Insert the Linux CD-ROM in the CD drive, and turn on the computer. On most computers, that's all it takes to boot the Red Hat installation program.
When you boot the installation program, a screen full of information is displayed, and you are given a boot prompt. The information you receive varies from distribution to distribution. In our example, Red Hat uses it to tell you about the installation program and to provide a help feature.
But the important thing on this screen is not the information; it's the boot prompt. Most distributions have it, and it serves the same purpose for them all: It gives the system administrator an opportunity to provide input to the boot process. For example, to run a expert-style text mode installation on a Red Hat system, enter boot: text expert
This bypasses the default graphical installation program, and starts the text-based installer. The expert option causes the installation program to present more configuration options. Most installations do not require either text mode or expert mode. This is just an example of how the boot prompt is used, and the boot prompt can do much more than control which installation program is used. For example, it can be used to input the correct geometry for the hard disk.
Most systems don't need input at the boot prompt, so the first time you run the installation program on any system, just press the Enter key. Only if the installation fails should you try it again with input at the boot prompt. An example of this is if the installation program was not able to automatically detect all of the storage on the hard drive through normal probing. This can happen with very large disk drives on systems in which the BIOS reports the wrong disk geometry to the operating system. See Chapter 1, "The Boot Process," for more information about the commands that can be entered at the boot prompt.
Responding to the boot prompt boots the Linux kernel, which in turn starts the Linux installer. Linux will detect some essential hardware during this phase. As it does, it displays information about that hardware on the screen. At this time, Linux also constructs a runtime environment for the installation program. If you used a boot floppy, the installation may prompt for a second disk to create the runtime environment or to install optional device drivers.
When the system completes the initialization of the runtime environment, the Red Hat installation asks you to select the language and keyboard you're using, and then displays a welcome message. If it detects a PCMCIA chipset, it asks you if you need PCMCIA support. At this point in the installation process, PCMCIA support is only required for laptops that have a CD-ROM attached through a PCMCIA interface card or for laptops that are installing over the network using a PCMCIA Ethernet card. If you answer that you do need PCMCIA support, you must provide a supplemental disk. Create it ahead of time with rawrite or dd by copying /images/ pcmcia.img from the Red Hat CD-ROM.
If the system did not boot from the CD-ROM, the Red Hat installation next asks for the installation method. (To force a system that boots from the CD-ROM to ask for an installation method, enter text expert at the boot prompt.) If you select the Local CDROM installation method, make sure the Red Hat CD-ROM is in the drive, because the installation program attempts to detect the correct drive by locating the CD-ROM. If the installation program cannot detect the CD-ROM in an IDE drive, it will ask you what type of CD-ROM drive you have. The choices are SCSI and Other. Some servers use a SCSI CD-ROM drive. If you select SCSI, you will be asked to tell the installation program which SCSI device is the CD-ROM. The Other category is reserved for CD-ROM drives connected through sound cards or other proprietary interfaces. If you select Other, you will be given a list of supported CD-ROM drives. Select the make and model of your drive from that list.
Note If you have an IDE drive that was not detected, you must reboot the system and pass the name of the CD-ROM device to the installation program at the boot prompt (for example, hdc=cdrom).
After the CD-ROM is located, the Red Hat installation asks if this is an install or an upgrade. The first time you install Red Hat on the hardware, select Install. For subsequent installs, you can use the Upgrade Existing System option to save the configuration files you have customized for your system, although some administrators prefer to always do a complete installation and to restore customized configuration files from the system backup. When an upgrade installation is used, new configuration files are given the extension .rpmnew so that they do not overwrite your configuration; or, in some cases, the old file is given the extension .rpmsave, and the new one is put into place. For example, after an upgrade the new httpd.conf file is stored in httpd.conf.rpmnew.
When you select Install, you must also select the type of installation:
• Workstation installation is designed for a desktop system.
• Server installation provides a predefined server configuration.
• Laptop installation ensures that packages needed for a laptop are installed.
• Custom installation gives you the most control over the installation and configuration process.
As the system administrator, you should know as much as possible about how your server is installed and configured, so select Custom. Choosing Server will work, but you won't have as much control over the installation.
Was this article helpful?