One of the most important things to understand before using GRUB is how the program refers to devices, such as hard drives and partitions. This information is particularly important when configuring GRUB to boot multiple operating systems.
The first hard drive of a system is called (hd0) by GRUB. The first partition on that drive is called (hd0,0), and the fifth partition on the second hard drive is called (hd1,4). In general, the naming convention for file systems when using GRUB breaks down in this way:
(<type-of-device><bios-device-number>,<partition-number>) |
The parentheses and comma are very important to the device naming conventions. The <type-of-device> refers to whether a hard disk (hd) or floppy disk (fd) is being specified.
The <bios-device-number> is the number of the device according to the system's BIOS, starting with 0. The primary IDE hard drive is numbered 0, while the secondary IDE hard drive is numbered 1. The ordering is roughly equivalent to the way the Linux kernel arranges the devices by letters, where the a in hda relates to 0, the b in hdb relates to 1, and so on.
Note | |
---|---|
GRUB's numbering system for devices starts with 0, not 1. Failing to make this distinction is one of the most common mistakes made by new GRUB users. |
The <partition-number> relates to the number of a specific partition on a disk device. Like the <bios-device-number>, the partition numbering starts at 0. While most partitions are specified by numbers, if a system uses BSD partitions, they are signified by letters, such as a or c.
GRUB uses the following rules when naming devices and partitions:
It does not matter if system hard drives are IDE or SCSI. All hard drives start with hd. Floppy disks start with fd.
To specify an entire device without respect to its partitions, simply leave off the comma and the partition number. This is important when telling GRUB to configure the MBR for a particular disk. For example, (hd0) specifies the MBR on the first device and (hd3) specifies the MBR on the fourth device.
If a system has multiple drive devices, it is very important to know the drive boot order set in the BIOS. This is rather simple to do if a system has only IDE or SCSI drives, but if there is a mix of devices, it can become confusing.
When typing commands to GRUB involving a file, such as a menu list to use when allowing the booting of multiple operating systems, it is necessary to include the file immediately after specifying the device and partition. A sample file specification to an absolute filename is organized as follows:
(<type-of-device><bios-device-number>,<partition-number>)/path/to/file |
Most of the time, a user will specify files by the directory path on that partition, plus the filename.
It is also possible to specify files to GRUB that do not actually appear in the filesystem, such as a chain loader that appears in the first few blocks of a partition. To specify these files, you must provide a blocklist, which tells GRUB, block by block, where the file is located in the partition. Since a file can be comprised of several different sets of blocks, there is a specific way to write blocklists. Each file's section location is described by an offset number of blocks and then a number of blocks from that offset point, and the sections are put together in a comma-delimited order.
Consider the following blocklist:
0+50,100+25,200+1 |
This blocklist tells GRUB to use a file that starts at the first block on the partition and uses blocks 0 through 49, 99 through 124, and 199.
Knowing how to write blocklists is useful when using GRUB to load operating systems that use chain loading, such as Microsoft Windows. It is possible to leave off the offset number of blocks if starting at block 0. As an example, the chain loading file in the first partition of the first hard drive would have the following name:
(hd0,0)+1 |
The following shows the chainloader command with a similar blocklist designation at the GRUB command line after setting the correct device and partition as root:
chainloader +1 |
Some users are confused by the use of the term "root filesystem" with GRUB. It is important to remember that GRUB's root filesystem has nothing to do with the Linux root filesystem.
The GRUB root filesystem is the root partition for a particular device. GRUB uses this information to mount the device and load files from it.
With Red Hat Linux, once GRUB has loaded its root partition, which equates to the /boot partition and contains the Linux kernel, the kernel command can be executed with the location of the kernel file as an option. Once the Linux kernel boots, it sets its own root filesystem. The original GRUB root filesystem and its mounts are forgotten; they only existed to boot the kernel file.
Refer to the root and kernel commands in the Section called GRUB Commands for more information.