Sams Teach Yourself MCSE Windows NT Server 4 in 14 Days
(Publisher: Macmillan Computer Publishing)
Author(s): David Schaer, et al
ISBN: 0672311283
Publication Date: 12/15/97

Appendix B
Answers to Review Questions

Chapter 2, “Windows NT Architecture and Domain Design”

Question 1

Answer: D

The server must be a primary domain controller to act as a security host. The only way for a stand-alone or member server to become a controller is to reinstall. (See section 2.6.1.)

Question 2

Answer: A

The controller being available to validate the users is a byproduct of a poor solution. Because the links are not in place when the PDC is installed in Los Angeles, it has no way of knowing that there is already a PDC existing for the VALIDATE domain. The system administrator is successful in installing a PDC in a domain called VALIDATE; however, it is not the same VALIDATE domain the rest of the company is using. Because the unique security identifier (SID) for a domain is created when the PDC is installed, even after the links are up the systems will fail to communicate properly. (See section 2.6.5.)

Question 3

Answer: D

Any account that is a member of the Domain Admins group on MASTERDOM could be used. (See section 2.6.4.)

Question 4

Answer: A, B, C

Because each of the machines is running Microsoft NT they can act as an application server. It is true that the most likely choice on which to install SQL would be the member server because it does not have to validate domain logons. The question does not ask which is the best choice, only which systems are capable of performing the task. (See section 2.6.)

Question 5

Answer: D

The PDC synchronizes its SAM database only with the backup domain controllers. Each BDC in a domain receives synchronization updates from the PDC to ensure that the SAM database held at the PDC is identical to that at each of the controllers. (See section 2.6.3.)

Question 6

Answer: A, B

The system administrator has caused his own problem. When a domain account is modified, the change must first be recorded in the SAM database at the PDC. Over the course of time the update would be sent back to the BDC. Because of the excessive number of BDCs, synchronizing the accounts between all controllers will take a long time. When a BDC is requested to validate an account within the domain, it checks the account information against its copy of the SAM database. If the password is incorrect the BDC can pass the account information to the PDC for verification. If the PDC is too busy to respond, a delay or logon failure could result. (See section 2.6.3.)

Question 7

Answer: D

The Security ID (SID) can be assigned only when an object is created. This is true of the SID assigned to controllers, users, and groups. When the system administrator installed the BDC to the wrong domain it was branded with the SID from that domain. Changing the name of the domain in the network properties would not replace the underlying SID. Unfortunately, reinstalling is the only way to move a BDC from one domain to another. (See section 2.6.5.)

Question 8

Answer: True. Through the process of pass-through authentication, her validation request is passed from the PDC of Domain B to the PDC of Domain A, where her user account resides. (See section 2.7.1.)

Question 9

Answer: B and D. Trust relationships require a permanent link between two NT servers that are PDCs. Trust relationships also require a common protocol. (See section 2.7.1.)

Question 10

Answer: False. You cannot do this. Only a user account in a trusted domain can access a resource in a trusting domain, not the other way around. (See section 2.7.1.)

Question 11

Answer: True. User accounts in a trusted domain can be put into a local group in a trusting domain. It is recommended, however, that the user account first be put into a global group and that the global group then be put into a local group. (See section 2.7.1.)

Question 12

Answer: False. Remember that trusts are non-transitive. Just because Domain A trusts Domain B and Domain B trusts Domain C does not mean that Domain A trusts Domain C. (See section 2.7.1.)

Question 13

Answer: True. Synchronization, however, may still occur. (See section 2.6.6.)

Question 14

Answer: B. 40MB (See section 2.6.2.)

Question 15

Answer: C. Seven. You need one domain controller for every 2,000 users. For 16,000 users, you need eight domain controllers: one PDC and seven BDCs. (See section 2.6.3.)

Chapter 3, “Installing NT Server 4.0”

Question 1

Answer: C

A RISC-based machine boots only from installed hardware. An NT installation is always done from the CD-ROM.

Question 2

Answer: B

Because each machine supports standard VGA drivers, you need only one unattend answer file.

Question 3

Answer: B

The NTHQ diagnostic disk is bootable and will confirm whether the hardware is NT compatible.

Question 4

Answer: D

Because NetBEUI is not routable, the BDC and PDC would not be able to communicate over the router.

Question 5

Answer: D

The repair disk utility enables you to update the repair information as well as create a new ERD.

Question 6

Answer: A

Only one hard disk is required, although two partitions are necessary—one partition formatted with FAT and one formatted with NTFS.

Question 7

Answer: B

In order to install a backup domain controller, it must be able to communicate with the primary domain controller.

Question 8

Answer: A, B

Before TCP/IP can be initialized, both an IP address and subnet mask must be configured.

Question 9

Answer: C

The DLC protocol is the protocol necessary to configure Hewlett-Packard JetDirect cards.

Question 10

Answer: A, B

Both TCP/IP and NWLink are routable. DLC and NetBEUI are not.

Question 11

Answer: All of the protocols can be forwarded by bridges.

Question 12

Answer: D. NetBEUI

Question 13

Answer: C

NWLink enables NetWare clients to communicate with NT servers.

Question 14

Answer: A, B

The Gateway Services for NetWare provides client services on an NT server as well as Gateway services.

Chapter 4, “Configuring the Environment”

Question 1

Answer: B, C

The DependOnService entry lists the services that must be configured before the Messenger service starts.

A is incorrect. The Alerter service is not listed as a dependency. The Messenger service is not dependent on the Alerter service. The Alerter service is, however, dependent on the Messenger service to deliver the alerts it generates. The recipients of the alert also must be running the Messenger service.

B is correct. The LanmanWorkstation is synonymous with workstation. Because it is in the list of dependencies it would be required. (See section 4.4.)

C is correct. NetBIOS is listed in the list of dependencies. If the NetBIOS interface does not load properly, the Messenger service will fail.

D is incorrect. The Server service is not included in the dependency list.

Question 2

Answer: E

The StartValue is used to categorize services.

Answer A is incorrect. A StartValue of 0×0 designates the service as one that will be activated at boot time. These services will be called by the NTLDR on x86-based machines.

Answer B is incorrect. A StartValue of 0×1 is used to represent a critical service that is called during the initialization phase.

Answer C is incorrect. A StartValue of 0×2 identifies services that are to be started automatically after NT has initialized. The Workstation and Server services receive a 0×2 StartValue by default.

Answer D is incorrect. A StartValue of 0×3 is assigned to services that must be started manually by the user or on demand by another process.

Answer E is correct. A StartValue of 0×4 is used to identify a service as disabled. (See section 4.5.)

Question 3

Answer: C

Pausing the Server service often is done before sending a message to users to log off the server. Pausing the Server service guarantees that no new users will attach.

Answer A is incorrect. Pausing the Server service will not disconnect the current users. Stopping the Server service would disconnect the users.

Answer B is incorrect. The Alerter service would not generate a message based on the Server service being paused. In any case, the messages generated by the Alerter service are directed to individual people or computers, not to all connected users.

Answer C is correct. Pausing the Server service prevents additional users from attaching. (See section 4.6.3.)

Answer D is incorrect. The Workstation service is independent of the Server service.

Question 4

Answer: A, B, D

Continuing the Server service will not disconnect the users.

Answers A, B, and D are all equally correct. (See section 4.6.3.)

Answer C is incorrect. The command issued in answer C would not work. Technically, the Server service does not need to be started because it was not stopped. The error message that would be received is shown in Figure B.1.

Figure B.1.  Answer C would return an error message.

Question 5

Answer: C

The NT server is a member of DomainC. The account assigned to a service must come from the local SAM database of the member server, the SAM database of the local domain (DomainC), or the SAM database of a trusted domain.

Neither DomainA nor DomainB are trusted domains. They are referred to as trusting domains. (See section 4.2.)

Question 6

Answer: D

NTBACKUP.EXE does not provide a scheduling function. It is quite common to create a batch file that contains the command-line parameters to execute NTBACKUP.EXE. To execute the batch file at specific times, the administrator creates a job to be carried out by the schedule service using either AT.EXE or WINAT.EXE. (See section 4.4.3.)

Question 7

Answer: A, D

Answer A is correct. The Messenger service delivers the alert message. The alerts can be sent to either a user or a computer. (See section 4.4.4.)

Answer B is incorrect. The schedule service is used with AT.EXE or WINAT.EXE to run scheduled events.

Answer C is incorrect. The signal service does not exist as an NT service.

Answer D is correct. The Alerter service is of course required to generate the alert message.

Question 8

Answer: C

By assigning the account to the directory replicator service through Server Manager or Control Panel, Services, the account would have been granted the right to log on as a service automatically. The right could also be granted to an account through User Manager, User Rights. (See section 4.6.2.)

When writing the question I thought it would be too much to expect people to be aware that an account is granted the right to log on as a service as a byproduct of being assigned to the Directory Replicator service. The question simply asks where one would go to confirm the right was assigned. This could only be done through User Manager, User Rights.

Answer A is incorrect. Server Manager could be used to assign the account to the Directory Replicator service but not to verify the right to log on as a service.

Question 9

Answer: B

Answer A is incorrect. If a service is set to Disabled it cannot be started by the user or at the request of another service.

Answer B is correct. Because the service was set to Manual, the server did not attempt to start it after rebooting. A Manual service can be started by the user or at the request of another service. (See section 4.6.2.)

Answer C is incorrect. If the service were set to Automatic, the system would have attempted to start it upon reboot. If the service had simply failed to start, an error message most likely would have been displayed.

Answer D is incorrect. If the service account was not properly configured, the Net Start command would not have worked properly.

Question 10

Answer: D

The Netlogon service is used to synchronize the SAM database between the PDC and BDCs. By pausing the Netlogon service, the PDC will not have to validate logon attempts to the domain but can concentrate on synchronizing the SAM database. (See section 4.6.3.)

Answer A is incorrect. Stopping the Server service also would stop the Netlogon service. This would prevent continued synchronization.

Answer B is incorrect. Pausing the Server service would not pause the Netlogon service. Although no additional users would be able to access the server, the PDC would continue to validate domain accounts.

Answer C is incorrect. Stopping the Netlogon service would stop synchronization.

Answer D is correct. The Netlogon service on the PDC can be paused to allow the system to concentrate on synchronizing the SAM database with the BDCs. The BDCs will continue to perform domain validation for the users.

Question 11

Answer: D

An example would be when you double-click a text file and it is loaded automatically into Notepad.exe. This is based on the association of text files with Notepad. (See section 4.3.)

File associations and information regarding how applications interact via OLE is implemented in HKEY_CLASSES_ROOT.

HKEY_CURRENT_CONFIG contains the current hardware configuration profile loaded during the boot process.

HKEY_CURRENT_USER is the profile for the user presently logged on to the system.

HKEY_LOCAL_MACHINE contains information about the local computer system’s hardware and operating system.

Question 12

Answer: C

If the system is mistakenly configured in a way that prevents the system from starting properly, you can attempt to initialize NT using the LastKnownGood control set. The system automatically creates the LastKnownGood control set each time a user logs on to the system following a reboot of the system. This is updated every time a user logs on. (See section 4.3.)

Question 13

Answer: D

The NT tape backup utility, NTBACKUP, can perform both online backups and restorations of the registry.

In almost all cases, the use of a tape backup utility will be more efficient than any means using disks. (See section 4.3.2.)

The Save Key feature in REGEDT32 will save only that key, not the entire registry.

Mirroring will not back up the entire registry because parts of it are stored in RAM.

The repair disk utility can be used to update the compressed copies of the registry databases in the %SYSTEMROOT%\repair directory, or create an emergency repair disk.

Question 14

Answer: A

The HARDWARE hive or key is built dynamically during the boot process, which is why the HARDWARE key is often referred to as being volatile. The other keys have both a database file and a log file. (See section 4.3.)

On Intel-based systems the HARDWARE key is built from information provided by NTDETECT.COM; RISC-based machines retrieve the information from firmware.

Question 15

Answer: B, D

Because the change involves modification of the physical hardware it will be recorded in HKEY_LOCAL_MACHINE and will affect all users. (See section 4.3.)

Question 16

Answer: A, C

Swapping the mouse buttons is simply a preference that the administrator has. Because this option is applied when the administrator is logged on, the settings are placed in HKEY_CURRENT_USER (which at the time of making the changes would be the administrator’s settings). The settings would be applied only when the administrator was logged on. (See section 4.3.)

Question 17

Answer: B

The names NTUSER.DAT and NTUSER.DAT.LOG can be used by all users of the system, because the system can maintain a separate set of the files in a unique directory for each user below \%SYSTEM_ROOT%\PROFILES\%USERNAME%. (See section 4.3.)

Answer B is correct. The NTUSER.DAT and NTUSER.DAT.LOG files represent the local user’s profile.

Question 18

Answer: C

The key to the question is in the fact that the administrator logged on to the system before rebooting and using the LastKnownGood control set. Logging on to the system would have updated the LastKnownGood control set to contain the improper parameters. (See section 4.3.)

When the administrator logged on successfully at the server, the Last Known Good configuration was saved, regardless of the fact that the configuration was incorrect.

Question 19

Answer: B

Although the emergency repair disk can be used to restore the configuration, and then only if it has been updated since the network card was installed, it is not the fastest method. The fastest and simplest method would be to simply boot using the Last Known Good configuration. (See section 4.3.)

Question 20

Answer: C

The failure of one service or driver, such as the network card, could cause a chain of dependency service failures. User rights would not affect the start of services, because they start as part of the system. (See section 4.6.)

Chapter 5, “Managing Local File Systems”

Question 1

Answer: A, B

X86-based computers can use either FAT or NTFS partitions to house the system or boot files. Because the NTLDR file on x86-based systems initializes the mini-file system drivers for both FAT and NTFS, the file can reside on either a FAT or NTFS partition. System files, however, may not reside on a stripe set or volume set. (See section 5.7.3.)

Question 2

Answer: A, B

The key is to first remember that the files will not be directly retrieved by the DOS-based clients. They simply make requests to the file server. Because the file server is running on top of NT all of these file systems can be accessed. The second issue is that only FAT and NTFS are file systems that can be applied to a hard disk; CDFS is for CD-ROMs. (See section 5.2.1.)

Question 3

Answer: D

When a file or directory exceeds the standard 8.3 naming conventions it receives both a long and short filename. The filename 1997AC~1.doc was derived by taking the first six valid characters of the filename (1997AC), appending a unique ID (~1, ~2, and so on) and finally attaching the extension (.doc). (See section 5.2.)

Question 4

Answer: A, C

Answer A is correct. A RISC-based computer does not utilize NTLDR and therefore does not initialize the mini-file system drivers required on boot to access NTFS as its system partition. Other partitions on a RISC-based computer can be either DOS or NTFS. (See section 5.2.1.)

Answer C is correct. Multi-booting into DOS and NT will require the system partition to be FAT. If the system partition were later converted to NTFS it would be inaccessible to DOS. (See section 5.2.1.)

Question 5

Answer: A, B, C

When a folder is moved to a different partition it does not maintain its ACL. Also, the original share would no longer be valid because it would point to a location that no longer exists. (See section 5.4.)

Answer A is correct. A new ACL would have been generated based on the ACL of the folder on the new partition it was made a subdirectory of.

Answer B is correct. The move would delete the original folder and corresponding share, so it would need to be re-created.

Answer C is correct. Because the share is being re-created it will initially have permissions that allow Everyone Full Control.

Question 6

Answer: C

Although the share has given administrators Full Control, the NTFS security will not allow the administrators access. If a user is a member of both the Administrators and Accounting groups, the effective rights would be Change. (See section 5.4.)

Question 7

Answer: B

Full Control in a directory grants the hidden right File Delete Child, which enables you to delete files to which you otherwise would not have permission. This is required for POSIX compliance.

You might have Delete—or Change, which encompasses Delete—on a directory level, but this is the right to delete the directory itself, not the files within it. (See section 5.4.)

Question 8

Answer: C, D

The question is testing your knowledge of the Take Ownership right. If you are the owner of a file or directory you have the ability to change the permissions. (See section 5.4.1.)

Question 9

Answer: B

All users, including guests, are members of the group Everyone. It is not uncommon to have individual file rights that exceed the rights against the directory itself. (See section 5.4.1.)

Question 10

Answer: C

The default ACL is designed to be most permissive. This is why it is often necessary to remove the Everyone group from the ACL when assigning restrictions. (See section 5.4.2.)

Question 11

Answer: A

Only one partition can be marked as active at one time. The active partition will be the one from which the system will attempt to boot. Remember that the partition from which NT boots is called the system partition. (See section 5.7.)

Question 12

Answer: A

A volume set requires from 2 to 32 segments; however, the segments do not need to all come from separate physical disks. A single volume set can span from 1 to 32 physical disks. (See section 5.7.3.)

Question 13

Answer: D

An equal amount of space is taken from each of the areas making up the stripe set. You might think that the answer should be C, or 300MB, but a stripe set can be created from between 2 and 32 disks. By utilizing only the two disks with the largest common space between them you can create a stripe set of 400MB. (See section 5.7.4.)

Question 14

Answer: A, D

A single physical drive can contain a maximum of four partitions. They can be either four primary partitions or three primaries and one extended. (See section 5.7.)

Question 15

Answer: C

Either file system can be used when the volume set is created from free space. However, in order to extend the volume set it must be NTFS. (See section 5.7.3.)

Question 16

Answer: B, C, D

When any disk that is a member of a volume set fails, the data across the entire volume set is lost. Because the scenario did not specify whether there was sufficient free space on the remaining drives on which to re-create the volume set it is necessary to replace the failed drive. The option to regenerate the volume set does not exist, so re-creating the volume set will be necessary. After the volume set has been re-created the data must be restored from backup. (See section 5.7.3.)

Question 17

Answer: A

A cluster size can be specified up to 128KB when formatting a partition from the command prompt. If a cluster size of 64KB was specified, a 128KB file could be stored using only two clusters versus requiring 32 4KB clusters. (See section 5.7.4.)

Question 18

Answer: A, C

The sole purpose of stripe sets without parity is to increase disk performance. Only NTFS volume sets can be extended. Watch out for answers like this one on the exam where the selection is offered in the negative. (See section 5.7.4.)

Question 19

Answer: C

You can use the Disk Administrator to assign a different drive letter to a CD-ROM drive. The change is immediately put into effect unless the drive is in use. (See section 5.7.1.)

Question 20

Answer: C

The existing 100MB FAT partition could be converted to NTFS. After the data partition is NTFS it could be extended over the two areas of free space to form a single volume set of 700MB. Because the 200MB NTFS partition contains system/boot files, it cannot become part of a volume set. (See section 5.7.2.)

Chapter 6, “Data Protection: Fault Tolerance and NT Backup”

Question 1

Answer: All of the above. Although not all of the levels provide fault tolerance, they can all be run on NT Server 4.0 regardless of the role they have in the domain. (See section 6.2.1.)

Question 2

Answer: B, D

Disk duplexing gains additional fault tolerance over disk mirroring by having two controllers. (See section 6.3.)

Question 3

Answer: C

You would need to restore the last full backup and then apply each of the successive incrementals. (See section 6.6.1.)

Question 4

Answer: A, D

Stripe sets with parity provide excellent read capabilities along with fault tolerance. Although disk mirroring will increase the read capacity to a certain extent, the gain is offset by a decrease in write speed. (See sections 6.3 and 6.4.)

Question 5

Answer: C

Stripe sets are the fastest disk configuration; however, they do not provide fault tolerance. (See section 6.4.2.)

Question 6

Answer: A, B, C

These files are the only required files. NTOSKRNL.EXE is located on the BOOT partition and NTBOOTDD.SYS is only required in the case of having a SCSI controller where the BIOS is not enabled. (See section 6.5.1.)

Question 7

Answer: C

In essence, the end result is the same as a copy. Because FAT does not support the permissions, they are lost. (See section 6.6.2.)

Question 8

Answer: A

If more than one drive fails, the stripe set with parity cannot be regenerated. (See section 6.4.)

Question 9

Answer: A

By restricting access to the owner or administrator, you limit the number of people who can restore the data, thereby increasing the security level. (See section 6.6.2.)

Question 10

Answer: B

The last tape contains the tape catalog. If the last tape in a family set is damaged, the /missingtape option can be used to bypass the catalog. (See section 6.6.2.)

Chapter 7, “Managing Users and Groups”

Question 1

Answer: D

The NetLogon service provides the synchronization. The frequency of replication can be controlled through registry settings.

Question 2

Answer: A, B

The administrator account is the only account that cannot be disabled. It is also not possible to rename the administrator account.

Question 3

Answer: B

Unlike the administrator account, the guest account can be disabled; it cannot, however, be deleted.

Question 4

Answer: True

The SID is a unique number that will not be regenerated based upon user name.

Question 5

Answer: True

The SID remains with the user account until it is deleted. This is helpful if the person in a particular job function changes.

Question 6

Answer: A

AGLP is designed around the ability to place global groups into local groups.

Question 7

Answer: A, B

Both global groups and users from trusted domains can be made local group members.

Question 8

Answer: D

A global group can contain only global users from the same domain.

Question 9

Answer: A, B

Auditing of the printer requires auditing of file and object access to report print jobs.

Question 10

Answer: A, D

Modifying the computer will affect HKEY_LOCAL_MACHINE, and modifying the user will change HKEY_CURRENT_USER.

Question 11

Answer: C

Users can be restricted to selected Windows-based applications in the policy settings.

Chapter 8, “Managing Network Resources”

Question 1

Answer: B, C, D

Remember that members of the Domain Admins global group are empowered by being members of the Administrators local group. (See section 8.2.)

Question 2

Answer: B

Only the Server service is required. (See section 8.2.2.)

Question 3

Answer: B

Permissions applied to the Network group apply only when you are gaining access from across the network. (See section 8.2.3.)

Question 4

Answer: A

The Server Manager allows an administrator to manage both local and remote share points. (See section 8.6.1.)

Question 5

Answer: D

The user’s effective rights can be blocked by NTFS settings. (See section 8.5.1.)

Question 6

Answer: C

Pausing the Server service is the safest method because it will not disconnect currently attached users. (See section 8.6.1.)

Question 7

Answer: B

Share names ending with a $ are always hidden; however, they do not necessarily imply that a share is for administrators only. (See section 8.6.1.)

Question 8

Answer: C

Only users with the right to control shares will see directories as shared. (See section 8.6.)

Question 9

Answer: C

Remember AGLP, which means Accounts go into Global groups, which go into Local groups, and Permissions are granted using the local groups. (See section 8.3.1.)

Question 10

Answer: C

By keeping the documents in the spooler even after they have been printed, they can be reprinted if the printer jams from outside the client application. (See section 8.8.)

Question 11

Answer: B

Although it is not required that each of the printer devices be physically identical, each of the print devices must support a common driver. (See section 8.8.)

Question 12

Answer: C

Until the DLC protocol is installed on the print server, the option to print to the HP network port does not appear. (See section 8.8.2.)

Question 13

Answer: B, C, D

The print server is the proper location to load the drivers for all clients that are capable of requesting the driver to be downloaded transparently. The Windows for Workgroups clients do not support the transparent download and so must be configured independently. (See section 8.8.4.)

Question 14

Answer: A, C, D

After the printer is installed, it is necessary to remove the default permissions granted to the Everyone group and to grant permissions to the Accounting group. If No Access were granted to Everyone this would also prohibit members of the Accounting group from printing. (See section 8.9.1.)

Question 15

Answer: A, B, C

When a printer is created, Full Control is granted to Administrators, Server Operators, and Print Managers.

Question 16

Answer: C

An unavailable printer will still accept print jobs but will hold them until it is scheduled to print before sending them to the print device. (See section 8.9.2.)

Question 17

Answer: C

By printing to a printer with a higher priority, jobs will be ordered ahead of those coming from a printer configured with default settings. (See section 8.9.2.)

Question 18

Answer: C

By having the right to Manage Documents a user is capable of managing other persons’ print jobs. (See section 8.9.2.)

Question 19

Answer: D

The key is that the client is gaining access via the gateway. The client can use any protocol supported by the gateway. Finally, although the NetWare server will not pass the print driver to the client, the gateway will provide the proper driver; the driver on the gateway would have been already installed manually. (See section 8.10.1.)

Chapter 9, “NT Server 4.0 Network Clients”

Question 1

Answers: A, B, D

Both Windows 95 and Windows NT support both TCP/IP for Internet as well as NT Server 4.0 exchanges. They also support IPX/SPX compatibility to allow NetWare access. LAN Manager 2.2c client for MS-DOS comes with NetWare connection software. It is also TCP/IP compatible to connect with NT Server 4.0 and the Internet. (See section 9.2.)

Question 2

Answer: C

The Pentium 133 can be configured for Windows NT 4.0 Workstation. Adding administration tools to the NT 4.0 Workstation will allow for management of DHCP and WINS, as well as Server Manager and User Manager for Domains. (See section 9.2.3.)

Question 3

Answer: D

Only NTFS can provide support for Services for Macintosh. (See section 9.2.5.)

Question 4

Answer: A

In Windows 95 with Server Tools, Windows Explorer is used to manage all file and print sharing assignments. (See section 9.2.4.)

Question 5

Answer: A, B, C

To install Windows 95 you must have shared the Win95 subdirectory of the Clients directory. You then must create a Network Installation Disk and boot from that disk the computer on which you want to install Windows 95. There is no Windows 95 Setup Disks option in the Disk Administrator Tools. (See section 9.2.1.)

Question 6

Answer: D

Network Client Administrator is used to create Network Installation Startup disks. (See section 9.2.1.)

Chapter 10, “Optimizing Your Server”

Question 1

Answer: A, C

By obtaining baseline information at nonpeak times, you can identify trends in network utilization. (See section 10.1.)

Question 2

Answer: C

The chart view provides real-time data for a quick view of the health of your server. (See section 10.2.1.)

Answer A. The Log view is best for obtaining data over time to be used for the creation of a baseline. (See section 10.2.2.)

Answer B. The Alert view provides notification when threshold parameters are exceeded. (See section 10.2.3.)

Answer D. The Report view displays summary information that is updated periodically. This allows the identification of objects that you might need to monitor further. (See section 10.2.4.)

Question 3

Answer: A

You must start DISKPERF before Performance Monitor can collect information on the physical disk. (See section 10.1.2.)

Question 4

Answer: True

A short explanation of the purpose of each counter is available from the Explain button. (See section 10.3.)

Question 5

Answer: C

After selecting the Processor object and the counter you are interested in monitoring, you must set this counter for each processor by selecting the appropriate instance. (See section 10.3.1.)

Question 6

Answer: A

Excessive paging shows up as increased physical disk activity and increased processor usage. The solution to this problem is to add more memory. (See section 10.3.5.)

Question 7

Answer: D

By setting a capture filter, you can acquire the information in which you are interested while keeping the size of the captured file small. Although you also can set a Display Filter to see the data in which you are interested, the size of your captured file would be considerably larger. (See section 10.4.1.)

Question 8

Answer: D

Performance monitor allows for the monitoring of remote computers. By configuring the alert to send you a message when the threshold has been exceeded, you will be notified promptly. (See section 10.3.11.)

Question 9

Answer: A, C, D

The Task Manager provides information on active applications and processes. In addition, it provides critical information on CPU and memory utilization. (See section 10.5.)

Question 10

Answer: B

The Process Viewer provides you with detailed information on the memory allocation of any active processes. (See section 10.5.2.)

Chapter 11, “Configuring and Optimizing NT 4.0 Services”

Question 1

Answer: B

The Dynamic Host Configuration Protocol is used to dynamically assign IP addresses used by TCP/IP. Both A and C are different protocols. (See section 11.2.)

Question 2

Answer: B

Your routers must be RFC-1542-compliant and be able to forward BOOTP packets. (See section 11.2.)

Question 3

Answer: A

IPCONFIG displays the IP address, subnet mask, and default gateway for your computer. (See section 11.2.2.)

Question 4

Answer: C

When providing redundancy, configure 75 percent of the local addresses on each DHCP server. The remaining 25 percent of the addresses can be configured on another DHCP server to provide backup address assignment. (See section 11.2.4.)

Question 5

Answer: C

As pull partners. You can set a time interval for replication by using a pull partner. This would enable you to configure replication to occur during off-peak periods of network utilization. (See section 11.4.3.)

Question 6

Answer: A

By using the WINS Manager, you could import the LMHOSTS file from the Static Mappings control dialog box. (See section 11.4.4.)

Question 7

Answer: A and D

You could use a HOSTS file located on each of the non-WINS-enabled client or use DNS. (See section 11.4.4.)

Question 8

Answer: A and B

It allows multiple computers on a network to have the same host name. It also places the responsibility for naming at the level of the organization controlling a particular domain level. (See section 11.5.)

Question 9

Answer: B

From a master name server. (See section 11.5.1.)

Question 10

Answer: D

A unique IP address for each of the virtual servers must be bound to your network adapter card. (See section 11.7.1.)

Question 11

Answer: A

Directory replication is used primarily for the replication of logon scripts, system policy files, and other commonly used information. (See section 11.6.2.)

Question 12

Answer: C

The Server Manager is used to configure directory replication. (See section 11.6.2.)

Question 13

Answer: B

The parameters for synchronization are located in the HKEY__LOCAL_MACHINE hive under System\CurrentControlSet\Service\Netlogon\Parameters.

Question 14

Answer: D

By comparing newly captured data with baseline data, you can determine the source of the traffic increase.

Question 15

Answer: C

By pointing your clients to the DNS most likely to be capable of resolving its name queries, you would decrease the need for recursive lookups.

Question 16

Answer: B

By limiting the amount of bandwidth that the IIS uses, you ensure the availability of adequate bandwidth for performing other network functions, including logon authentication.

Question 17

Answer: C

By lengthening the time between WINS replication, you reduce the amount of traffic on the WAN link.

Question 18

Answer: C

By lengthening the Interval parameter on the export server and the Pulse parameter on the import workstation, you decrease the frequency of replication resulting in decreased traffic across the WAN link.

Question 19

Answer: C

The Replication Governor determines the amount of bandwidth that can be utilized for synchronization, whereas the Pulse Concurrency determines the number of BDCs that can be simultaneously synchronized. By adjusting these two parameters, you would decrease the amount of bandwidth utilized for synchronization at any one time. However, you would lengthen the amount of time it takes to synchronize your entire domain.

Question 20

Answer: B

Generally, if NT servers or workstations are available to act as browsers, you should exclude Windows 95 clients from being master or backup browsers.

Question 21

Answer: C

Stopping unnecessary services decreases the number of registrations each computer must make with the WINS server.

Chapter 12, “Managing a Heterogeneous Environment”

Question 1

Answer: D

Installing the NWLink protocol is the only thing you must install for Windows NT machines to access client/server applications on a NetWare server.

Question 2

Answer: B

When you install Gateway Services for NetWare, NWLink is also installed.

Installing the GSNW also gives you client capabilities to the NetWare server. For this service to work, you must also install NWLink. Note that if you install the GSNW, NWLink is installed automatically.

Question 3

Answer: A

802.2 is the default protocol. (See section 12.3.3.)

Question 4

Answer: A

802.2 is the standard frame type for NetWare 3.12 and later versions. For NetWare 3.11 and earlier versions, 802.3 is the standard. (See section 12.3.3.)

Question 5

Answer: B

The GSNW acts as a translator between the SMBs used by Microsoft networks and the NCPs used on NetWare networks. (See section 12.2.)

Question 6

Answer: C

There must be a group called NTGATEWAY on the NetWare server, of which the gateway user must be a member. All security can then be set on the NTGATEWAY group. (See section 12.2.2.)

Question 7

Answer: D

Installing the NWLink protocol is the only thing you must do for NetWare clients to access client/server applications on a Windows NT server. Expect to see client/server applications listed on the exam rather than the words “client/server.” (See section 12.3.1.)

Question 8

Answer: D

In this example, the NetWare logon script was constructed to identify the Engineering group and map a virtual drive to its files. NWCONV does not migrate logon scripts. The conditional MAP instruction was overlooked by the administrator and must be reconstructed as either a logon script or profile under Windows NT. (See section 12.7.1.)

Question 9

Answer: D

The ERROR.LOG file is generated each time NWCONV is executed and explicitly lists the errors encountered while performing a migration. (See section 12.9.)

Question 10

Answer: C

The 3.1x servers have a separate and individual bindery and therefore each server has its own user and group information. Each must be identified for migration under this situation. Meanwhile, only one 4.1x server must be accessed because all user objects are specified in the NDS. (See section 12.6.)

Question 11

Answer: A

You do not need to be directly logged onto the NetWare server; the administrator simply requires the supervisor account for access. As long as NT can see the NetWare server on the network and the administrator has supervisor access, you can execute NWCONV. (See section 12.4.)

Question 12

Answer: B

The Modify button on the File Options dialog box allows the administrator to define new paths and share names. Mapping files are not constructed or used during file migration. (See section 12.7.1.)

Question 13

Answer: False. Supervisor access is required.

Question 14

Answer: True. If NetWare defaults are migrated, they are expressed as individual restrictions to Windows NT and thus supersede the domain account policy. (See section 12.2.2.)

Chapter 13, “RAS Dial-In Solutions”

Question 1

Answer: B

NT Server can support up to 256 simultaneous questions. NT Workstation can support only one. (See section 13.2.)

Question 2

Answer: A, B, and E

Serial Line Interface Protocol (SLIP), Point-to-Point Protocol (PPP), RAS Protocol, and NetBIOS Gateway are the supported remote access protocols. (See section 13.3.1.)

Question 3

Answer: A and C

NetBEUI, TCP/IP, and IPX are supported networking protocols. (See section 13.3.2.)

Question 4

Answer: A and C

There are two steps to this. First, you must decide the range of IP addresses you want to assign and enter the values in the RAS Server TCP/IP Configuration dialog box (see Figure 13.7). You must also exclude this range of addresses from the scope on the DHCP server, if the ranges you selected fall within that scope. (See section 13.3.2.)

Question 5

Answer: B and D

The two best things you can do to increase the security of dial-in access are to set predetermined callback numbers and enable auditing of your resources. Note that the unified Windows NT logon authentication provides the best security. This feature is enabled by default. (See section 13.4.)

Question 6

Answer: A

You must configure both the RAS Server and the remote client machine with the Point-to-Point Tunneling Protocol. This creates a secure connection by using one protocol to tunnel through the TCP/IP protocol of the Internet. (See section 13.3.2.)

Question 7

Answer: C

Although you could use the Dial-Up Networking Monitor utility to see a snapshot of the current connections and connection times, it probably is better to create a log by using Performance Monitor. This will enable you to see specifically how many of your incoming lines are utilized and how much each is being used. (See section 13.6.2.)

Question 8

Answer: B and C

The most likely problem is that this user’s user account has not been granted dial-in access. It also is possible that there is a problem with the user account itself, such as an incorrect password or invalid logon times set. (See section 13.8.1.)

Chapter 14, “NT Server Troubleshooting Guidelines”

Question 1

Answer: A, C, D, E

These four files are necessary for you to boot up Windows NT. You do not need the others because either they are not necessary or pertain only to RISC-based machines. Read the question carefully. (See section 14.3.)

Question 2

Answer: B, E, G

You need these three files to boot up Windows NT. The rest of the files pertain only to Intel-based machines. (See section 14.3.)

Question 3

Answer: B, E

You can copy the file from a boot disk created for this server or you can replace it by performing an emergency repair. The BOOTSECT.DOS file is computer-specific, so you cannot copy it from another server. (See section 14.3.)

Question 4

Answer: D

BIOS is not on the SCSI adapters, and so the ARC name begins with scsi. Windows NT is on the first controller. Numbering starts with 0, and so scsi is (0). Because it is on the first hard disk and the disk is SCSI, disk is (0). The rdisk only pertains when you use multi, so you don’t need it, and thus it is (0). Partition numbering begins with 1, not 0. Because Windows NT is on the second partition, it is (2). (See section 14.3.2.)

Question 5

Answer: B

Although an emergency repair would fix the problem (if you did not update after you installed the drive), the easiest way is to use the Last Known Configuration option, which would have removed the CD-ROM configuration. (See section 14.3.3.)

Question 6

Answer: A

The Last Known Good was saved when you logged on. Therefore, it will not remove the new configuration information. You can’t get the computer to run long enough to run the backup utility, and so your only resort is to perform an emergency repair. (See section 14.3.3.)

Question 7

Answer: D

Using a boot disk is the fastest way to bypass missing boot files. (See section 14.3.)

Question 8

Answer: D

Using a boot disk is the fastest way to bypass missing boot files. (See section 14.3.)