Free MCSE Braindumps, the best braindumps site for certification exams like 070-290, 070-270 and more!

QUESTION 1
You are the cluster administrator for Alecnet .s T he company wants its file
services to be highly available. You need to evenly migrate all of the data on 10
dedicated file servers across an active/active cluster. The old servers will be
deployed throughout the network for various noncritical functions.
Many users have mapped drives an embedded UNIC references to the existing
servers name and share names. You need to ensure that those existing connections
continue to function properly after the migration.
The cluster is configured to have three disk groups. Cluster group, Disk group 1,
and Disk group 2. You have created File Share resources in the appropriate disk
groups by using the existing share names.
You want to redirect user access to the new Server cluster without changing the
settings on the user's computers. What should you do?
A. For the Cluster group, create 10 IP Address resources and 10 Network Name
resources that use the original service names.
Rename the old servers.
B. For Disk group1, create one IP Address resource and five Network Name resources
that use the original Server names.
For Disk group2, create one IP Address resource and five Network Name resources
that use the original Server names.
Rename the old servers.
C. For Disk group1, create one IP Address resource and five Network Name resources
that use the original Server names.
For Disk group2, create one IP Address resource and five Network Name resources
that use the original Server names.
Change the IP Addresses on the old servers.
D. For the Cluster group, create 10 IP Address resources that have the same IP addresses
as the original servers.
For Disk group1, create Network Name resources that have unique names.
For Disk group2, create Network Name resources that have unique names.
Change the IP Addresses on the old servers.
Answer: B
Explanation: To ensure that users can still use their old settings with the new Server
cluster it is important that the new disk groups IP Addresses resource and Network Name
resources use the original server name. Then the old servers need to be renamed.
Incorrect answers:
A. To ensure that the file service is highly available you must create at least 2 disk
groups and not just provide the Cluster group with 10 IP address resources and 10
Network Name resources.
C. To ensure there are no problems with the new cluster, the old servers we need to be
renamed and not just have their IP addresses changed.
D. To ensure there are no problems with the new cluster, the old servers we need to be

renamed and not just have their IP addresses changed.
http://www.microsoft.com/windows2000/techinfo/planning/server/clustersteps.asp
QUESTION 2
You are the administrator of a Windows 2000 Advanced Server cluster. The cluster
provides file and print share resources for Alecnet .s Y ou add 350 cluster File
Share resources to support a new research project.
After adding the resources, you decide to test the cluster for failover. During the
test, the users of the new shares are denied access to their files after failover.
You need to ensure that users can access the shares after failover. What should you
do?
A. Use Windows Explorer to reassign share-level permissions.
B. Use the cluster administrator to reassign share-level permissions.
C. Use dynamic file shares to remove and then re-create the cluster File Share resources.
D. In the Advanced File Share Properties dialog box, clear the Hide subdirectory
shares check box.
Answer: B
Explanation: In order for users to access the shares after a failover the share-level
permissions must be reassigned. In order to change the share-level permissions you will
need to do so from the cluster administrator account.
Incorrect answers:
A. Windows Explorer does not have the required permissions to reassign share-level
permissions.
C. Dynamic file shares is not one of the three File Share resources types: basic, share
subdirectories, and DFS root.
E. This action will not have the desired effect.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, page 94-106.
QUESTION 3
You are the administrator of a new Windows 2000 Advanced Server cluster. You
want to configure the cluster to support both DHCP service resource and a WINS
service resource. You create a Services Resource group. The group contains the
resources shown in the following table.
DHCP Service resource
WINS Service resource
Physical Disk resource
IP Address resource
The services resource group fails when you try to bring it online. What should you
do to bring the group online?
A. Add a Network Name resource to the Services Resource group.
B. Add one Generic Service resource to the Services Resource group.

C. Add a second IP Address resource to the Services Resource group.
D. Add two Generic Service resources to the Services Resource group.
Answer: A
Explanation: Whenever you create a resource on a cluster you need to include a
Network Name resource. The Network Name resource is used to assign a name to a
resource.
Incorrect answers:
B. The Generic Service resource is used to support a cluster-unaware service and to
provide basic cluster functionality.
C. A second IP Address Resource will not allow the group online.
D. Two Generic Service resources will not allow the group to come online.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 97,
164, and 177.
QUESTION 4
You are the Cluster Administrator for Alecnet .s Y ou need to create a new
cluster. First, you install and configure the cluster storage. Next, you configure two
network adapters each in two Windows 2000 Advanced Server computers named
Server1 and Server2. Server1 and Server2 are member servers in Windows 2000
domain
and will be configured as cluster nodes.
On Server1, you create a local cluster named Cluster Account. You enable the User
cannot change password option. You place the Cluster Account user in the
Administrators group. You assign the necessary rights to the Cluster Account user.
You install the Cluster service on Server1. You specify the Cluster Account user as
the account under which the Cluster service will run. You install the Cluster service
on Server2, specifying the same user name and password as you did for Server1.
You receive an error message stating that the Server cannot join the cluster. You
want to ensure that Server2 can join the cluster. What should you do?
A. For the Cluster Account user on Server1, enable the Password never expires option.
B. Create a domain user under which the Cluster service will run on Server1 and
Server2.
C. For the Cluster Account user on Server1, disable the User cannot change password
option.
D. Add the Cluster Account user on Server1 to the Administrators group on Server2. On
Server2, assign the necessary rights to the Cluster Account.
Answer: B
Explanation: In order of Cluster Service to run it requires a domain user account under
which the Cluster Service can run. In this question the Cluster account was created as a
local account and not a domain account.
Incorrect answers:

A. It is recommended that the Password never expires option is selected but it is not
required. When you do no select this option you must ensure that you renew the
password and update the cluster service configuration on each node before the passwords
expires.
C. Microsoft recommends that the Cluster user account have the User cannot change
password selected.
D. The Cluster user account must be a domain user account. Adding the Cluster
Account user on Server1 to the Administrators group on Server2. On Server2, assign the
necessary rights to the Cluster Account will not correct the problem.
http://www.microsoft.com/windows2000/techinfo/planning/server/clustersteps.asp
QUESTION 5
You are the administrator of Windows 2000 Advanced Server computer. You need
to prepare the Server for use as a cluster node. The Server contains a SCSI
controller and three internal 4.3 GB hard disks. You plan to attach the Server to an
external storage subsystem that contains eight 18.3-GB hard disks.
You connect the three 4.3-GB hard disks to the SCSI controller's internal
connector. Next, you connect the external storage subsystem to the SCSI controller's
external connector. When you start the Server, you receive numerous BIOS error
messages that indicate errors on the SCSI bus.
Before installing the Cluster service, you need to be able to access the internal and
external disks attached to the Server. What should you do?
A. Replace the SCSI controller.
B. Replace the external SCSI cable.
C. Set the SCSI ID to 6.
D. Install an additional SCSI controller in the Server.
Connect the external drive array to the new controller's external connector.
E. Disable the on-board SCSI termination on the SCSI controller.
F. Connect a SCSI terminator to the SCSI-out connector on the external storage
subsystem.
Answer: D
Explanation: When you use SCSI hardware in a cluster there is a requirement for two
SCSI adapter cards. One card for the internal disks and one for the external disk
Incorrect answers:
A. If you replaced the SCSI controller and still connect internal and external drives the
same problem will still exist.
B. As the errors are in the BIOS the problem could not be related to the external SCSI
cable.
C. Each SCSI device will require it's own SCSI ID and changing just one of them will
not achieve the desired result.
E. The on-board SCSI termination on the SCSI controller only needs to be disabled when
you terminate externally.

F. As the errors are in the BIOS the problem could not be corrected by connecting an
external SCSI terminator.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 44-46.
QUESTION 6
You are the cluster administrator for Alecnet .s Y ou need to configure the two
Windows 2000 Advanced Server computers for cluster nodes. You need to
implement fault-tolerance methods for the local storage. You also need to implement
fault-tolerance methods for the external storage subsystem.
Each Server contains two 9.3 GB hard disks. The external storage subsystem
contains eight 10-GB hard disks. You need to ensure that the cluster can tolerate the
failure of a single hard disk in either Server and in the external storage subsystem.
Which two fault-tolerance methods should you use?
A. Configure the hard disks in each Server as Windows 2000 striped volumes.
B. Configure the hard disks in each Server as Windows 2000 RAID 5 volumes.
C. Configure the hard disks in each Server as Windows 2000 mirrored volumes.
D. Configure the hard disks in the external storage subsystem to use a hardware RAID 5
array.
E. Configure the hard disks in the external storage subsystem as Windows 2000 striped
volumes.
F. Configure the hard disks in the external storage subsystem as Windows 2000 RAID 5
volumes.
Answer: C, D
Explanation: In order to provide fault tolerance to the internal disk drives they will need
to be configured as mirrored volumes. With mirrored volumes when data is saved it is
written on both drives. To offer the best fault tolerance to the external storages device
RAID 5 is the best but this is done at the expense of performance.
Incorrect answers:
A. Stripped volumes do not provide fault tolerance.
B. RAID 5 requires at least 3 disks to function. There are only two internal disks.
E. Clustering service does not support the software implementation of RAID. RAID
needs to be implemented at the hardware level.
F. Clustering service does not support the software implementation of RAID. RAID
needs to be implemented at the hardware level.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 46-48.
QUESTION 7
You are the administrator of two Windows 2000 Advanced Server computers
named Server1 and Server2. You install two network adapters in each Server and
then connect them as shown in the following diagram.

You turn off Server2 and begin installing the Cluster service on server1. In the
Cluster service Configuration wizard, you designate Client access only for Adapter
1B. the wizard does not allow you to configure Adapter 1A, however.
You verify that the crossover cable between Adapter 1A and Adapter 2A is
connected properly.
You need to designate Internal cluster communications only for Adapter 1
A. What
should you do?
A. Start Server2, and enter the BIOS setup screen.
Run the Cluster Service Configuration Wizard on Server1.
Designate Adapter 1A for Internal cluster communications only.
B. Start Server2, and start Windows 2000 Advanced Server, but do not log on.
Run the Cluster Service Configuration Wizard on Server1.
Designate Adapter 1A for Internal cluster communications only.
C. Disconnect crossover cable from Adapter 1A.
Run the Cluster Service Configuration Wizard on Server1.
Designate Adapter 1A for Internal cluster communications only.
Connect the crossover cable to Adapter 1A.
D. Disconnect crossover cable from Adapter 1A.
Run the Cluster Service Configuration Wizard on Server1.
Designate Adapter 1A for All communications.
Connect the crossover cable to Adapter 1A.
Use the Cluster administrator to designate Adapter 1A for Internal cluster
communications only.
Answer: A
Explanation: As a Windows 2000 Media sense prevents you from configuring Adapter
1

A. There are two ways to resolve this problem: during the cluster installation, the other
node must be booted but stopped at the Boot menu so that there is power to the network
adapter but it does not have write access to the shared disks or edit the registry.
Incorrect answers:
B. Server 2 had to be stopped at the boot menu. By allowing Server 2 to boot fully you
are likely causing more problems and not fixing any.
C. Even with the crossover cable disconnected, Media sense will consider the adapter
down. Therefore you will not be able to configure it.
D. Even with the crossover cable disconnected, Media sense will consider the adapter
down. Therefore you will not be able to configure it.
http://support.microsoft.com/default.aspx?scid=KB;EN-US;q242430&
QUESTION 8
You are the administrator of a Windows 2000 Advanced Server cluster. The cluster
nodes are two member servers in the same domain. These nodes are designated
NodeA and NodeB.
You want to delegate control of the cluster to the global group cluster operations.
You do not want the members of the Cluster Operations group t have
Administrator permission on the cluster nodes.
You want to accomplish the following goals:
Enable members of the Cluster Operations group to change the location that is
used for cluster log file.
Enable members of the Cluster Operations group to add and remove resources.
Enable the members of the Cluster Operations group to bring resources online
and take resources offline.
Enable the members of the Cluster Operations group to stop and start the
Cluster service.
You take the following actions:
Create a new domain local group named Node Operations.
Add the Cluster Operations group to the Node Operators group.
Use the Cluster Administrator tool to grant the Node Operations group
permissions to administer the cluster.
Which result or results do these actions produce?
A. Members of the Cluster Operations group can change the location that is used for
the clustering log file.
B. Members of the Cluster Operations group can add and remove resources.
C. Members of the Cluster Operations group can bring resources online and take
resources offline.
D. Members of the Cluster Operations group can stop and start the Cluster service.
Answer: B, C
Explanation: As you granted the Nodes Operation group permissions to administer the
cluster the will be able to manage resources (add/remove and bring on/offline).
Incorrect answers:

A. As you only granted the Nodes Operation group permissions to administer the cluster
they are not able to change the location of the log file.
D. As you only granted the Nodes Operation group permissions to administer the cluster
they are not able to stop and start the Cluster service.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 86-90.
QUESTION 9
You are the administrator of a Windows 2000 Advanced Server cluster. The cluster
consists of two nodes that are designated NodeA and NodeB. You configure several
groups and resources, but you cannot bring any of the Network Name resources
online.
Windows 2000 is installed in the default location on both NodeA and NodeB, and the
Quorum resource is located on logical drive Q in the default path. You attempt to
view the log of cluster activity, but the log does not appear.
You want to be able to review as much detail of the cluster activity as possible so
that you can identify the problem.
What should you do?
A. Increase the size of quorum log.
B. Increase the size of the System event log.
C. Configure the Application event log to not overwrite events.
D. On both nodes, configure a system environment variable named ClusterLog with
a value of C:\winnt\cluster\cluster.log.
Answer: D
Explanation: The quorum log file does not appear. A likely cause of this problem could
be a nonexistent or a misconfigured environment variable.
In Microsoft Cluster Server, you can turn Cluster Logging on as a way of troubleshooting
problems with the clustering software. To enable Cluster Logging, there are two variables
you need to place in the Windows NT environment variable list. These variables are
ClusterLog and ClusterLogLevel.
Procedure adding System environment variables,:
1. In the System tool in Control Panel, click the Environment tab.
2. Click an entry in the System Environment Variable window.
3. Click the Variable and Value text boxes to clear them.
4. Type ClusterLog in the Variable box, type path\cluster.log in the
Value box, and then click Set, where path is the drive and folder to
store the Cluster Server log file.
NOTE: The recommended path is %SystemRoot%\Cluster. For
example, C:\WinNT\Cluster\Cluster.log. Cluster logging is enabled
by default in Windows 2000, and this is the default path.
5. Type ClusterLogLevel in the Variable box, type 0, 1, 2, or 3 in the
Value box, and then click Set.
6. Click OK.

7. Restart your computer for Cluster Server to read the variables
correctly.
Reference: How to Enable Cluster Logging in Microsoft Cluster Server, Microsoft
Knowledge Base Article - Q168801
Incorrect answers:
A. The scenario states that the quorum log does not appear. Increasing the size of the
quorum log would not make it appear.
B. The System event log does not contain a copy of the Quorum log file nor does it
contain cluster activity. This will not allow you to see as much detail as possible of the
cluster activity.
C. The Application event log cannot overwrite cluster activity.
QUESTION 10
You are the administrator of a Windows 2000 Advanced Server cluster. The cluster
contains two nodes that are designated NodeA and NodeB. A third Windows 2000
Server computer on your network is named BackupSvr.
You need to implement a disaster recovery plan. You decide to use Windows 2000
Backup to perform the backup operations.
You want to accomplish the following goals:
Backup the 80-GB cluster storage subsystem to a single tape daily.
Complete the backup even if a single node of the cluster fails.
Backup the cluster information databases on NodeA and NodeB daily.
Alecnet us s es 100-GB tape drives and 30-GB tape drives. You need to decide
which tape drive you will use for the backups. You need to use the smallest possible
tape drives for NodeA, NodeB and BackupSvr. You can attach up to two tape drives
to each Server.
To answer click the Select and Place button. Drag the appropriate tape drives to
appropriate answers.
Answer:

Explanation: As the cluster information databases on NodeA and NodeB will not be
that large a 30 GB tape drive will suffice. Each node requires its own tape drive in case
one of the nodes fails. In addition, in order to backup the 80 GB cluster storage
subsystem to one tape then the 100 GB tape drive is required.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 237-
244.
QUESTION 11
You are the administrator of a Windows 2000 Advanced Server cluster named
Cluster1. Cluster1 consists of two nodes that are designated NodeA and NodeB. Each
node has a 8 GB of local disk storage. The cluster storage is composed of eight
10 GB hard drives in a hardware RAID 5 array.
The local disk storage on both nodes consists of a single volume, which is shared as
C$. The cluster storage is shared as Z$.
You need to implement a disaster recovery strategy that includes regular backups
of the cluster. You decide to use Windows 2000 Backup.
You want to accomplish the following goals:
Perform a full daily backup of the cluster storage, even if one node in the cluster
fails.
Perform a full daily backup of the System State data on the node that owns the
Quorum resource.
Perform a full daily backup of the local disk storage on the node that owns the
Quorum resource.
Save all of the backed-up data to a single tape.
You take the following actions:
Connect a 100-GB tape drive to NodeA.
Use Windows 2000 backup to create a backup selection file that includes
\\Cluster1\Z$ and \\NodeA\c$. Save the backup selection file in
C:\ClusterBack.bks on NodeA.

Use the Scheduler service on BackupSvr to launch Windows 2000 Bcakup daily.
Use the following command line parameters:
ntbackup backup systemstate c:\clusterback.bks
Which result or results do these actions produce? (Choose all that apply)
A. Cluster storage is fully backed up daily, even if one node in the cluster fails.
B. The System State data on the node that owns the Quorum resource is fully backed
up.
C. The local disk storage on the node that owns the Quorum resource is fully backed
up daily.
D. All of the backed-up data is saved to a single tape.
Answer: D
Explanation: All backup-up data is saved to a single tape.
Incorrect Answers
A:: You save the backup selection on Node
A. If node a fails, the backup command would
not run.
B, C:: Only node a is backed up (including the system state). If NodeA fails, NodeB
becomes the owner of the quorum resource. But NodeB\C$ is not included in the
backup. So it cannot be guaranteed that the system state and the local disk storage
of the quorum owner will be backed up, if the owner is NodeB.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 236-
241.
QUESTION 12
You are the administrator of a Windows 2000 network. The network contains a twonode
cluster that functions as a file Server for 500 users in the manufacturing
department. The cluster consists of two nodes that are designated NodeA and
NodeB. Both cluster nodes are single-processor, Hardware Compatibility List
(HCL)-compliant systems. Both nodes contain 128 MB of RAM, 500 MB of available
local storage and 3.5 GB of available cluster storage. The CPU utilization for each
node average 70 percent.
The accounting department requests that you add a new application to the cluster.
The new application will support 500 additional users. The application requires
multiprocessing. It also requires 128 MB of RAM, 750 MB of local storage space for
installation, and 1.5 GB of cluster storage. The application is cluster unaware, but it
meets the requirements for a cluster application.
You want NodeA to be the preferred owner for all File Share resources. You want
NodeB to be the preferred owner for the accounting Application resource.
You want to accomplish the following goals:
Reduce processor utilization on both nodes.
Optimize NodeB for running the accounting application.
Optimize NodeA for use as a file Server.
Enable NodeA and NodeB to provide for failover support for each other.

You take the following actions:
Add 512 MB of RAM to each node.
Add another processor on NodeB.
Upgrade the processor on NodeA.
Add 2 GB of local storage on NodeB.
Which result or results do these actions produce?
A. Processor utilization on both nodes is reduced.
B. NodeB is optimized for running the accounting application.
C. NodeA is optimized for use as a file Server.
D. NodeA and NodeB can provide for failover support for each other.
Answer: A, B
Explanation:
A: Processor utilization on both nodes is reduced.
B: NodeB is optimized for running the accounting application.
Incorrect answers:
C: NodeA was not optimized with the processor upgrade.
D: No steps has been done to provide failover support for both nodes.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 40-41.
QUESTION 13
You are the administrator of a two-node Windows 2000 Advanced Server cluster
that serves as an application Server for Alecnet .s T he cluster runs a
resourceintensive, critical application. You want to be able to run diagnostics on the
application and the cluster nodes after failover. Business hours for Alecnet asr e
8:00 A.M. to 6:00 P.M. and the application must be available during these hours.
You want to configure the failback policy for the application. What should you do?
A. Configure the Application resource group properties to Prevent failback.
B. In the Application resource group properties, configure both nodes as preferred
owners.
C. Configure the Application resource group properties to Allow failback between 1
and 4 hours.
D. Configure the Application resource group properties to Allow failback between 4
and 1 hours.
Answer: A
Explanation: If you want to run diagnostics on the application and cluster nodes after a
failover it is necessary that you prevent failback.
Incorrect answers:
B. If both nodes were preferred owners it is very likely that the application could become
unavailable during working hours.
C. It is possible with these setting that the application could become unavailable during

work hours.
D. It is possible with these setting that the application could become unavailable during
work hours.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 220-
222.
QUESTION 14
You are the Cluster Administrator for Alecnet .s Y ou need to create a cluster that
consists of two Windows 2000 Advanced Server computers that are designated
NodeA and NodeB. Your network infrastructure includes Windows 2000 WINS
servers, Windows 2000 DNS servers that have dynamic updates enabled for all
zones and Windows 2000 DHCP servers that have client DNS registration enabled.
Client computers in the network include Windows 2000 computers, Windows NT
Workstation 4.0 computers and Windows 95 computers. These client computers are
dispersed over multiple subnets. All of the client computers receive DNS and WINS
Server assignments by using DHCP and do not require additional manual
configuration. All of the Administrators on the network use Windows 2000 client
computers.
You want the cluster to provide high availability for a cluster-aware Windows
sockets application and for file shares.
For each service you configure a resource group that contains a Network Name
resource. You manually configure the IP address, the subnet mask and the default
gateway TCP/IP properties on NodeA and NodeB.
You install the cluster service on NodeA and NodeB. After the installation, none of
the users can connect to any of the virtual servers by name.
You want to accomplish the following goals regardless of which node owns the
resources.
Provide the ability to administer the cluster remotely by using the cluster name.
Enable all of the users to access the Windows sockets application by using a
virtual Server name.
Enable all of the users to access file shares by using a virtual Server name.
Provide the ability to dynamically register the virtual Server names.
You take the following actions.
Configure NodeA as a WINS client computer.
Configure NodeB as a WINS client computer.
Which result or results do these actions produce? (Choose all that apply)
A. The cluster can be administered remotely by using the cluster name.
B. All the users can access the Windows sockets application by using a virtual
Server name.
C. All the users can access file shares by using a virtual Server name.
D. The virtual Server names can be registered dynamically.
Answer: A, B, C, D
Explanation: By configuring NodeA and NodeB as a WINS client computer you

accomplished all of the goals that you wanted to achieve.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 174-
178.
QUESTION 15
You are the Cluster Administrator for Alecnet .s Y ou need to install the Cluster
service on two servers named Server1 and Server2. Both servers have one network
adapter that is attached to the company LAN and a second network adapter that is
attached to a private cluster network, as shown in the exhibit. (Click the exhibit
button.)
During the installation of cluster service on Server1, you provide an IP address for
the cluster and receive the following error message:
How should you configure the IP address for the cluster?
A. Configure the IP address for the cluster to be 127.0.0.1.
B. Configure the Server1 and Server2 to be DHCP client computers prior to cluster
installation.
C. Configure the IP address for the cluster to be unique IP address on the 10.0.5.0
subnet.
D. Configure the IP address for the cluster to be a unique IP address on the
192.168.5.0 subnet.
Answer: D
Explanation: It is very important that the cluster receive a unique IP address. As you
were configuring an IP address for the cluster it can be assumed that you were
configuring the Public Cluster Connection. As this is the Public Cluster Connection, it
will require a public IP address. Therefore, the IP address should be in the 192.168.5.0
subnet.
Incorrect answers:
A. This is the IP address of the local loop back. It cannot be used as a valid IP address.
B. It is highly recommended that a cluster have a static IP address. DHCP servers are
not a valid solution.
C. The public cluster connection should not have a private IP address.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 64-68.
QUESTION 16
You are the Cluster Administrator for Alecnet .s Y ou create a file and print
cluster. The cluster contains two nodes that are designated NodeA and NodeB. Both
nodes are directly connected to a SCSI storage enclosure.
The cluster is configured to have five disk groups. All five-disk groups are online on

NodeA, and none of the disk groups have errors. You decide to test failover. When
you shut down NodeA, you cannot access the cluster storage form NodeB.
You want to ensure that you can access the cluster storage from one node when the
other node is shut down.
What should you do?
A. Enable passive termination on the SCSI bus.
B. Configure all of the cluster groups to allow immediate failback.
C. Attach a Y cable and a terminator to the SCSI adapter in each node.
D. Attach a Y cable and a terminator to each SCSI storage enclosure connection.
Answer: C
Explanation: As with as SCSI devices it is important to ensure that each end of the SCSI
chain. In this case it would be best to attach a Y cable to each adapter. One part of the
cable would attach to the device and the other would be terminated. This would ensure
access to the cluster storage from one node when the other node is down.
Incorrect answers:
A. Passive termination would achieve the desired effect.
B. Failback would not allow access to the storage devices without proper termination.
D. The termination must be off the SCSI adapter on each node.
Microsoft Windows 2000 Advance Server Clustering Services Training Kit, pages 202-
206.