Understanding the Red Hat Linux Installation Process
Before starting a big job, I always find it helpful to visualize the entire sequence of tasks I must perform. The process is similar to studying a map before you drive to a place you have never been. Red Hat Linux installation can be a big job, especially if you run into snags. This section shows you the road map for the installation process. After reading this section, you should be mentally prepared to install Red Hat Linux.
Here are the general steps for installing Red Hat Linux:
Gather information about your PC’s hardware before you install Red Hat Linux. The Linux operating system accesses and uses various PC peripherals through software components called drivers. You have to make sure that the version of Red Hat Linux you are about to install has the necessary drivers for your system’s hardware configuration. Conversely, if you do not have a system yet, look at the list of hardware that Linux supports, and make sure you buy a PC with components that Red Hat Linux supports.
Because most PCs come with Microsoft Windows preinstalled on the hard disk, you have to perform a step known as partitioning to allocate parts of your hard disk for Linux’s use. If you have a spare hard disk, you should keep Windows on the first hard disk and install Linux on the second hard disk. With a spare second disk, you don’t need to worry about partitioning under DOS or Windows. If you have only one hard disk, however, you have to partition that disk into several parts. Use a part for Windows, and leave the rest for Linux. For Windows 95/98/Me, you can use the nondestructive repartitioning program FIPS to repartition your hard disk without destroying the existing contents. FIPS creates a new partition by shrinking the existing DOS partition. For Windows NT/2000/XP systems with disk partitions that use the NT file system (NTFS), use a commercial hard-drive partitioning tool such as PartitionMagic from PowerQuest. You can, of course, install Linux as the sole operating system on a PC; in that case, you can ignore this step and simply let the Red Hat Linux installation program automatically create the necessary partitions.
Under DOS or Windows, create a Red Hat Linux installer boot disk. This boot disk is used to boot your PC and to start an initial version of the Linux operating system. If your PC is configured to boot from the CD-ROM drive, you can skip this step—instead, you can boot Linux directly from the CD-ROM. Even if your PC normally does not boot from the CD-ROM drive, you can usually press one or more keys as the PC boots and enter the BIOS setup screen from which you can select the CD-ROM drive as the boot device. You may want to consult your PC’s manual to see how you can boot from the CD-ROM.
Boot your PC from the first CD-ROM or with the Red Hat Linux installer boot disk. This procedure automatically loads the Linux kernel and runs anaconda, the Red Hat Linux installation program. From this point on, you respond to a number of dialog boxes as the Red Hat installation program takes you through the steps. You have the option of using a text mode or a graphical user interface (GUI). You have to use the text mode if, for some reason, the installation program fails to start the X Window System. In addition, you have to type the linux noprobe command if the Linux kernel does not detect some of the older hardware, such as the SCSI controller and network adapter installed in your system. When you type the linux noprobe command, you have to either provide a driver disk or load the drivers for your SCSI controller and network adapter by selecting them from a list of drivers.
Respond to the dialog box that asks you to choose a language to be used during installation. From subsequent dialog boxes, select the keyboard type, and the mouse type. If you have already placed the CD-ROM in the CD-ROM drive, the installation program uses the CD-ROM as the source of all Red Hat Linux files. Otherwise, the installation program starts in text mode and displays a dialog box from which you have to select the storage medium in which the Red Hat Linux files are located.
Choose whether you want to install a new system or to upgrade an existing installation. For a new installation, you also have to decide whether you want to set up a Personal Desktop, Workstation, Server, or a Custom system. Select a Custom installation for maximum flexibility.
Prepare the hard disk partitions on which you plan to install Linux. If you have created space for Linux by reducing the size of an existing Windows partition, now you have to create the partitions for Linux. Typically, you need at least two partitions: one for the Linux files and the other for use as the swap partition, which is a form of virtual memory. You can have the installation program automatically perform this disk-partitioning step for you, or you can manually create the partition using Disk Druid, a graphical partitioning utility. If you manually create the hard disk partitions, indicate which partition is the swap, and specify the partition on which you want to install Linux (this is called the root partition).
Specify options for installing a boot loader—the GNU GRand Unified Bootloader (GRUB) or the Linux Loader (LILO)—on your hard disk, so you can boot Linux when you power up your PC after shutting it down. For the GRUB boot loader, you can also enable a password that has to be typed every time the system boots. If you are squeamish about installing a boot loader, you can skip installing the boot loader. If you do so, you have to boot Linux using a boot disk that you create a few steps later during the installation process.
If the initial Linux kernel detects a network card (assuming that you have one installed on your system), the installation program lets you configure the network (the local area network, not the dial-up network). If the Linux kernel does not detect your network card, you can type the linux noprobe command and select your network card from a list. To configure the network, you can either identify a DHCP server (a DHCP server provides IP addresses to systems in a network) or specify a number of parameters, including an IP address, a host name, the IP address of name servers, and a domain name for your Linux system.
Configure the firewall security level for your system by selecting one of three predefined levels of security—high, medium, or none.
Select one or more languages to be used on your Linux system.
Specify the local time zone. For the United States, you can also enable the use of daylight saving time.
Select a root password. The root user is the super user—a user who can do anything—in Linux. You can also add one or more other users. Then, select the password-authentication method.
Select various software package groups to install. Each package represents a part of Red Hat Linux, from the base operating system to packages such as the GNOME and KDE graphical desktops, the Emacs editor, programming tools, and the X Window System (a graphical windowing system that GNOME and KDE require). Select the package groups you need, and let the Red Hat installation program do its job.
The installation program formats the disk partitions and installs the selected package groups.
Create a boot disk you can use to boot your Red Hat Linux system if the Linux kernel on the hard disk is damaged or if the boot loader—LILO or GRUB—does not work or if you have chosen not to install any boot loader.
Configure the X Window System. In response to dialog boxes that the installation program presents, provide information about your video card and monitor. In most cases, the installation program automatically detects the video card and the monitor. The installation program also lets you enable graphical login, so that when you boot your Linux system, it displays a graphical login window and (after you enter your user name and password) starts the GNOME or KDE graphical desktop after successful login.
If you find that Linux does not work properly with one or more of your system components (such as the network card or sound card), you may have to reconfigure the Linux operating system to add support for those system components.
The following sections guide you through the basic installation steps and the initial booting of Linux.
Insider Insight Your PC must have a CD-ROM drive—one that Linux supports—to install Linux from this book’s companion CD-ROM. Most new PCs have CD-ROM drives that connect to the hard disk controller (called IDE for Integrated Drive Electronics). Any IDE CD-ROM works in Red Hat Linux. If your PC does not have a CD-ROM drive but is on a network, you can use another PC’s CD-ROM drive and can install Linux over the network by using NFS or FTP. To install Linux over the network, you must use the Red Hat installer boot disk along with a driver disk—one that contains the drvnet.img file from the \images directory of the Red Hat Linux CD-ROM.
Preparing Your PC for Linux Installation
Before you install Linux, you should prepare your PC for the installation. You can be in either of two situations:
You already have a PC that runs some version of Microsoft Windows.
You are about to buy a new PC, and you plan to run Red Hat Linux on that PC at least some of the time.
If you are about to purchase a PC, you are lucky because you can get a PC configured with peripherals that Red Hat Linux supports. To pick a Red Hat Linux-compatible PC, simply consult the current list of hardware that Red Hat Linux supports, and select a PC certified by Red Hat. If you don’t see the PC model you are considering listed as Red Hat–certified, you can still make sure that key hardware components in the PC will work with Red Hat Linux. You may have to ask the PC vendor explicitly for detailed information about hardware, such as the video card, CD-ROM drive, and networking card, to ensure that you can use the peripherals under Red Hat Linux. Selecting a PC with Red Hat Linux–supported hardware minimizes the potential for problems when you install Red Hat Linux.
If you want to install Red Hat Linux on an existing PC, verify that the latest Red Hat Linux distribution supports all of the hardware on your PC. In other words, you have to take an inventory of your PC’s hardware components and determine whether Red Hat Linux currently supports all of them.
Checking the Red Hat Hardware Compatibility List
Like many other operating systems, Linux supports various types of hardware through device drivers. For each type of peripheral device, such as a networking card or a CD-ROM drive, Linux needs a driver. In fact, each kind of peripheral needs a separate driver. Because Linux is available free (or relatively inexpensively) and because many programmers scattered throughout the world cooperate to develop Linux, you cannot demand support for a specific kind of hardware. You can only hope that someone who can write a Linux driver has the same hardware you do. In all likelihood, that person will write a driver, which eventually will find its way into a version of Linux, then you can use that hardware under Linux. It may take a while for Linux to support new interfaces.
Secret
Red Hat categorizes all hardware into four categories depending on their compatibility with Red Hat Linux:
Certified hardware has been tested and certified by Red Hat. This type of hardware is fully supported. The installation program automatically detects such hardware and installs the appropriate drivers.
Compatible hardware has been reviewed by Red Hat and is supported. Such hardware is generally known to work, but Red Hat has not certified the hardware. The installation program should be able to detect and use compatible hardware.
Community Knowledge hardware is not tested and not supported by Red Hat. To support Community Knowledge hardware, you may have to perform some explicit steps, such as manually loading the driver module or editing the /etc/ modules.conf file.
Not supported hardware may be compatible with Linux but is not officially supported by Red Hat Linux.
For the most recent and detailed list of hardware Red Hat Linux supports, as well as the level on which a specific hardware belongs, check Red Hat’s Hardware Compatibility List at http://hardware.redhat.com/hcl/.
Cross Ref Appendixes B through F provide more information about whether or not Linux supports your system’s unique hardware configuration. In these Appendixes, you can also find information about how to get the most from your PC’s hardware under Linux.
Making a Hardware Checklist
After you have seen a list of various hardware peripherals Red Hat Linux supports, you should have a rough idea of whether or not you have the right PC hardware to use it. If you are buying a new PC to run Red Hat Linux, the hardware-compatibility list at http://hardware.redhat.com/hcl/ should help you decide the hardware configuration of your new PC.
To summarize, go through the following checklist to see whether or not you are ready to install Red Hat Linux:
Does your PC have an 80386 or better processor, with the ISA, EISA (Extended Industry Standard Architecture), VLB (VESA Local-Bus), MCA (Micro Channel Architecture), or PCI (Peripheral Component Interconnect) bus; at least 64MB of RAM; a high-density floppy disk drive; and a large hard drive (at least 4GB)? Remember that to comfortably run both Windows and Red Hat Linux on your PC, you need at least 10GB of disk space.
Does your PC have a CD-ROM drive that Red Hat Linux supports? (This is not a problem for new PCs with IDE CD-ROM drives. You need a CD-ROM drive to install Red Hat Linux from this book’s companion CD-ROMs.)
Can you get a second hard drive? (If so, you can install Red Hat Linux on that hard drive. Installing Red Hat Linux on a second drive prevents you from having to repartition your first hard drive, which usually has Windows loaded on it.)
If you have a SCSI (Small Computer System Interface) controller with any SCSI devices that you want to use under Red Hat Linux, does Linux support the SCSI controller?
Does XFree86 support your video card? (If not, you won’t be able to set up and run the X Window System.)
As the comments after the questions indicate, you do not necessarily have to answer yes to each question. You must answer yes to the first two items, however, because without that basic hardware configuration, Red Hat Linux cannot run on your system.
Insider Insight If you plan to install Red Hat Linux on an empty second hard disk or if you want to install Linux over an existing DOS/Windows partition, you do not have to go through the process of partitioning (dividing) your hard disk under MS-DOS. You can skip the next few sections and proceed to “Creating the Red Hat Installer Boot Disk.” Then you can boot Linux from the CD-ROM or the installer boot disk and proceed to install Red Hat Linux.
Repartitioning Your Hard Drive
If your PC has a single hard disk drive, chances are good that you have some version of Microsoft Windows installed on that drive. If your hard drive is at least 10GB, I recommend that you keep Windows installed on your system, even if you want to work mostly in Linux. After all, you may have Windows applications that you paid for and would like to use now and then. You can access the Windows files from Linux. You get the best of both worlds if you keep Windows around when you install Linux.
Secret
Typically, your PC hard disk is set up as a single large partition, designated by the drive letter C (Windows XP installations often have a hidden first partition and the second partition designated as drive C). Unless you can scrounge up a second hard disk for your PC, or you already have a second disk, your first task is to shrink the existing partition on your one and only hard disk and create free space on the disk for Red Hat Linux.
You can use one of the following tools to resize the existing partition on your hard disk:
PartitionMagic: This is a commercial product that can resize hard disk partitions and create new partitions on any version of Microsoft Windows and Linux. In particular, PartitionMagic can resize partitions that contain NTFS file systems used in Windows NT/2000/XP.
FIPS: This is a free program that comes with Red Hat Linux and runs in MS-DOS mode and can split an existing partition into two. FIPS works only with Windows 95/98/Me systems that use the FAT or FAT32 file system. FIPS does not work with the NTFS file system that’s often used in Windows NT/2000/XP systems. For those systems, your best bet is PartitionMagic.
With both of these tools, the idea is to shrink the existing Windows partition and to create unused disk space for the Linux partition. Later on, during Red Hat Linux installation, you have to create new partitions for Red Hat Linux in the unused space. You can either let the Red Hat installer automatically partition the unused disk space or create the partitions yourself by using the Disk Druid tool.
The bottom line is that once you have successfully shrunk the existing Microsoft Windows partition and created a block of unused disk space for Red Hat Linux, you are past the biggest hurdle in installing Red Hat Linux while retaining Windows intact.
Repartitioning with PartitionMagic
PartitionMagic, from PowerQuest, can resize and split disk partitions in all Microsoft operating systems from Windows 95/98/Me to Windows NT/2000/XP. It’s a commercial product, so you have to buy it to use it. At the time I’m writing this, the list price of PartitionMagic 8.0 is $69.95. You can read about it and buy it at http://www.powerquest.com/partitionmagic.
Caution Resizing the disk partition always involves the risk of losing all data on the hard disk. Therefore, before you resize hard disk partitions using a disk partitioning tool such as PartitionMagic, you should back up your hard disk. After making your backup, please make sure that you can restore files from the backup.
When you run PartitionMagic, it shows the current partitions in a window. If you are running Windows XP, you probably have two partitions—one small hidden partition that contains Windows XP installation files and a huge second NTFS partition that serves as the C drive. You have to reduce the size of the existing C drive, which creates unused space following that partition. Then, during Red Hat Linux installation, the installation program can create new Linux partitions in the unused space.
To reduce the size of the partition, follow these steps:
In the partition map in PartitionMagic’s main window, right-click the partition and select Resize/Move from the menu (see Figure 2-1). The Resize Partition dialog box appears.
Figure 2-1: Right-Clicking on the Partition to Resize in PartitionMagic 8.0.
In the Resize/Move Partition dialog box, click and drag the right edge of the partition to a smaller size. For a large hard disk (anything over 10GB), reduce the Windows partition to 5GB and leave the rest for Red Hat Linux. If possible, try to leave 5GB or more for Red Hat Linux.
Click OK and then Apply to apply the changes. After PartitionMagic has made the changes, click OK.
Reboot the PC.
You do not have to do anything with the disk space left over after shrinking the partition that used to be the C drive. Later, in the “Partitioning and Using the Hard Disk” section, you learn to use the free disk space to install Red Hat Linux.
Repartitioning with FIPS
Red Hat Linux comes with a utility program called FIPS (The First Nondestructive Interactive Partition Splitting Program), which can split an existing primary DOS partition into two partitions. FIPS cordons off the unused part of a hard disk, making a new partition out of it without destroying any existing data.
Insider Insight Before using FIPS, make sure that the hard disk has only a FAT or FAT32 partition and that there is enough free space to install Linux. You need 3GB to 4GB of free space for a useful Red Hat Linux installation.
Caution Resizing partitions with FIPS may cause loss of all data on the hard disk. Therefore, you should back up your hard disk before using FIPS.
The FIPS.EXE program and related files are located in the \DOSUTIL subdirectory of the first CD-ROM. To use FIPS, follow these steps:
For FIPS to work, all used areas of the disk must be contiguous or at least as tightly packed as possible. You can prepare the disk for FIPS by running a defragmenter. In Windows 95/98/Me, click the right mouse button on the disk symbol in the Explorer window, select Properties from the pop-up menu, click the Tools tab, and click the Defragment Now button. Another way to start the defragmenter is to select Start>Programs>Accessories>System Tools>Disk Defragmenter.
Create a bootable disk by using the command FORMAT A: /S. In Windows 95/98/Me, create a startup disk by using the Add/Remove Programs option in the Control Panel and then following the instructions on the Startup Disk tab.
Copy the following files from the CD-ROM to the formatted disk (the following example assumes that D: is the CD-ROM drive):
COPY D:\DOSUTILS\FIPS.EXE A:
COPY D:\DOSUTILS\RESTORRB.EXE A:
COPY D:\DOSUTILS\FIPSDOCS\ERRORS.TXT A:
FIPS.EXE is the program that splits partitions. ERRORS.TXT is a list of FIPS error messages. You consult this list for an explanation of any error messages displayed by FIPS. RESTORRB.EXE is a program that allows you to restore certain important parts of your hard disk from a backup of those areas FIPS has created.
Leave the bootable disk in the A drive, and restart the PC. The PC boots from the A drive.
Type FIPS. The FIPS program runs and shows you information about your hard disk. FIPS gives you an opportunity to save a backup copy of important disk areas before proceeding. After that, FIPS displays the first free cylinder on which the new partition can start (as well as the size of the partition, in megabytes).
Use the left and right arrow keys to adjust the starting cylinder of the new partition (the one that results from splitting the existing partition) to change the partition size. Press the right arrow to increase the starting cylinder number (this leaves more room in the existing partition and reduces the size of the new partition you are creating).
When you are satisfied with the size of the new partition, press Enter. FIPS displays the modified partition table and prompts you to enter C to continue or R to re-edit the partition table.
Press C to continue. FIPS displays some information about the disk and asks whether you want to write the new partition information to the disk.
Press Y. FIPS writes the new partition table to the hard disk and exits.
Remove the disk from the A drive, and reboot the PC. When the system comes up, everything in your hard disk should be intact, but the C drive will be smaller. You have created a new partition from the unused parts of the old C drive.
You needn’t do anything with the newly created partition under DOS. Later, in the “Partitioning and Using the Hard Disk” section, you will learn to use the new partition under Linux.
Creating the Red Hat Installer Boot Disk
After you repartition the hard disk and make room for Linux, you can begin the next step of installing Red Hat Linux from this book’s CD-ROM: creating the Red Hat installer boot disk. (For this step, you should turn on your PC without any disk in the A drive and run Windows as usual.)
Note This book comes with the Publisher’s Edition version of Red Hat Linux 9 that was provided by Red Hat. Unfortunately, these CDs do not include the DOSUTILS directory. If you have full Red Hat Linux distribution, you can use the following directions to create the boot disk. Otherwise, please download rawrite.exe from one of the FTP sites listed at
http://www.redhat.com/download/mirror.html
Click on one of the Distribution links and then look in the directory corresponding to the Red Hat Linux version number. For example, rawrite.exe for Red Hat Linux 9 should be in the 9/en/os/i386/dosutils directory. Use a Web browser to download the rawrite.exe file. Then, follow the steps outlined next, but run the version of rawrite.exe that you have downloaded from the FTP site.
Insider Insight You do not need the Red Hat installer boot disk if you can boot your PC from the CD-ROM. You may have press a key (such as the F2 key) to go into the SETUP mode as the PC powers up and change the boot device to the CD-ROM. If you can boot directly from the first companion CD-ROM, skip this section and proceed to the section entitled “Booting the Red Hat Installer.”
Like the MS-DOS or Windows boot disk, the Red Hat installer boot disk is used to start your PC, start Linux, and run the Red Hat installation program. Once you have installed Red Hat Linux, you no longer need the Red Hat installer boot disk.
The Red Hat installer boot disk contains an initial version of Linux that you use to start Linux, run the installer, prepare the hard disk, and load the rest of the files from the CD-ROMs to the hard disk. Creating the Red Hat installer boot disk involves using a utility program called RAWRITE.EXE to copy a special file called the Red Hat Linux boot image to a disk.
To create the Red Hat boot disk under Windows, follow these steps:
In Windows 95/98/Me, open an MS-DOS Prompt window by selecting Start>Run and then typing the word Command in the text field.
In Windows NT/2000/XP, select Start>Run. Click the Browse button and go to the WINNT folder; then open the SYSTEM32 folder, and select the CMD.EXE file in that folder. Click OK to run that program.
Put the first companion CD-ROM in the CD-ROM drive. Then, type the following commands in the window you opened in Step 1 or 2 (my comments are in parentheses, and your input is in boldface):
d: (use the drive letter for the CD-ROM drive)
cd \dosutils
rawrite
Enter disk image source filename: \images\bootdisk.img
Enter target diskette drive: a
Please insert a formatted diskette into drive A: and press -ENTER- :
As instructed, you should put a formatted disk into your PC’s A drive and press Enter. RAWRITE copies the boot-image file to the disk.
After you see the DOS prompt again, you can take the Red Hat boot disk out of the A drive and (if you haven’t done so already) label it so you know what it is.
Booting the Red Hat Linux Installer
The Red Hat Linux installer runs under Linux; therefore, you need to boot Linux on your PC before you can go through the installation steps. This initial version of Linux can come from the installer boot floppy or the frist CD-ROM. The initial Linux operating system, in turn, runs the Red Hat installation program, which prepares the disk partitions and copies all necessary files from the CD-ROM to the disk.
You can boot your PC with an initial version of the Linux operating system in one of the following ways:
Boot your PC from the first CD-ROM (this works only if your PC is bootable from the CD-ROM; most PCs can boot from the CD-ROM drive, as I explain a little later).
Load the initial Linux kernel by executing the AUTOBOOT.BAT command file (from the DOSUTILS directory of the CD-ROM) while your PC is running MS-DOS.
Boot your PC from the Red Hat installer boot floppy you created earlier.
The following sections describe these approaches to booting Linux and initiating the Red Hat installation.
Booting from the Red Hat Linux CD-ROM
Most new PCs can boot directly from the CD-ROM. To do so, you have to go into SETUP as the PC powers up. The exact steps for entering SETUP and setting the boot device depend on the PC, but they typically involve pressing a key such as F2. As the PC powers up, a brief message should tell you what key to press to enter SETUP. Once you are in SETUP, you can designate the CD-ROM drive as the boot device.
After your PC is set up to boot directly from the CD-ROM drive, place the first Red Hat Linux CD-ROM into the CD-ROM drive, and reboot the PC. The PC should power up and start the Linux kernel from the CD-ROM. After Linux starts from the CD-ROM, the Red Hat installation program begins to run. The section “Installing from the Red Hat Linux CD-ROM” describes this process in detail.
Starting the Red Hat Linux Installer from DOS
You can start Linux directly from the CD-ROM while your PC is running MS-DOS. An MS-DOS program called LOADLIN.EXE can load a Linux kernel into memory and begin running Linux. The Linux kernel itself is in another file. You do not need to understand all of the details about how LOADLIN starts Linux. In fact, the Red Hat CD-ROM provides a DOS batch file, AUTOBOOT.BAT, in the \DOSUTILS directory that runs LOADLIN with appropriate arguments.
Secret
You can use AUTOBOOT to start Linux directly from the CD-ROM only if your PC is running MS-DOS alone (not an MS-DOS window under Windows 95/98 or a command window under Windows NT/2000/XP). In addition, you must be able to use the CD-ROM from MS-DOS. If your PC runs Windows, select Shutdown from the Start menu, and click the button labeled Restart the Computer in MS-DOS Mode. From the DOS prompt, try to see the directory of the CD-ROM with the command DIR D: where D is the drive letter of the CD-ROM drive. If this works, you can start Linux directly from the CD-ROM. Otherwise, you have to use a boot floppy.
To start Linux, place the Red Hat CD-ROM in the CD-ROM drive, and use the following commands from the DOS prompt:
D: (use the drive letter for the CD-ROM drive)
cd \dosutils
autoboot
After Linux starts, the Red Hat installation program begins to run. The section “Installing Linux from the Red Hat CD-ROM” describes this process in detail.
The AUTOBOOT.BAT file runs the LOADLIN program with two arguments: the name of a Linux kernel file and a file that contains an initial file system (known as initial RAM disk, or initrd, because this file system is loaded into memory). If the autoboot command gives you an error message, type the following command to run loadlin and to start Linux from the CD-ROM:
loadlin autoboot\vmlinuz initrd=autoboot\initrd.img ramdisk_size= 9216 %1 %2 %3 %4 %5 %5 %7 %8 %9
Booting from the Red Hat Installer Boot Floppy
To start Linux for installation, put the Red Hat boot floppy in your PC’s A drive and restart your PC. Your PC goes through its normal startup sequence, such as checking memory and running the ROM BIOS code. Then, the PC loads Linux from the floppy and begins running the Red Hat installation program.
Watching the Boot Process during Installation
A few moments after you start the boot process, an initial screen appears—the screen displays a welcome message and ends with a boot: prompt. The welcome message tells you that help is available by pressing one of the function keys, F1 through F5.
If you want to read the help screens, press the function key corresponding to the help you want. If you don’t press any keys, after a minute the boot process proceeds with the loading of the Linux kernel into the PC’s memory. To start booting Linux immediately, press Enter. After the Linux kernel loads, it automatically starts the Red Hat Linux installation program, which, in turn, starts the X Window System and provides a GUI for the installation.
Insider Insight As the Linux kernel begins to run, various messages appear on the screen. These boot messages tell you whether or not the Linux kernel has detected your hardware. The messages typically flash by too quickly for you to follow. Afterward, the screen shows a dialog box with a welcome message and some helpful information about the installation. At this point, you can read the messages about your hardware by pressing Ctrl-Alt-F4—this switches the display to another virtual screen where all kernel messages appear in a form slightly different from what you see on the main installation screen. In particular, look for a message about the CD-ROM, because the kernel has to detect the CD-ROM to proceed with the rest of the installation. To return to the graphical installation screen, press Ctrl-Alt-F7.
Installing from the Red Hat Linux CD-ROM
After you start the initial version of Linux following the procedures described in the section “Booting the Red Hat Installer,” Linux runs the Red Hat Linux installation program—called anaconda—from the CD-ROM. The rest of the installation occurs under the control of the installation program. The default Red Hat installation program uses a GUI. You can go through the installation steps by pointing and clicking with the mouse.
Secret
anaconda is the Red Hat Linux installation program. It runs after the initial Linux kernel boots and shows the text or GUI screens through which you perform the installation. anaconda is written in the Python programming language, and the anaconda source code is included in each Red Hat Linux in the anaconda RPM.
To learn more about anaconda, visit Red Hat’s website on anaconda at http:// rhlinux.redhat.com/anaconda/.
If for some reason anaconda fails to start the X Window System–based graphical interface, press Ctrl-Alt-Del to restart the installation and type linux text at the boot prompt to activate the text-mode installer. In text mode, the Red Hat installer uses a full-screen text-based interface. Each screen typically presents a dialog box with various elements, such as lists of items from which you select one or more buttons to indicate action. Typically, the buttons are labeled OK and Cancel. The bottom of the screen displays a help message that shows you how to navigate around the text screen.
If, for some reason, the installer fails to detect your IDE CD-ROM drive and prompts you for the type of CD-ROM drive, restart the installation by pressing Ctrl-Alt-Del. When you see the boot prompt, type linux hdx=cdrom where x is a single letter—a, b, c, or d—identifying the IDE interface to which the CD-ROM drive is connected. The letter a refers to the master or primary connector of the first IDE controller, b is the slave on the first IDE controller. The letters c and d refer to the master and slave of the second IDE controller. Thus, if your CD-ROM drive is connected as the master on the second IDE controller, you would type the following at the boot prompt:
linux hdc=cdrom
Monitoring the Installation Process
As the installation progresses, you respond primarily to various dialog boxes, entering information the installation program needs. The installation program displays useful information about a number of virtual consoles—these are screens of text in memory that you can view on the physical screen by pressing the key sequences shown in Table 2-1.
Insider Insight You work mostly in the main console—virtual console 7 in GUI installation and 1 in text mode. To switch to another virtual console, press the appropriate keystroke shown in Table 2-1. For example, to view the install log on virtual console 3, press Ctrl-Alt-F3. After you are done viewing the log, press Ctrl-Alt-F7 to return to the GUI console so you can continue with the installation.
Typically, you can get by without ever having to switch to the other screens, but if something goes wrong, you can switch to the install log screen by pressing Ctrl-Alt-F3, where you can get more information about the problem.
Table 2-1: Virtual Consoles Available During Red Hat Linux Installation Virtual
Keystroke
Description Console
1
Ctrl-Alt-F1
This is the main console on which the installation program displays the text-based user interface through which you install Linux.
2
Ctrl-Alt-F2
This console displays a shell prompt from which you can use Linux commands to monitor the progress of installation. The shell prompt appears only after you insert the CD-ROM and press Enter in response to a dialog box that the installation program displays.
3
Ctrl-Alt-F3
This is the install log. Messages from the installation program appear here.
4
Ctrl-Alt-F4
The Linux kernel displays its messages on this console. After Linux initially boots, you may want to switch to this console to see the kernel messages because they include information about hardware that Linux detects in your PC.
5
Ctrl-Alt-F5
This console shows the output of any other programs run during the installation process.
7
Ctrl-Alt-F7
This is where the anaconda installer displays the X Window System-based graphical user interface.
Understanding the Red Hat Installation Phases
Installing Red Hat Linux is a fairly lengthy process that contains the following major phases:
Getting Ready to Install—Choose the language to be used during the installation process. Indicate the type of keyboard. If you haven’t already placed the first Red Hat CD-ROM into the CD-ROM drive, you also have to specify the media from which Red Hat Linux is to be loaded (for example, the CD-ROM). After the graphical installation screen appears, configure the mouse. Then specify whether you are installing or upgrading and, if you are installing, the type of installation: workstation, server, or custom. I show a custom system installation in this chapter.
Partitioning and Using the Hard Disk—This step is to prepare the hard disk space you plan to use for Red Hat Linux. Select the automatic partition option to let the installation program perform this step, or partition the disk manually using Disk Druid. At minimum, you need two partitions—one to be used as swap area and the other for the Linux root file system (represented by /). The partitions are also formatted before use. Appendix B discusses typical partitioning strategies in an operational Linux system.
Configuring Linux—In this phase, specify where the GRUB boot loader or Linux Loader (LILO) should be loaded. For GRUB, you can also specify a password that has to be entered to load Linux at system startup. You set up the TCP/IP network, the firewall, the language support, and the time zone, and you specify a password for the root—the super user. The final configuration step is to set up the authentication mechanism.
Selecting the Package Groups to Install—Select which package groups—such as X Window System, GNOME desktop, KDE Desktop, Editors, and Web server—you want to install. Up to this point, you can abort the installation without writing anything to the hard disk. After you select the package groups, the installation program prepares the hard disk and installs the selected packages on the hard disk. At the end of the installation, create a boot disk.
Completing the Installation—Configure the X Window System. This is the X configuration used when you reboot the system after the installation finishes. You have to identify your video card, monitor, select your GUI (GNOME or KDE), and indicate whether you want a graphical or text login screen.
If you have all configuration information handy (such as video card, monitor, network card details, IP addresses, and host names for the TCP/IP network configuration) and all goes well, installing Red Hat Linux from the companion CD-ROM on a fast Pentium PC should take approximately an hour (assuming that you select nearly all packages). For example, on 1 GHz Pentium PC with 128MB RAM and a 6GB-disk partition devoted to Linux, the entire installation took about an hour.
Secret
The Red Hat installer uses kudzu (a utility program) to probe—attempt to determine the presence of—specific hardware, and tailors the installation steps accordingly. For example, if the installation program detects a network card, the program automatically displays the screens on which you can configure the TCP/IP network. Therefore, you may see some variation in the sequence of steps, depending on your specific hardware configuration.
Kudzu is a configuration program as well as library of object code that can probe and determine what hardware is installed on your system. The anaconda installation program uses kudzu to detect hardware during installation.
To learn more about kudzu, visit Red Hat’s kudzu website at http://rhlinux. redhat.com/kudzu/.
The following sections describe each of the Red Hat Linux installation phases in detail.
Getting Ready to Install
In this phase, you perform the following steps before moving on to disk setup and the actual installation of Linux:
A text screen appears with a welcome message and a prompt. Press Enter to begin installing in GUI mode or type linux text to install in text mode. If you have a SCSI controller or network card that the Linux kernel may not automatically detect, you should type linux noprobe to manually install the hardware (in this case, you have to select the hardware from a list or provide a driver disk; see Appendix B for more information on setting up SCSI controllers). Typically, you press Enter to continue the installation in GUI mode.
The installation program starts the X Window System and displays a list of languages to be used during the installation. The list includes languages such as English, French, German, Icelandic, Italian, Norwegian, Romanian, Russian, Slovak, Spanish, and Ukrainian. Use your mouse to select the language you want to use for the installation screens, then click the Next button to continue.
In the graphical mode installation, each screen has online help available on the left side of the screen. You can read the help message to learn more about what you are supposed to select in a specific screen.
The installation program displays a list of keyboard layouts. Select a keyboard layout suitable for your language’s character set (for example, U.S. English in the United States).
Note that you can always reconfigure your keyboard after finishing the installation. Simply log in as root at a text console (you can get one by pressing, for example, Ctrl-Alt-F2) and type setup. This runs a text-mode setup utility that enables you to configure, among other things, the keyboard, the mouse, and the printer.
The installation program displays a screen from which you configure the mouse in your system. The various mouse types are listed in a tree structure organized alphabetically by manufacturer. You need to know your mouse type and whether or not it is connected to the PC’s serial port or the PS/2 port (this uses a small round connector). If your mouse type appears in the list, select it. Otherwise, select a generic mouse type. Most new PCs have a PS/2 mouse. Finally, for a two-button mouse, you should select the Emulate 3 Buttons option. Because many X applications assume that you are using a three-button mouse, go ahead and select this option. On a typical two-button mouse, you can simulate a middle-button click by pressing both buttons simultaneously. On a Microsoft Intellimouse, the wheel acts as the middle button.
If you select a mouse with a serial interface, you have to specify the serial port where the mouse is connected. For COM1, specify /dev/ttyS0 as the device; for COM2, the device name is /dev/ttyS1.
The installation program searches for any existing Red Hat Linux installation on your PC and then prompts you for the installation type—whether you want to install a new system or upgrade any older Red Hat installation. If you are performing a fresh installation, you have to select the installation type: Personal Desktop, Workstation, Server, or Custom. The Personal Desktop, Workstation, and Server installations simplify the installation process by partitioning the disk in a predefined manner. The Personal Desktop installation creates a Red Hat Linux system for home, laptop, or desktop use. A graphical environment is installed along with productivity applications.
A Workstation installation installs graphical environment as well as software development tools. This type of installation also deletes all currently existing Linux-related partitions and creates a set of new partitions for Linux. A Server installation deletes all existing disk partitions, including any existing Windows partitions, and creates a whole slew of Linux partitions. By default, the Server installation does not install the graphical environment, but you can always add the packages for the GUI desktops. For maximum flexibility, select the Custom installation, but note that the other installation types can be easier because many choices are already made for you.
Caution Do not select the Server installation if you want to retain any existing Windows partitions.
The next major phase of installation involves partitioning the hard disk for use in Linux.
Partitioning and Using the Hard Disk
Like Windows, Linux requires you to partition and prepare a hard disk before you can install Red Hat Linux. You usually do not perform this step; when you buy your PC from a vendor, the vendor takes care of preparing the hard disk and installing Windows and all other applications on the hard disk. Because you are installing Red Hat Linux from scratch, however, you have to perform this crucial step yourself. As you see in the following sections, this task is just a matter of following instructions.
When the Red Hat Linux installation program reaches the disk-partitioning phase, it displays a screen from which you can select a partitioning strategy. The screen gives you three options for partitioning and using the hard disk:
Automatically partition: This option causes the Red Hat installation program to create new partitions for installing Linux, based on your installation type, such as workstation or server. After the automatic partitioning, you get a chance to customize the partitions. This is the option most users choose.
Manually partition with Disk Druid: With this option, you can use the Disk Druid program that lets you partition the disk any way you want and, at the same time, specify which parts of the Linux file system are to be loaded on which partition.
From the disk-partitioning strategy screen, select the Automatically partition option to have the installer automatically partition the disk for you. The Red Hat Linux installation program, then displays another screen (see Figure 2-2) that asks you how you want the automatic partitioning to be done.
Figure 2-2: Selecting Automatic Partitioning Options during Red Hat Linux Installation.
As Figure 2-2 shows, you can select from three automatic partitioning options:
Remove all Linux partitions on this system—This option causes the Red Hat installation program to remove all existing Linux partitions and to create new partitions for installing Red Hat Linux. You can use this option if you already have Linux installed on your PC and want to wipe it out and install the latest version of Red Hat Linux.
Remove all partitions on this system—This option is similar to the first option, except that the installation program removes all partitions, including those used by other operating systems such as Microsoft Windows.
Keep all partitions and use existing free space—If you have created space for Linux by using PartitionMagic or the FIPS utility, select this option to create the Linux partitions using the free space on the hard disk. If you are installing Red Hat Linux on a new PC after resizing the partition, this is the option to choose.
Select the appropriate option, and click Next. For example, if you select the first option, the Red Hat Linux installation program displays a dialog to confirm your choice and to point out that all data in the existing Linux partitions will be lost. Click Yes to continue. In the next screen (see Figure 2-3) the installer shows the partitions it has prepared. The exact appearance of this screen depends on your hard disk’s capacity and its current partitions.
Figure 2-3: Partitions Automatically Created by the Red Hat Linux Installer.
The screen displays the disk drives and the current partition information for the drives. If you want to accept the partitions as is, click Next to proceed.
Caution If your PC does not have enough memory (typically less than 128MB), the installer asks if it can write the partition table and activate the swap partition. Once you do this, your hard disk partitions will be changed. Click Yes only if you are committed to the new partitions and definitely want to install Red Hat Linux.
You can also perform specific disk-setup tasks through the six buttons that run across the middle of the screen. Specifically, the buttons perform the following actions:
New lets you create a new partition, assuming there is enough free disk space available to create a partition. When you click this button, another dialog box appears, in which you can fill in information necessary to create a partition.
Edit lets you alter the attributes of the partition currently highlighted in the partitions list. You make changes to the current attribute in another dialog box that appears when you click the Edit button.
Delete is used to delete the partition currently highlighted in the partitions list.
Reset causes the installation program to ignore any changes you may have made.
RAID is used to set up a RAID (Redundant Array of Independent Disks) device—a technique that combines multiple disks to improve reliability and data transfer rates. There are several types of RAID configurations. You have to select at least two partitions to make a RAID device.
LVM creates a logical volume for Logical Volume Management (LVM).
For this discussion, I assume that the installer-created partitions are acceptable and that you click Next to continue with the installation.
Configuring Red Hat Linux
When the installation program finishes loading all of the selected packages, it moves on to some configuration steps. The typical configuration steps are as follows:
Install boot loader
Configure the network
Configure the firewall
Select languages to support
Set the time zone
Set the root password
Configure password authentication
The following sections describe each of these configuration steps.
Installing a Boot Loader
You can install one of two boot loaders—GRUB or LILO. GRUB stands for Grand Unified Bootloader, and LILO stands for Linux Loader—each of these boot loader programs can reside on your hard disk, and each can start Linux when you power up your PC. If you have Microsoft Windows on your hard disk, you can configure the boot loader to load Windows as well. You perform this step from the boot loader installation screen (see Figure 2-4).
Figure 2-4: Selecting and Configuring a Boot Loader.
The boot loader installation screen (see Figure 2-4) shows you information about the boot loader to be installed and the partitions from which that the boot loader boots the PC. You also get a chance to change the boot loader or add other partitions that contain non-Linux operating systems.
GRUB is the default boot loader, and it’s installed by default in the master boot record (MBR) of the hard disk. If you want to change the boot loader from GRUB to LILO or not install it at all, click the “Change boot loader” button near the top of the window. From the dialog box that appears, either select the older LILO boot loader or skip the boot loader installation entirely. If you choose not to install any boot loader, you should definitely create a boot disk later on. Otherwise, you won’t be able to start Red Hat Linux when you reboot the PC. You get a chance to create the boot disk at the very end of the installation.
In the middle of the boot loader installation screen a table lists the disk partitions from which the boot loader can boot the PC. That table lists the Linux partition and any other partitions that may contain another operating system (such as Windows XP or 2000). Each entry in that table is an operating system that the boot loader can load and start. The default operating system is the one with a check mark in the Default column.
Secret
You must assign a label to the Windows XP partition that appears in the list of partitions—click that line and enter the label. If there is no label, GRUB will not display that partition as a choice in its boot screen, and you won’t be able to boot Windows XP after you finish installing Red Hat Linux.
If GRUB fails to detect and list an existing Windows XP partition in the list, you should click the Add button and fill in the information about the Windows XP partition, as shown in Figure 2-5. Typically, the Windows XP partition should be the /dev/hda2 device (the first partition—/dev/hda1—is usually a hidden partition that holds files used to install Windows XP).
Figure 2-5: Adding a Windows XP Partition to the List of Partitions to Boot.
If you forget to enter a label for the Windows XP partition or add the XP partition to the list, that does not mean that all is lost. The XP partition is still there on the hard disk. It’s just that GRUB does not show it as an option to boot from. To fix the problem, you have to manually edit the GRUB configuration file, /etc/grub.conf. See Appendix B for information on configuring bootloaders such as GRUB and LILO. After you fix the /etc/grub.conf file, you should get the option to boot XP from the initial GRUB screen the next time you reboot the PC.
If you decide to install the GRUB bootloader, consider adding a password for GRUB. Click the “Use a boot loader password” check box and make sure that a check mark appears there. The installer displays a dialog box where you can specify a password for GRUB. With a GRUB password, your system is more secure because no one can boot your system without a password given that GRUB will not load the operating system without the password.
If you select the “Configure advanced boot loader options” check box and click Next, the installer displays the Advanced Boot Loader Configuration screen, as shown in Figure 2-6.
Figure 2-6: Selecting Where to Install Boot Loader and Providing Kernel Options.
The screen of Figure 2-6 gives you the option to install the boot loader in one of two locations:
Master Boot Record (MBR), which is located in the first sector of your PC’s hard disk (the C drive)
First sector of the Linux boot partition
You can also add any necessary boot options in the text field labeled “General kernel parameters.” These are options passed to the kernel as the system boots.
You should install the boot loader in the Master Boot Record unless you are using another operating system loader, such as BootMagic or Windows NT/2000/XP Boot Manager. After making your selections for the boot loader, click Next to continue.
Configuring the Network
Assuming that the Linux kernel has detected a network card, the Red Hat installation program displays the Network Configuration screen, which enables you to configure the local area network (LAN) parameters for your Linux system.
This step is not for configuring the dial-up networking. You need to perform this step if your Linux system is connected to a TCP/IP LAN through an Ethernet card.
Cross Ref If the Red Hat installation program does not detect your network card, you should restart the installation and type linux noprobe at the boot prompt. Then, you will be to select your network card manually. See the “Troubleshooting the Installation” section for more information.
The Network Configuration screen displays a list of the network devices (for example, Ethernet cards) installed in your PC. For each network device, you can indicate how the IP (Internet Protocol) address is set. Click the Edit button next to the list, and a dialog box appears from which you can specify the options.
You have two choices for specifying the IP address for the network card:
Configure using DHCP: Enable this option if your PC gets its IP address and other network information from a Dynamic Host Configuration Protocol (DHCP) server. This is often the case if your PC is connected to a DSL or cable modem router.
Activate on boot: Enable this option to turn on the network when your system boots.
You should select DHCP only if a DHCP server is running on your local area network. If you choose DHCP, your network configuration is set automatically, and you can skip the rest of this section.
If you do not select the “Configure using DHCP” option, you have to provide an IP address and a network mask that indicates which part of the IP address represents the network and which part the host address. After entering the requested parameters in the text input fields, click OK to close the dialog box.
Cross Ref If you have a private LAN (one that is not directly connected to the Internet), you may use an IP address from a range that has been designated for private use. Common IP addresses for private LANs are in the range 192.168.0.1 through 192.168.0.254. You will learn more about TCP/IP networking and IP addresses in Chapter 6.
In the rest of the Network Configuration screen, you have to specify how to set the host name. You have two options to set the host name:
Automatically via DHCP: Enable this option to assign a host name automatically using DHCP (the name is of the form dhcppc1, dhcppc2, and so on).
Manually: Use this option to manually specify a host name—type the name in the text field next to the radio button.
After you enter the requested parameters, click the Next button to proceed to the firewall configuration step.
Configuring the Firewall
In this step, you configure the built-in packet filter that comes with Red Hat Linux and that enables you to control how various types of IP packets are handled. You can select a predefined level of security from the Firewall Configuration screen (see Figure 2-7), and customize these security levels to suit your needs.
Figure 2-7: Setting up the Packet Filtering Firewall Built into Your System.
The installer offers the following predefined security levels:
High security means that your system does not accept any connections other than those that you explicitly enable. By default, a system with high security accepts only Domain Name Service (DNS) requests and DHCP requests.
Medium security means that the system does not accept any connection requests for ports numbered 1023 and lower, or connections to the NFS server at port 2049, remote connections to the X Window System, and the X Font Server port.
No firewall security means your system accepts connections at all ports and does not perform any security checking. Use this option only if your system runs in a trusted network, if it is already inside a firewall, or if you plan to set up a more elaborate firewall configuration later on.
Insider Insight Think about how you plan to use the Red Hat Linux system before selecting a predefined security level. For example, if you plan to use the system as a boot server on an internal network, you have to set the security level to No firewall security. Otherwise, the other PCs on the network cannot download the kernel using TFTP. If one of your system’s network interface such as eth0 is only connected to your trusted internal network, you can indicate eth0 as a trusted device. If you do not configure the firewall correctly here, you can do so late ron by following instructions outlined in Chapter 22.
After you select a predefined security level, you can click the Customize button and select other services that you might want to enable. When you’re done configuring the firewall, click the Next button to proceed to the next step.
Selecting Languages to Support
In this step, you select one or more languages that your Red Hat Linux system will support once the installation is complete. This step is similar to the language selection step you perform at the beginning of the installation when you select the language to be used during the installation. Now, you have to select the languages the system will support when you reboot the PC after completing the Red Hat Linux installation. From the Language Support Selection screen, select one or more languages to support. You must also select a default language. Then click the Next button to continue.
Setting the Time Zone
After completing the network configuration, you have to select the time zone—the difference between the local time and the current time in Greenwich, England, which is the standard reference time (also known as Greenwich Mean Time, or GMT, and UTC, or Universal Coordinated Time). The installation program shows you a screen with two tabs—Location and UTC Offset—from which you can select the time zone, either in terms of a geographic location or as an offset from the UTC. Initially, the screen shows the Location tab. This tab lets you pick a time zone by simply clicking your geographic location. As you move the mouse over the map, the currently selected location’s name appears in a text field. If you want, you can also select your location from a long list of countries and regions. If you live on the East Coast of the United States, for example, select USA/Eastern. Of course, the easiest way is to simply click the eastern United States on the map.
If the World view of the map is too large for you to select your location, click the View button on top of the map. A drop-down list of views appears with the following options: World, North America, South America, Pacific Rim, Europe, Africa, and Asia. You can then click the view appropriate for your location.
The other way to set a time zone is to specify the time difference between your local time and UTC. Click the UTC Offset tab to select the time zone this way. For example, if you are in the eastern part of the United States, select UTC-05:00 as the time zone. This tab also allows you to enable Daylight Saving Time, which applies to the United States only.
After you select your time zone, click the Next button to proceed to the next configuration step.
Setting the Root Password
After completing the time zone selection, the installation program displays a screen from which you can set the root password. Earlier versions of the Red Hat installer enabled you to add one or more user accounts at the step, but now you get a chance to add user accounts when the system boots for the first time. You can also use the useradd command or the User Manager tool to add more user accounts later on.
The root user is the super user in Linux. Because the super user can do anything in the system, you should assign a password that you can remember but that others cannot guess easily. Make the password at least eight characters long, include a mix of letters and numbers, and, for good measure, throw in some special characters, such as + or *.
Type the password on the first line, and reenter the password on the next line. Each character in the password appears as an asterisk (*) on the screen. You have to type the password twice, and both entries must match before the installation program accepts it. This ensures that you do not make any typing mistakes.
After you type the root password twice, click the Next button to continue with the installation.
Configuring Password Authentication
The installation program displays a screen from which you can enable or disable several password-authentication options.
The first two check boxes that are checked by default have the following purposes:
Enable MD5 passwords: Select this option to enable users to use long passwords of up to 256 characters instead of the standard password that can be, at most, eight characters. Note that MD5 refers to Message Digest 5, an algorithm developed by RSA, Inc., to compute the digest of the entire data of a message. Essentially, MD5 reduces a message to a digest consisting of four 32-bit numbers.
Enable shadow passwords: This option causes the password to be stored in the /etc/shadow file instead of the older /etc/passwd file (the /etc/passwd file is still used to store account information). The /etc/shadow file can only be read by the super user (root). This provides an added level of security against someone getting access to the encrypted passwords (crackers may use tools to break these password, so it’s critical to protect the password even though they are already encrypted).
The remaining authentication options on this screen are for different types of network-wide authentication protocols. The options appear in four tabs:
NIS: Click this tab and select the Enable NIS check box to authenticate users on multiple systems in the same Network Information Service (NIS) domain with a common password and group file. Select the NIS domain option to specify the domain to which your system belongs. Otherwise, select the NIS server option to use a specific NIS server for authentication. To enable NIS, your Linux system must be connected to a network with an NIS server. Chapter 18 has more information on NIS.
LDAP: Click this tab and select the Enable LDAP check box to use the Lightweight Directory Access Protocol (LDAP) as a unified authentication mechanism for logging into your Linux system, as well into other systems on the network. If you enable LDAP, you have to specify the LDAP Server and LDAP Base DN. For the LDAP Server, specify the full domain name of a server running LDAP. In the LDAP Base DN text field, enter the distinguished name (DN) to look up the authentication information. To learn more about using LDAP in Linux, consult the LDAP Linux HOWTO on the Web (http://www.ibiblio.org/pub/Linux/docs/HOWTO/LDAP-HOWTO).
Kerberos 5: Click this tab and select the Enable Kerberos check box to use Kerberos for secure, network-based authentication. If you enable Kerberos, you have to specify the Realm, KDC, and Admin Server. The Realm is the network that uses Kerberos. KDC is the Key Distribution Center, also known as the Ticket Granting Server, or TGS. Admin Server runs kadmind, through which the Kerberos realm is managed. To learn more about Kerberos, visit the Kerberos home page at the Massachusetts Institute of Technology’s (MIT) Website (http://web.mit.edu/kerberos/www/).
SMB: Click this tab and select the Enable SMB authentication to use an SMB server running on a Windows system to authenticate users. If you enable SMB authentication, you have to specify the name of an SMB server and the workgroup to which the SMB server belongs. Chapter 19 shows you how to set up a Samba server on your Red Hat Linux system.
By default, only the Enable Shadow Passwords and Enable MD5 Passwords options are selected. Unless your network is set up to use NIS, LDAP, Kerberos, or SMB for authentication, you should use these default settings for authentication; click the Next button to proceed to the next configuration step.
Selecting the Package Groups to Install
After you complete the key configuration steps, the installation program displays a message about reading package information and, after a short while, shows a screen (see Figure 2-8) from which you can select the Red Hat Linux package groups you want to install.
Figure 2-8: Selecting Package Groups to Install.
After you select the package groups, you can take a coffee break, and the Red Hat installation program can format the disk partitions and copy all selected files to those partitions.
Cross Ref Red Hat uses special files called packages to bundle a number of files that make up specific software. For example, all configuration files, documentation, and binary files for the Perl programming language come in a Red Hat package. You use a special program called Red Hat Package Manager (RPM) to install, uninstall, and get information about packages. Chapter 21 shows you how to use RPM. For now, just remember that a package group is made up of several Red Hat packages.
As Figure 2-8 shows, the package groups are organized into broad categories such as Desktops, Applications, Servers, and so on. Within each category, each package group is shown with a label, an icon, a brief description, and a check box. To select a package group, you should click on the check box so that a check mark appears in the box. As you select package groups, the installer shows the total disk space needed for the installation.
Some of the package groups are already selected, as indicated by the check marks in the check boxes. You can think of the selected package groups as the minimal set of packages recommended by Red Hat for installation for the type of installation (personal desktop, workstation, server, or custom) you have chosen. You can, however, choose to install any or all of the components. Use the mouse to move up and down in the scrolling list and click the mouse on a check box to select or deselect that package group.
Secret
In an actual production installation of Red Hat Linux, you should install exactly those package groups that you need. However, when you are trying to learn everything about Red Hat Linux, you need many different packages. If you have enough disk space (at least 6GB) for the Linux partition, go ahead and select the package group labeled Everything—this installs all the package groups. Being able to install everything makes it easy for you to try out everything. Note that Everything is much more than the package groups that you see listed during installation—everything includes nearly all the packages on the companion CD-ROMs.
I must repeat that you should install Everything only when you are trying to play with all the packages on a system that is specifically meant for experimenting with Red Hat Linux. Do not install Everything on a system that will be used for anything critical. Installing everything is likely to open up too many security holes in your system.
Table 2-2 shows the package groups organized by the categories as they appear on the Package Group Selection screen. In addition to these user-selectable package groups, the Red Hat installation program automatically installs a number of packages needed to run Linux and the applications you select. In other words, even if you do not select any of the package groups, the installation program installs a number of packages needed simply to run the core Linux operating system and a minimal set of utilities.
Insider Insight Because each package group is a collection of many different Red Hat packages, the installation program also gives you the option to select individual packages. If you select the button labeled “Select Individual Packages” that appears below the list and click the Next button, the installation program takes you to another screen from which you can select individual packages.
If you are installing Linux for the first time, you do not need to go down to this level of detail to install specific packages. Simply pick the package groups you think you need from Table 2-2. If you forget something, you can always install additional packages later with the RPM utility program or the graphical Red Hat Package Management utility, described in Chapter 21. After you have selected the package groups you want, click the Next button to continue with the rest of the installation.
Table 2-2: List of Package Groups in Red Hat Linux Package Group
Description
Desktops
X Window System
Packages that make up XFree86, the X Window System for Intel x86 systems; includes all fonts, libraries, and GUI tools for configuring X, adding users, configuring printers, and so on
GNOME Desktop Environment
Packages for the GNOME (GNU Network Object Model Environment) graphical desktop, including the Nautilus graphical shell
KDE Desktop Environment
Packages for the KDE—K Desktop Environment—another graphical desktop for Linux
Applications
Editors
Text editors such as Emacs and vi
Engineering and Scientific
Packages such as blas, octave, and gnuplot for performing math and science computations and plotting
Graphical Internet
Graphical email, Web, chat, and instant messaging clients included in packages such as evolution, mozilla, xchat, and gaim
Text-based Internet
Text-based email, Web, FTP, and news reader clients included in packages such as pine, lynx, xchat, ncftp, and slrn
Office/Productivity
Office suites, PDF viewers, and more in packages such as openoffice, abiword, xpdf, koffice, and kghostview
Sound and Video
Multimedia applications for tasks such as playing audio CDs and burning CDs in packages such as cdda2wav, cdp, cdrecord, dvdrecord, grip, xcdroast, and xmms
Authoring and Publishing
Tools to create documentation in TeX (pronounced “tech”) and DocBook document-formatting system, provided in packages such as docbook-utils, tetex-latex, and xmlto
Graphics
Packages such as xsane, gimp, ImageMagick, and gtkam to scan and manipulate images and work with digital camera
Games and Entertainment
A number of computer games provided in packages such as Maelstrom, gnome-games, and kdegames
Servers
Server Configuration Tools
Red Hat’s custom server configuration tools in packages such as redhat-config-bind, redhat-config-httpd, redhat-config-nfs, redhat-config-network, redhat-config-printer, redhat-config-printer-gui, redhat-config-samba, redhat-config-securitylevel, and redhat-config-services
Web Server
Packages such as httpd, php, mod_perl, mod_ssl, and php-pgsql needed to run the Apache Web server
Mail Server
Packages such as sendmail, imap, and sendmail-cf used to set up a mail server
Servers
Windows File Server
Samba and other supporting packages needed to use the Linux PC as a LAN Manager server
DNS Name Server
The bind and caching-nameserver packages needed to provide Domain Name Service (DNS) on the Red Hat Linux system
FTP Server
The vsftpd and anonftp packages needed to run an FTP server
SQL Database Server
Packages such as postgresql-server and mysql-server that enable you to implement an SQL server (a database) on the Red Hat Linux system
News Server
The inn package for setting up an Internet News server
Network Servers
Packages such as cipe, dhcp, and telnet-server that enable you to run other services such as CIPE (Cryptographic IP Encapsulation), DHCP (Dynamic Host Configuration Protocol), and Telnet
Development
Development Tools
Software development tools in packages such as gcc, gcc-c++, gcc-java, gdb, perl-CPAN, python-devel, cvs, rcs, and so on
Kernel Development
The kernel-source package and other libraries needed to rebuild the kernel
X Software Development
Packages such as XFree86-devel, lesstif-devel, and openmotif-devel needed to develop GUI applications for the X Window System
GNOME Software Development
Packages such as gtk+-devel, gtk2-devel, libgnome-devel, and bonobo-devel needed to develop GTK+ and GNOME graphical applications
KDE Software Development
Packages such as PyQt-devel, kdebase-devel, kdelibs-devel, qt-designer, and qt-devel needed to develop QT and KDE graphical applications
System
Administration Tools
Graphical system administration tools in packages such as redhat-config-date, redhat-config-keyboard, redhat-config-kickstart, redhat-config-language, redhat-config-mouse, redhat-config-rootpassword, redhat-config-soundcard, redhat-config-users, redhat-config-packages, redhat-config-proc, and redhat-logviewer
System Tools
System administration and networking tools in packages such as ethereal, gnome-lokkit, nmap, samba-client, and vnc
Printing Support
Packages such as cups, cups-drivers, LPRng, and ghostscript needed to print from the system
Miscellaneous
Minimal
Select this to install a minimal system.
Everything
Select this item to install all of the package groups plus many more additional packages.
Completing the Installation
After you complete the key configuration steps and select the package groups to install, the installer displays a screen informing you that installation is about to begin and that a log of the installation will be in the /root/install.log file. That file essentially lists all the Red Hat packages installed in your system. You can review the install log later and move the file to another directory for future reference. The content of the install log depends on the exact packages you choose to install.
The message also tells you that a kickstart file containing the installation options used in this installation is stored in the /root/anaconda-ks.cfg file. If you are going to repeat the same installation configuration on many more PCs, you can use this file along with a kickstart installation. See the “Using Kickstart Installation” section for a discussion of kickstart installation.
Caution This is your last chance to abort the installation without doing anything to the hard disk. To abort, simply press Ctrl-Alt-Delete to restart the PC. To proceed with installation click Next.
After you click Next at this step, the Red Hat installer formats the disk partitions and installs the packages one by one. As it installs packages, the installation program displays a status screen showing the progress of the installation, including information such as total number of packages to install, number installed so far, estimated amount of disk space needed, and estimated time remaining to install.
Secret
After the Red Hat Linux installation begins, the hard disk formatting and package installation can take quite a bit of time—so you can take a break and check back in 20 minutes or so (don’t take too long a break though or you may come back to find the installer waiting for you to insert CD #2). When you come back, you should be able to get a sense of the time remaining from the status screen, which is updated continually.
After all the packages are installed, the Red Hat installer displays a screen that asks you to insert a blank floppy into your PC’s A drive. This floppy is the boot disk that you can use to start Red Hat Linux if something happens to the hard disk or if you have not installed the boot loader. You should definitely create this boot disk if you did not install the boot loader. Otherwise, you will have no way to boot Red Hat Linux.
Insert a blank floppy into your PC’s A drive, and click the Next button (note that all data on the floppy is destroyed). The installer copies the Linux kernel and some other files to the floppy so that you can boot Linux from that floppy.
After preparing the boot disk, the installation program configures the X Window System. In this step, the installation program prepares the configuration file to be used by the X server when your system reboots.
The installation program tries to detect the video card and displays the result in a screen, as shown in Figure 2-9.
Figure 2-9: Selecting the Video Card for X Configuration.
The detected card appears as the selected item in a long list of video cards. If you know the exact name of the video card, or the name of the video chipset used in the video card, select that item from the list. Also select the amount video memory from the drop-down list at the bottom of the screen. Click Next to continue.
Next, the installer tries to detect the monitor, displaying a screen (see Figure 2-10) with the results, whether successful or not.
Figure 2-10: Configuring the Monitor.
If the installation program displays a wrong monitor or a generic one as the choice, you should enter a range of values for the two parameters that appear along the bottom of the screen:
Horizontal Sync: This is the number of times per second that the monitor can display a horizontal raster line in kilohertz (kHz). A typical range might be 30–64 kHz.
Vertical Sync: This is how many times a second the monitor can display the entire screen in Hertz (Hz). Also known as vertical refresh rate, the typical range is 50–90 Hz.
Typically, the monitor’s documentation includes all this information. If you have bought your PC recently, you may still have the documentation. If you have lost your monitor’s documentation, one way to find the information might be from your Microsoft Windows setup. If your system comes with a Windows driver for the display, that driver may display information about the monitor. Another possibility is to visit your computer vendor’s website and look for the technical specification of the monitor. I was able to locate useful information about my system’s monitor from the vendor’s website.
Caution Do not specify a horizontal synchronization range beyond the capabilities of your monitor. A wrong value can damage the monitor.
Next, the installation program displays a screen from which you can customize the system’s GUI, as shown in Figure 2-11.
Figure 2-11: Customizing X and the Login Type.
Select the color depth and screen resolution from drop-down lists. Then, click one of two check boxes—Text or Graphical—for a text or graphical login screen when the system restarts.
After you finish selecting X configuration options, click the Next button. At this point, you are basically done. Remove the floppy from the A drive (if you used a boot floppy), and click Exit to reboot your PC.
Insider Insight When Red Hat Linux boots for the first time, a server named firstboot runs and gives you an opportunity to perform some initial setup tasks such as setting the date and time and installing any other CDs that you might have.
Subscribe to:
Post Comments (Atom)
No comments:
Post a Comment