Managing Basic Disks and Dynamic Disks


Objective:

Manage basic disks and dynamic disks

Windows Server 2003 supports two types of physical disk configurations: basic and dynamic. A single physical disk must be one type or the other; however, you can intermingle the physical disk types in a multiple disk server.

Introduction to Basic Disks

When a new disk is installed in Windows Server 2003, it is installed as a basic disk. The basic disk type has been used in all versions of Microsoft Windows dating back to 1.0, OS/2, and in MS-DOS. This configuration allows a basic disk created in Windows Server 2003 to be recognized by these earlier operating systems.

A basic disk splits a physical disk into units called partitions. Partitions allow you to subdivide your physical disk into separate units of storage. There are two types of partitions: primary and extended.

A primary partition can be used to store a boot record so that you can boot your server from that partition. A hard disk can be configured with one to four primary partitions. You can use these partitions to boot multiple operating systems, or to just contain data. Although multiple primary partitions can contain boot records, only one primary partition can be marked active. When a primary partition is marked as active, the system BIOS looks for the boot files needed to start the system in that partition.

The other type of partition is the extended partition, which allows you to create a theoretically unlimited number of logical drives inside that partition. Using older operating systems limits the number of drive letters that are available; however, in the later versions of Windows NT/2000/2003/XP, these logical drives can be mounted without a drive letter. The downside is that logical drives created inside an extended partition are not bootable.

You can have up to four primary partitions, or three primary partitions and an extended partition, on a single physical hard disk. The basic disk was the only type supported in versions of Windows prior to Windows 2000.

Note: Watch the Terminology

Beginning with Windows XP, Microsoft has started to refer to primary partitions and the logical drives that are contained in extended partitions as basic volumes. Do not confuse this with simple volumes, which are discussed later in the chapter. In addition, remember that a basic disk is a physical entity and that a basic volume is a logical one.


Exam Alert: Know Your Partitions

Be familiar with the two types of partitions, logical drives, and which drives are bootable.


Before a physical hard disk can be used with an operating system, it must be initialized. The initialization process is used to write the master boot record (MBR) to the first sector of the hard drive. The MBR contains a small amount of startup code and a partition table that lists the configuration of the partitions. The hard drive can be initialized during a clean installation of Windows 2003 or by using the command-line utility FDISK.

When a Windows Server 2003 server is started, its basic input/output system (BIOS) reads the MBR on the physical hard drive that is used to start the server. The MBR contains a pointer to the location of the active partition on the hard drive and the code needed to begin the startup process. The active partition is also referred to as the system partition. On an Intel-based system, the system partition contains the BOOT.INI, NtdETECT.COM, and NTLDR files. These files tell the server how to start the operating system.

Note: NTBOOTDD.SYS

Another file that might be included is NTBOOTDD.SYS. This file is needed only if you are using a SCSI controller that has its BIOS disabled.


The BOOT.INI file, shown in Figure 12.1, contains the physical path to the location of the folder that contains the operating system files. The BOOT.INI file can contain paths to multiple operating systems. At boot, you are presented with a menu that allows you to choose which operating system to start. In the example shown, there is an additional option to boot to the Recovery Console.

Figure 12.1. A typical BOOT.INI file, showing the location of the system partition.


Note: More on the Recovery Console

The Recovery Console is covered in Chapter 17, "Managing and Implementing Disaster Recovery."


The partition that contains the operating system files is referred to as the boot partition. Of course, you don't have to have separate boot and system partitions. You can use a single partition to contain both the boot and the operating system files; for example, the common configuration for most servers is for both partitions to be located on the C: drive.

Exam Alert:

There might be scenarios on the exam where you must understand the logical relationship between boot partitions and system partitions.


Although the system partition must be located on a physical partition, the boot partition can be contained on a logical drive in the extended partition. It is important to remember that although physical and extended partitions are physical entities, active, system, and boot partitions are logical entities.

Using the DISKPART Utility

A new feature in Windows Server 2003 is the capability to extend partitions on basic disks. This allows you to extend either primary or logical partitions into unallocated, adjacent, contiguous space. In Windows 2000, only dynamic volumes could be extended.

Extending partitions on a basic disk is performed using the DISKPART utility. DISKPART is a command-line utility that can be used to perform a variety of disk-management tasks. Because it is a command-line utility, it can be included in scripts or run from a remote session.

Although you use the Disk Management snap-in to perform the majority of disk-related tasks, DISKPART can perform more operations than the Disk Management snap-in. For example, extending a basic volume can be performed only through the DISKPART utility.

However, whereas the Disk Management snap-in prevents you from performing operations that might result in data loss, DISKPART has no safeguards and allows you to perform just about any operation on your disk with little or no warning. Here is a partial list of the operations that can be performed using the DISKPART utility:

  • Set a partition to active.

  • Add, create, and delete disks, partitions, and volumes, both basic and fault tolerant.

  • Add or break mirrored volumes (RAID-1).

  • Repair mirrored sets (RAID-1) or fault-tolerant volumes (RAID-5).

  • Import foreign disks.

  • List the size, configuration, and status of the physical and logical disks in the server.

In Windows Server 2003, a basic volume can be extended into the next contiguous, unallocated space on that disk. The limitations are that you cannot extend the system or boot partitions you used to start the current session, and the partition must be formatted with NTFS.

Note: Required Permissions

To perform disk-management tasks on a Windows Server 2003 server, you must be a member of either the Administrators or Server Operators local group on the server. Alternatively, if the server is a member of a domain, you should be a member of the Domain Admins group.


To extend a basic volume on a basic disk using the DISKPART utility, follow Step by Step 12.1.

Step by Step

12.1 Using DISKPART to extend a basic volume

1.

From the Start menu, select Run. From the command line, enter DISKPART and press the Enter key.

2.

A command window opens with DISKPART in command mode. You need to select the disk and partition on which you want to perform operations. You can accomplish this by entering the list disk command to obtain a list of physical disks in your server, as shown in Figure 12.2. Note that an asterisk appears next to the disk, if any, that has the current focus.

Figure 12.2. Output from the DISKPART utility, showing the results of the list disk command.


3.

After you determine the number of the disk you want to perform the operation on, you can set the focus to that disk by entering select disk x, where x is the number of the disk.

4.

After you have set the focus to the desired disk, you can list the partitions or volumes on the disk by entering the list partition or the list volume command, as shown in Figure 12.3.

Figure 12.3. Output from the DISKPART utility, showing the results of the list partition command.


5.

After you determine the number of the partition or volume you want to perform the operation on, set the focus to it by entering either select partition x or select volume x, as appropriate, where x is the number of the partition or volume.

6.

After the desired disk and partition/volume are selected, use the extend command to extend the selected partition into the contiguous unallocated space. At the command prompt, enter extend size=x, where x is the amount in megabytes (MB) that you want to extend the partition or volume, as shown in Figure 12.4. If no size is specified, all the contiguous allocated space is added to the existing partition or volume.

Figure 12.4. Output from the DISKPART utility, showing the results of the extend command. The asterisk is displayed next to the partition or volume that has the focus.


7.

You can verify the results by entering the list partition or list volume command and then checking the new size.

Note: Use the Proper Identification

On basic disks, you can specify the volume or partition by using the number, drive letter, or mount point. On dynamic disks, you can use only the volume number.


Exam Alert: Know the Uses for DISKPART

Be familiar with the things that DISKPART can do that can't be accomplished via the Disk Management MMC.


Introduction to Dynamic Disks

Dynamic disks were first introduced in Windows 2000 and are the preferred disk type for Windows Server 2003. Unlike a basic disk, a dynamic disk is divided into volumes instead of partitions. Although a clean installation of Windows Server 2003 creates a basic disk by default, any additional disks can be added as basic or dynamic disks. In addition, after the initial installation, the basic disk can be converted to a dynamic disk.

Unlike basic disks, which use the original MS-DOSstyle master boot record (MBR) partition tables to store primary and logical disk-partitioning information, dynamic disks use a private RAID-5 set of the disk to maintain a Logical Disk Manager (LDM) database. The LDM contains volume types, offsets, memberships, and drive letters of each volume on that physical disk. The information in the LDM database is also replicated to all the other dynamic disks so that each dynamic disk knows the configuration of every other dynamic disk. The disks that share the same configuration information are known as a Disk Group.

The Disk Group feature makes dynamic disks more reliable and recoverable than basic disks. This extra reliability and recoverability means that a disk move or failure does not cause the disk configuration to be lost. The LDM database is stored in a 1MB reserved area at the end of every dynamic disk.

In addition, configuration changes can be made to dynamic disks without rebooting the server. Dynamic volumes can be created, deleted, and expanded, all without a reboot.

The following are the five types of volumes available using dynamic storage disks:

  • Simple volumes Simple volumes are similar to a partition on a basic disk.

  • Spanned volumes Spanned volumes can take various amounts of disk space, from 2 to 32 physical disks, and use that space to create a single large volume. Spanned volumes provide no fault tolerance. In fact, they can be more prone to failure than other types of volume because if any disk fails, the entire set is lost. The advantage of spanned volumes is that you can quickly add more storage space.

  • Striped volumes Striped volumes can be created from 2 to 32 physical disks. Striped volumes write data to these disks in 64KB sequential stripes. The first stripe is written to the first disk, the second stripe is written to the second disk, and so forth. Striped volumes, also known as RAID-0, provide no fault tolerance. The advantage provided by striped volumes lies in the overall disk I/O performance increase of the computer because the total disk I/O is split among all the disks in the volume.

  • Mirrored volumes Mirrored volumes, also known as RAID-1 volumes, provide fault-tolerant data storage using two physical disks. Data is written simultaneously to each physical disk so that they contain identical information. If one of the drives in a mirrored volume fails, the system continues to run using the other volume. The total volume capacity will be equal to that provided by one of the physical disks.

  • RAID-5 volumes RAID-5 volumes are similar to striped volumes in that they use multiple disksin this case, from 3 to 32 physical disks of the same size. The total volume capacity is equal to that provided by the number of physical disks minus one. Data is written sequentially across each physical disk and contains both data and parity information. For example, if you create a volume using four 1GB disks, your usable storage would be 3GB, because 1GB is devoted to storing the parity information. The parity information from the set is used to rebuild the set should one disk fail, thus providing fault tolerance. RAID-5 volumes in Windows Server 2003 cannot sustain the loss of more than one disk in the set while still providing fault tolerance.

After Windows Server 2003 is installed, you can convert a basic disk to a dynamic disk using the Disk Management snap-in or the DISKPART command-line utility. Using the Disk Management utility to convert disks is covered later in this chapter.

Limitations of Dynamic Disks

After reading about the wonderful features of dynamic disks, you're probably thinking that you should always use dynamic disks instead of basic disks. However, dynamic disks have some limitations, and there are situations in which they can't be used at all. Here are some of the limitations of dynamic disks:

  • You cannot use dynamic disks if you want to dual boot a computer with an older operating system such as Windows 9x or Windows NT 4.0. Unfortunately, these older operating systems cannot access a dynamic disk, either as a boot device or for file storage. Only Windows 2000, 2003, and XP machines recognize dynamic disks. However, a dynamic disk can be accessed over the network by older operating systems.

  • You can't set up multiple boot partitions on a dynamic disk.

  • If you convert a basic disk that was configured to multiboot two or more operating systems to a dynamic disk, that disk is no longer bootable from any of the operating systems on it.

  • Dynamic disks are not supported in laptops or removable disks, such as Zip disks or disks connected via a Universal Serial Bus (USB) or FireWire (IEEE 1394) interface. In addition, dynamic disks are not supported on disks used in shared Cluster Array configurations.

  • Windows Server 2003 cannot be installed on a dynamic volume that was created from unallocated space on a dynamic disk.

  • You can install Windows Server 2003 only on a basic disk, or on a dynamic volume that was converted from a basic boot volume/partition.

  • After a basic disk is converted to a dynamic disk, it can't be converted back to a basic disk. The only way to revert to a basic disk is to back up the data, reinitialize the disk, repartition it, and restore the data.

  • When installing Windows Server 2003 on a dynamic disk, you can't change the volume or partition sizes during the setup procedure.

Exam Alert: Accessing Disks Over the Network

When accessing a disk over the network, the client PC doesn't know or care whether it's accessing a basic or dynamic disk, using FAT or NTFS. This could easily be part of an exam question.


Most of the dynamic disk limitations in the Windows 2000/2003/XP family are due to the requirement of these operating systems that the boot partition have an entry in the partition table. After a basic disk is converted to a dynamic disk, you cannot change the partition table to mark a volume as an active volume.

Note: Legacy Volumes

In Windows NT 4.0, basic disks could be created in the following configurations:

  • Spanned volume sets

  • Mirrored volumes (RAID-1)

  • Striped volumes (RAID-0)

  • Striped volumes with parity (RAID-5)

Windows Server 2003 does not support the creation or use of any of these configurations on a basic disk. If any of these volumes are present in a server that is upgraded to Windows Server 2003 or are added after Windows 2003 is installed, they will no longer be accessible. These volumes must be mounted in either Windows NT 4.0 or Windows 2000 Server to recover the data. For more information, see the Microsoft Knowledge Base Article Q328520, "Disk Management Displays Previous FTDisk Volumes As 'Unknown' with No Drive Letter."





MCSA. MCSE 70-290 Exam Prep. Managing and Maintaining a MicrosoftR Windows ServerT 2003 Environment
MCSA/MCSE 70-290 Exam Prep: Managing and Maintaining a Microsoft Windows Server 2003 Environment (2nd Edition)
ISBN: 0789736489
EAN: 2147483647
Year: 2006
Pages: 219
Authors: Lee Scales

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net