSTS file with entries for the domain controllers on each computer.

c. Install the Microsoft DNS service on one of the domain controllers and configure each client computer with the DNS server's IP address.

d. Install the Microsoft DNS service on one of the domain controllers and configure each domain controller with the DNS server's IP address.

 

Choice b is correct. On a small, stable network, an LMHOSTS file listing the domain controllers on each computer will ensure that either a PDC or BDC can validate that computer's logon. A computer that is promoted to be the master browser of its subnet will automatically poll the subnet and forward its browse list to the PDC. In turn, the PDC will send the merged browse list for the entire network back to the BDC. The format for adding a domain controller to an LMHOSTS file is:

10.10.10.10 DomainControllerName #PRE #DOM:DomainName

The #PRE tag ensures that the entry will be cached by the host computer upon initialization.

Microsoft states that the future of network browsing using LMHOSTS files is uncertain.

For this and other reasons, administrators depending upon LMHOSTS files should start looking at WINS as a browsing alternative. In a strictly Windows-based environment, there is no need to install DNS, which is used for hostname resolution, not NetBIOS name resolution. RKNG, pp. 508-510, Chp. 10, Using LMHOSTS File to Find Computers and Services; TPWIN2, p. 358, Chp. 11, Managing LMHOSTS Files; "Domain Browsing with TCP/IP and LMHOSTS Files" on TechNet.