|
Managing and
Maintaining a
Microsoft Windows Server 2003 Environment
Microsoft Exam 70-290
Managing Windows Server 2003
Control Panel
can be used to configure hardware settings, manage user-specific
settings, and manage computer-specific settings.
Installing New Hardware
Installing a
new device to a Windows Server 2003 computer typically involves
physically connecting the device to the computer; loading the
appropriate device drivers; and configuring the device properties and
settings if required.
Note:
To be able to install a device you must be logged on as an
administrator
or as a member of the
Administrators group.
When you install a
Plug and Play
device, Windows Server 2003 automatically configures the device so
that it works properly with the other devices that are already
installed on the computer. This includes assigning the appropriate
system resources, such as Interrupt Request (IRQ)
line number, Direct Memory Access (DMA)
channels, Input/Output (I/O)
port addresses and
Memory Address
ranges, to the device. Each device must be assigned a unique system
resource or the device will not function properly. When you install a
non-Plug and Play, or a legacy device, you must use the
Add/Remove Hardware Wizard.
If Windows 2000 does not detect the device you must configure the
system resources for the device manually. You can assign system
resources to the device in Device Manager.
Note:
Some old
legacy ISA
devices require the use of a specific IRQ number that Windows Server
2003 may have assigned to a Plug and Play device. In this event you
should
reserve
the IRQ that is required by the device in your
system BIOS.
Windows Server 2003 will then assign another an IRQ to the Plug and
Play device that was using the IRQ that you have reserved.
Note:
When you install Windows Server 2003 on a new computer that does not
have a standard
Hardware Abstraction Layer
(HAL) or a
RAID
device that is
not detected
by the Windows Setup program, you must install the drivers for these
devices during the
text portion
of the Windows Server 2003 Setup program.
Using Driver Signing
Some device
drivers and some applications overwrite existing operating files as
part of their installation process. These files can cause system
errors that are difficult to troubleshoot. Microsoft has simplified
the tracking and troubleshooting of altered files by digitally signing
the original operating system files and allowing you to verify these
signatures.
Configuring Driver Signing
You can configure how the computer responds to unsigned files on
HARDWARE
tab of
SYSTEM.
Here you can configure one of three responses:
Ignore
allows any files to be installed regardless of whether they are
digital signature or not.
Warn
displays a warning message before allowing the installation of an
unsigned file. This is the default option.
Block
prevents the installation of unsigned files.
When you change the default Driver Signing option, you must select the
Apply setting as system default
check box in the
Driver Signing Options
dialog box. This will make the new settings the default system
setting. If you do not select the
Apply setting as system default
check box, the settings will revert to the old setting when the
computer is next rebooted.
You will want to become familiar with the
driver signing
dialog box.
The File Signature Verification Utility
Windows Server 2003 also provides a File Signature Verification
utility,
sigverif,
that allows you to view the file's name, its location, its
modification date, its type, and its version number.
Adding Hardware
Adding Additional CPUs
When you
install Windows Server 2003 on a multiprocessor computer with only one
central processor unit (CPU), the Windows Server 2003 Setup program
will install a uniprocessor kernel on the computer. You can add a
second CPU to the computer by inserting the CPU in the motherboard,
according to the manufacturer's instructions, and restarting the
computer. On reboot, the system will detect the additional CPU and
will replace the uniprocessor system drivers with the multiprocessor
equivalents.
You can verify that the system sees the second CPU by opening the
Performance
window in
Task Manager.
The
Performance
window should show graphs for each CPU. If the
Performance
window does not display a graph for the second CPU, make sure it is
fully seated and that no changes are required in CMOS. Also, make sure
the voltage regulators or other motherboard devices are in place and
firmly seated.
Adding Removable Media Drives
Windows Server 2003 supports a new service called the
Removable Storage Manager (RSM).
The RSM service simplifies managing removable media if you have large
CD-ROM jukeboxes or a robotic tape library. In previous versions of
Windows, each disk in a jukebox or disc changer is assigned a separate
drive letter, which could be problematic when configuring large CD
libraries because there are not enough drive letters to accommodate a
25-disc library. RSM solves this problem by assigning a single drive
letter to the device itself and manipulating the CDs in the
background.
However, each
disk must be mounted by name using the RSM Mount command. This
requires that you know the logical media ID assigned to the CD by RSM.
This information is available in the RSM snap-in within the Computer
Management console.
Configuring Hard Disks
Disk Storage Types
Windows Server 2003 provides support for two types of disk storage:
basic storage,
which uses basic disks and is the standard storage type; and
dynamic storage,
which uses dynamic disks. Basic disks can be divided into up to four
partitions that can either be
primary partitions
or
extended partitions.
You can have multiple primary partitions but only one extended
partition. You can create multiple primary partitions to which enables
you to
dual boot
between Windows Server 2003 and other operating systems such as
Windows XP Professional and Windows 98. One of the primary partitions
must be set in
fdisk
as the
active partition
as the
boot files
required to start the operating systems must be located on the active
partition.
Note:
If you plan to dual boot between Windows Server 2003 and
Windows 95,
Windows 95 OSR2,
or
Windows 98
the primary partition must be formatted with the
FAT
or
FAT32
file system.
Basic disks can be converted to dynamic disks from which
dynamic volumes
can be created. Windows Server 2003 supports five types of dynamic
volumes: simple volumes; spanned volumes; striped volumes; mirrored
volumes; and striped volumes with parity.
Converting to Dynamic Disk Status
By default, all
disks in Windows Server 2003 are configured as basic disks but you can
convert them to dynamic disks. No data loss is incurred when
converting to dynamic disks, however, if you convert a disk that is
currently being accessed, such as the boot or system disks, you must
reboot the computer in order to perform the conversion.
Once you have converted to dynamic disks, you cannot covert them back
to basic disks without incurring data loss. Thus, to convert back to
basic disks, you must back up your data, and then delete the dynamic
volumes. You can then convert the disk back to basic using the disk
management MMC or the
diskpart
utility.
Simple Volumes
Simple volumes
are the default volume type on a dynamic disk. A simple volume is
created from the free space on a single physical disk and is not fault
tolerant.
Spanned Volumes
Spanned volumes
contain disk space from up to 32 physical disks. This enables you to
group different disks of the same or different sizes and access them
as if they were one disk. However, only one disk in the volume is
written to at a time. Spanned volumes provide 100 percent drive
utilization but is not fault tolerant.
Striped Volumes
Like spanned
volumes, striped volumes allow you to combine the free space from two
to 32 physical disks into one logical volume. Unlike spanned volumes,
the amount of unallocated space on each of the disks that is part of
the volume must be of the same size. Striped volumes increase both
read and write performance when accessing the volume by utilizing all
the disks at one time. Striped volumes cannot be extended.
Mirrored Volumes
Mirrored
volumes require exactly two disks that are of the same size. When
information is written to a mirror volume, the same information is
written to each disk. This provides fault tolerance and complete
redundancy for your data. Should one disk fail, you can use the
mirrored copy. Because the same information is written to two disks,
mirrored volumes provide only 50% disk utilization. Mirrored volumes
also cannot be extended.
Striped Volumes with Parity RAID-5 Volumes
Striped volumes
with parity are also called RAID-5 volumes and can be created using
three to 32 disks. They provide fault tolerance by calculating parity
information, which can be used to recreate the data on the other
disks, and writing it to a block on one disk as part of the striping
operation. Data is striped across all the disks in the volume, while
parity information is written to one disk in each stripe. The parity
information can be used to regenerate the missing data should one disk
fail. If you lose more than one disk, however, all your data will be
lost.
As with
mirrored volumes, RAID-5 volumes cannot be extended. However, RAID-5
volumes offer more efficient disk utilization than mirrored volumes.
You lose the storage space of one disk in the RAID-5 volume because it
is used for parity information.
Configuring File Systems
Windows Server 2003 supports the
FAT,
FAT32
and
NTFS
file systems. A computer can contain a combination of file systems but
each file system must be located on a separate partition or volume.
Note:
MS-DOS, Windows 95, Windows 98 and Windows Millennium Edition cannot
access data on NTFS formatted disks.
The NTFS file system used by Windows 2000 and Windows Server 2003 is
version 5.
This version of NTFS has features that were not available in NTFS
version 4 used by Windows NT 4.0. Windows NT 4.0 cannot therefore
fully support all the features of NTFS version 5. NTFS version 5
offers a number of benefits that include:
Note:
You can
convert
a disk from the FAT and FAT32 file to NTFS at any time without data
loss by using the
convert
command from a command prompt and using the
fs:/ntfs
switch. When you format the data on the disk is lost.
Encrypting File System (EFS)
Windows Server 2003 supports Encrypting File System (EFS) which allows
users to encrypt their files and folders. In Windows Server 2003,
users can encrypt files and folders on the local computer and across
the network. In addition, users can also encrypt offline files.
However, EFS is only supported on NTFS volumes. When a user encrypts a
file, only that user will be able to use the file. They can use the
encrypted file without having to decrypt the file first. EFS can be
implemented from Windows Explorer or from the command prompt using the
Cipher
command.
-
EFS
is only supported on
NTFS version 5 (Windows 2000 or 2003)
-
Compressed
files
cannot be encrypted using EFS
-
System
files
cannot be encrypted
-
Encrypted
files cannot be
shared
-
Encrypted
files or folders that are moved or copied to partitions or volumes
that are not formatted with the NTFS file system will become
decrypted
-
Files
and folders on network computers can be encrypted if you have the
necessary access
permissions
to the network computers NTFS volume and if file
encryption is enabled on the network computer.
See:
230520 - HOW TO Encrypt Data Using EFS in Windows 2000
Windows 2003
also supports encryption of offline files. This enables users to use
offline file storage while retaining the ability to protect their
files with encryption.
Encrypting Files Across the Network
File encryption and decryption requires the presence of EFS keys on
the local computer where the files reside. When a user encrypts a file
on a local desktop or laptop, EFS works with the
Microsoft Crypto Provider
to create EFS keys and to place those keys in the user's local
profile. If the user attempts to encrypt a file across the network,
EFS running at the server looks for the user's local profile at the
server. EFS cannot access keys at a user's desktop because it does not
have a security context anywhere except at the machine where it's
running. This means that the server must have a local profile for the
user that contains both the EFS public key to encrypt the file and the
EFS private key to open the encrypted file. To build the local private
key, the Protected Storage service at the server must have a copy of
the user's password hash so it can encrypt the Master key that
protects the user's private key. It obtains this information by "user
impersonation". This requires obtaining a
Kerberos session ticket
on behalf of the user to present when requesting the user's security
credentials from a domain controller. A server has two ways of
obtaining this session ticket:
It can ask the Kerberos client at the user's desktop to
obtain the session ticket and pass it over to the server. Such a
ticket would be marked as
forwardable;
or
The server can ask the Kerberos client for a
ticket-granting ticket (TGT)
that it can use to obtain its own session tickets as if the server
were the user. The TGT would be flagged as
proxiable.
However, before a server submitting forwardable and proxiable Kerberos
tickets and TGTs to create encrypted files for network users, it must
be
Trusted For Delegation.
This option is configured in the server's
Computer
object in
Active Directory.
Encrypted File Recovery
If a user leaves the company or goes on vacation, the administrator
can access the user's encrypted files by resetting the user's password
in Active Directory and then logging on as the user. Windows Server
2003 will build a new encryption key with the new password hash to
re-encrypt the private keys. Alternatively, you can open the user's
encrypted files using the credentials of the
Data Recovery Agent (DRA).
The default DRA is the domain
Administrator
account.
Volume Mounting
The Disk Management snap-in can be used to mount local drives to an
empty folder
on an NTFS volume.
This empty folder becomes the mount point. When a physical disk is
mounted to a folder, it is assigned a
drive path
rather than a drive letter. The Administrator can identify and manage
volume mount points by using the
mountvol.exe
command-line tool.
File Compression
Windows Server
2003 supports file and folder level compression. Compressed files can
be read and written to by any Windows-based or MS-DOS-based
application without first having to be uncompressed by another
program. When you access a file via a Windows-based or MS-DOS-based
application, NTFS automatically uncompresses the file. When you save
or close the file again, NTFS compresses it again.
Therefore NTFS allocates
disk space
based on the
uncompressed file size
and not on the compressed file size.
Copying and Moving Compressed Files and Folders
When copying a
file within an NTFS volume, the file inherits the compression state of
the target folder.
When moving a file or folder within an NTFS volume, the
file or folder retains its original compression state.
When copying a file or folder to another NTFS volume,
the file or folder inherits the compression state of the target
folder.
When moving a file or folder to another NTFS volume,
the file or folder inherits the compression state of the target
folder. Because Windows Server 2003 treats a move as a copy and then a
delete, the files inherit the compression state of the target folder.
When moving or copying a file or folder to a
FAT volume,
Windows Server 2003 automatically uncompresses the file or folder.
This is because Windows Server 2003 only supports file and folder
compression on NTFS volumes.
When moving or copying a compressed file or folder to a
floppy disk,
Windows Server 2003 automatically uncompresses the file or folder, as
floppy disks are formatted with the FAT file system. Floppy disks
cannot support the NTFS file system.
Defragmenting Volumes and Partitions
Defragmenting the disks on all your servers can ensure optimal
performance of your disks. Defragmentation is the process of
reorganizing your disk so that clusters that make up each file are
stored together, instead of being spread around the disk. Windows
Server 2003 provides two tools that work with both basic and dynamic
disks that are formatted with the FAT, FAT32, or NTFS file systems for
performing defragmentation. These tools are the
Disk Defragmenter
and the
defrag
command-line utility.
You can also use the
defrag
command-line utility to defragment a disk.
Backing Up and Restoring Data
Performing regular back ups of the data on hard disks prevents
data loss
due to disk drive failures, power outages, virus infections, and other
such incidents. If data loss occurs, and you have performed regular
backup jobs, you can restore the lost data.
Performing regular back ups of the data on hard disks prevents
data loss
due to disk drive failures, power outages, virus infections, and other
such incidents. If data loss occurs, and you have performed regular
backup jobs, you can restore the lost data.
Windows Server 2003 provides
Backup And Recovery Tools.
This includes the
Backup Wizard,
which you can use to easily back up and restore data.
You can use
Backup to back up data manually or you can schedule regular unattended
backup jobs. You can back up data to a file or to a tape. Files can be
stored on hard disks, removable disks, and recordable compact discs
and optical drives.
To successfully
back up and restore data on a Windows Server 2003 computer, you must
have the appropriate permissions and user rights.
All users
can back up their own files and folders, and files for which they have
the
Read,
Read & Execute,
Modify,
or
Full Control
permission.
All users
can restore files and folders for which they have the
Write,
Modify,
or
Full Control
permission.
By default, members of the
Administrators
and
Backup Operators
groups have the
Backup Files and Directories
and the
Restore Files and Directories
user rights and can therefore back up and restore all files regardless
of the assigned permissions.
Backup Types
Backup Wizard provides five types of backup that define which data is
backed up. Some backup types use backup
markers,
also known as archive bits, which mark a file as having changed. When
a file changes, an attribute is set on the file that indicates that
the file has changed since the last backup. When you back up the file,
this
clears
or resets the attribute.
Normal-
which backs up all selected files and folders and does not rely on
markers to determine which files to back up. During a normal backup,
any existing marks are cleared and each file is marked as having been
backed up. Normal backups speed up the restore process because the as
the files are the most current therefore you do not need to restore
multiple backup jobs.
Copy-
which backs up all selected files and folders without looking for or
clearing markers.
Incremental-
which only backs up selected files and folders that have a marker and
then
clears
the markers. Thus, if you did two incremental backups in a row on a
file and nothing changed in the file, the file would not be backed up
the second time.
Differential-
which only backs up selected files and folders that have a marker but
does not clear markers. Thus if you did two differential backups in a
row on a file and nothing changed in the file, the entire file would
be backed up each time.
Daily-
backs up all selected files and folders that have changed during the
day and does not look for or clear markers.
Backing Up System State Data
You can use the Backup utility to back up the system state data. The
system state data includes: the Registry; Component Services Class
Registration database; System startup files; Certificate Services
database; Active Directory directory services; and the
Sysvol
folder.
To back up the
system state data on a local computer, do one of the following:
In the
Backup
wizard, on the
What To Back Up
page, click
Only Back Up The System State Data.
In the
Backup
wizard, on the
Items To Back Up
page, expand
My Computer,
and then select the check box to the left of
System State.
In the
Backup
utility, on the
Backup
tab, expand
My Computer,
and then select the
System State
check box.
Restoring Files and Folders
You can use the
Backup utility in Windows Server 2003 to restore files and folders.
The Backup utility includes a Restore wizard that steps you through
the entire restore process. However, you can also restore files and
folders without using the wizard.
To restore
files and folders without using the wizard, open the Backup utility,
and specify the folders or files to restore; a restore location; and
the Restore options, such as whether to replace existing files with
backup files.
Note:
To avoid data loss and to preserve file and folder features, such as
encryption and permissions, you should restore data that is archived
from a Windows 2000 or Windows Server 2003 NTFS volume to another
Windows 2000 or Windows Server 2003 NTFS volume. Restoring the data to
a FAT volume or to a Windows NT 4.0 NTFS volume might result in the
loss of access permissions, Encrypting File System (EFS) settings,
disk quota information, mounted drive information, or Remote Storage
information.
Restoring Active Directory Directory Services
You can use the
Backup utility to restore Active Directory directory services during
the process of replacing a failed domain controller, to repair a
damaged Active Directory database, or to recover one or more objects
that are accidentally deleted from Active Directory directory
services.
Failed Domain Controllers
If a domain
controller fails completely, you must first restart the computer and
make sure Windows Server 2003 is running. Then you can use the Backup
utility to restore the latest version of the system state data, which
includes Active Directory directory services. After you have restored
Active Directory directory services, Windows Server 2003 automatically
performs a consistency check on the Active Directory database and
re-indexes it. Windows Server 2003 then updates Active Directory
directory services and the
File
Replication Service with data from their replication partners.
Damaged Active Directory Databases
If the operating system on a domain controller is functioning
normally, but the Active Directory database is damaged, you must
restart the computer, select the
Directory Services Restore Mode
advanced startup option, and then use the Backup utility to restore
the system state data.
After you have
restored the Active Directory database, restart the computer, and
Windows Server 2003 will automatically re-index the Active Directory
database and update Active Directory directory services and the File
Replication Service.
Authoritative Restores
If you restore the most recent copy of the Active Directory database
that contains the deleted objects, those objects will be deleted when
replication occurs because the objects are marked for deletion in the
replicas of the database. To prevent this from occurring, you can
perform an
authoritative restore.
When you restore an object authoritatively, it persists after
replication even though it is marked for deletion in the replicas of
the database.
See:
241594 - HOW TO Perform an Authoritative Restore to a Domain
Controller
2.7 Automated System Recovery
Every time you successfully edit your systems configuration, you
should back the configuration up against the time when you
unsuccessfully edit the settings. This backup disk is called the
automated system recovery (ASR)
disk. The ASR is not a boot disk but contains pointers to the system
state backup. You can create an ASR disk in Windows Backup.
To use the ASR
to repair a damaged Windows Server 2003 installation,
you will boot from the
Windows Server 2003 Installation CD, not the ASR disk.
In Windows 2003 setup, you will choose to repair an existing
installation, and provide the ASR disk which contains recovery
information. The recovery process will use the CD ROM to recover
system files.
Note:
The ASR performs a nonauthoritative restore of System State data.
Thus, if you are restoring a domain controller you must use
Ntdsutil.exe to make the restore authoritative for the domain.
|