ToDo Übersetzen!
Dokumentation zur FAI Harddisk Konfiguration
Das alte skript "setup_harddisks.pl" zur Festplattenpartitionierung wurde ersetzt durch das neue fai-setup-storage ( http://fai-project.org/doc/man/setup-storage.html).
Die Konfigurations files für die Festplatten stehen unter fai/disk_config/ und werden auch wie eine Klasse definiert. Zeile die mit einem # beginnen sind Kommentare.
Die folgenden Beispiele gelten nur für das alte Tool (setup_harddisks)!!!
Das folgende Beispiel ist eine konfiguration für eine IDE Hardisk "Primary Master" von Linux "hda" genannt. Und für die zweite SCSI disk genannt "sdb":
# <Type> <mount point> <Size in mb> [mount options] [;extra options] disk_config hda primary / 200 defaults,errors=remount-ro logical /home 100-300 logical /scratch1 10- defaults,nosuid ; -i 15000 -m 0 disk_config sdb primary /tmp 300-500 rw ;ext2 primary /backup preserve2 rw logical swap 50-100 logical /scratch2 100-300 rw ;-m 30 logical - preserve7 logical /var 100 ;-j logical /var/tmp preserve9 ;format primary /tmp/mytmp -300
Every disk configuration starts with the command "disk_config" followed by the Linux device name without "/dev/". After this command for each partition a line containing the type, mount point and size is added. Mount options and additional parameters for mke2fs -- separated from the mount options by a semicolon -- can be added.
Type: There are two types of partitions: primary and logical. Primary partitions are bootable, but there is a maximum of four primary partitions on each disk. The Linux root filesystem must be of this type. (Wirklich?)
All other partitions are called logical. Because logical partitions are gathered internally in one big primary partition, only three primary partitions can be used if logical partitions are defined. Normally only one primary partition for the root file system is created and all others are logical, like hda in the example above.
Mount point: The mount point is the full path (beginning with a slash) for the filesystem. The value "swap" defines a Linux swap partition. Both types will be automatically added to /etc/fstab. A dash "-" indicates that the partition will not be mounted and can be used for other types of filesystems (VAT, UFS, MINIX, ...)
Size: This is the size of the partition in megabytes. This value is rounded up to fit to a cylinder number. There are several forms to define the size:
- "200" means about 200MB, no more no less
- "100-300" sets a 100MB minimum and a 300MB maximum
- "10-" sets a minimum of 10MB and a maximum of the disk size
- "-300" sets a minimum of 1MB and a 300MB maximum
By default, a new filesystem (currently of type ext2 or swap) will be created, and all data on the partition is lost. The meaning of "preserve<no>" will be described later.
Calculating the partition size: If an interval is defined for several partition sizes, the script maximizes the values by preserving the ratio between them.
Mount options: The mount options will be copied to /etc/fstab. An empty field sets the option to "defaults" (see mount(8)).
Extra options: The last field is a space separated extra options list. The following options are known:
-i <bytes> : Bytes per inode
(only ext2/3 filesystem)
-m <blocks> : Reserved blocks percentage for superuser
(only ext2/3 filesystem)
-j : Create the filesystem with an ext3 journal
-c : Check for bad blocks
ext2 : flag the partition as ext2 instead of auto in /etc/fstab
ext3 : flag the partition as ext3 instead of auto in /etc/fstab
format : Always format this partition even if preserve
reiser : create a reiser file system, not an ext2The order of the extra options is not relevant. For more information see mke2fs(8).
Thus, we have the following interactions between -j, ext2 and ext3 :
<no option> : an ext2 fs flagged as auto in the fstab -j : an ext3 fs flagged as auto in the fstab ext2 : an ext2 fs flagged as ext2 in the fstab -j ext2 : an ext3 fs flagged as ext2 in the fstab -j ext3 : an ext3 fs flagged as ext3 in the fstab ext3 : an ext2 fs flagged as ext3 in the fstab !!BAD!!
The use of auto in the fstab for ext3fs enable a non-ext3 enabled kernel or tool to cope with these partitions.
Preserving partitions:
It is possible to preserve the size of a partition or additionally to preserve the existing data on this partition. To preserve only the partition size, the number of the partition must be unchanged and the size must be specified as "preserve<no>". The number <no> is the device number (used by /dev/hda*, or see output of "df") of the partition. Primary partitions are numbered from one to four, the numbers for logical partitions begin at five.
Problems were reported (feb 2003) when using more than two primary partitions and trying to preserve a logical partition. It you have this problem, try to use only two primary partitions.
In this example, the partitions' numbers (= device number) are also shown for disk sdb:
primary /tmp 300-500 # 1 primary /backup preserve2 # 2 logical swap 50-100 # (3) 5 logical /scratch2 100-300 # (3) 6 logical - preserve7 # (3) 7 logical /var 100 # (3) 8 logical /var/tmp preserve9 # (3) 9 primary /tmp/mytmp -300 # 4
The first two partitions are of type primary, so they get the numbers 1 and 2. The logical partitions start at 5 and the last gets number 8. All logical partitions define the primary partition 3, but this number is not used. So, if you want to preserve /dev/hda7 you have to insert a minimum of two logical partitions before it. asc@lap:/usr/doc/fai$ zcat README.disk_config.gz |more Documentation for hard disk configuration
The script setup_harddisks.pl partitions and formats the local disks. It uses all configuration files in /fai/disk_config/ which are also defined as classes. Lines beginning with # are comments. The following example is a configuration for the first IDE disk "hda" and for the second SCSI disk "sdb":
# <Type> <mount point> <Size in mb> [mount options] [;extra options] disk_config hda primary / 200 defaults,errors=remount-ro logical /home 100-300 logical /scratch1 10- defaults,nosuid ; -i 15000 -m 0 disk_config sdb primary /tmp 300-500 rw ;ext2 primary /backup preserve2 rw logical swap 50-100 logical /scratch2 100-300 rw ;-m 30 logical - preserve7 logical /var 100 ;-j logical /var/tmp preserve9 ;format primary /tmp/mytmp -300
Every disk configuration starts with the command "disk_config" followed by the Linux device name without "/dev/". After this command for each partition a line containing the type, mount point and size is added. Mount options and additional parameters for mke2fs -- separated from the mount options by a semicolon -- can be added.
Type: There are two types of partitions: primary and logical. Primary partitions are bootable, but there is a maximum of four primary partitions on each disk. The Linux root filesystem must be of this type.
All other partitions are called logical. Because logical partitions are gathered internally in one big primary partition, only three primary partitions can be used if logical partitions are defined. Normally only one primary partition for the root file system is created and all others are logical, like hda in the example above.
Mount point: The mount point is the full path (beginning with a slash) for the filesystem. The value "swap" defines a Linux swap partition. Both types will be automatically added to /etc/fstab. A dash "-" indicates that the partition will not be mounted and can be used for other types of filesystems (VAT, UFS, MINIX, ...)
Size: This is the size of the partition in megabytes. This value is rounded up to fit to a cylinder number. There are several forms to define the size:
"200" means about 200MB, no more no less
"100-300" sets a 100MB minimum and a 300MB maximum
"10-" sets a minimum of 10MB and a maximum of the disk size
"-300" sets a minimum of 1MB and a 300MB maximumBy default, a new filesystem (currently of type ext2 or swap) will be created, and all data on the partition is lost. The meaning of "preserve<no>" will be described later.
Calculating the partition size: If an interval is defined for several partition sizes, the script maximizes the values by preserving the ratio between them.
Mount options: The mount options will be copied to /etc/fstab. An empty field sets the option to "defaults" (see mount(8)).
Extra options: The last field is a space separated extra options list. The following options are known:
-i <bytes> : Bytes per inode
(only ext2/3 filesystem)
-m <blocks> : Reserved blocks percentage for superuser
(only ext2/3 filesystem)
-j : Create the filesystem with an ext3 journal
-c : Check for bad blocks
ext2 : flag the partition as ext2 instead of auto in /etc/fstab
ext3 : flag the partition as ext3 instead of auto in /etc/fstab
format : Always format this partition even if preserve
reiser : create a reiser file system, not an ext2The order of the extra options is not relevant. For more information see mke2fs(8).
Thus, we have the following interactions between -j, ext2 and ext3 :
<no option> : an ext2 fs flagged as auto in the fstab -j : an ext3 fs flagged as auto in the fstab ext2 : an ext2 fs flagged as ext2 in the fstab -j ext2 : an ext3 fs flagged as ext2 in the fstab -j ext3 : an ext3 fs flagged as ext3 in the fstab ext3 : an ext2 fs flagged as ext3 in the fstab !!BAD!!
The use of auto in the fstab for ext3fs enable a non-ext3 enabled kernel or tool to cope with these partitions.