View source for Setting up disks manually


The menu choices that are provided by <code>setup-alpine</code> can not cover all needs, however it's possible to:

* Call [[Alpine_setup_scripts#setup-disk|setup-disk]] directly and pass partitioning parameters to it.
* Or to set up the partitions and filesystems manually (to the desired details) and [[https://wiki.alpinelinux.org/wiki/Alpine_setup_scripts#Partitioning|tell setup-disk]] to install with the prepared mountpoint as target.




(Beta refenence: [https://beta.docs.alpinelinux.org/user-handbook/0.1a/Installing/manual.html#_finalizing beta.docs.alpinelinux.org] )


= Using setup-alpine (setup-disk) features =

== For "diskless" and "data" disk mode installs ==

This requires some manual steps, as long as the setup-scripts only auto-detect usable partitions on *removable* devices (not internal harddisks), and they do not support creating new, writable partitions to be used by Alpine.

How to prepare writable storage space is explained in the section about [[Setting_up_disks_manually#Manual_partitioning| manual partitioning]] below.

== For "sys" disk mode installs ==

{{Draft|Some information on this page may be incomplete or outdate.}}

=== Disk layouts ===

It is possible to have one or more of RAID, encryption, and/or LVM on your {{Path|/}} (root) volume. However, the Alpine init script only knows how to handle them when they're layered in that order, and your initram and extlinux.conf file in the {{Path|/boot}} partition are configured properly.

Your {{Path|/boot}} cannot reside on an encrypted or LVM volume, at least not with Alpine's default bootloader (extlinux). (Grub2 can deal with {{Path|/boot}} being on an LVM volume.) The usual practice is to create a small partition for {{Path|/boot}}, and then devote the rest of your disk to a separate partition on which you layer one or more of RAID, encryption, and/or LVM.

Sometimes {{Path|/boot}} is also setup as a mirrored (RAID1) volume, however this is just for post-init access. That way, when you write a new kernel or bootloader config file to {{Path|/boot}}, it gets written to multiple physical partitions. During the pre-init, bootloader phase, only one of those partitions will be read from.

So, typical setups might look like this:

<pre>
One-disk system
---------------
  +------------------------------------------------+
  |  small partition (32--100M), holding           |
  |  only /boot, filesystem needn't be journaled   |
  +------------------------------------------------+
  |  rest of disk in second partition              |
  |  +------------------------------------------+  |
  |  | cryptsetup volume                        |  |
  |  |  +-------------------------------------+ |  |
  |  |  |  LVM PV, containing single VG,      | |  |
  |  |  |  containing multiple LVs, holding   | |  |
  |  |  |  /, /home, swap, etc                | |  |
  |  |  +-------------------------------------+ |  |
  |  +------------------------------------------+  |
  +------------------------------------------------+


Two-disk system
---------------
  +------------------------------------------------+  +------------------------------------------------+
  |  small partition (32--100M), holding           |  |  small partition (32--100M), holding           | These 2 partitions might
  |  only /boot, filesystem needn't be journaled   |  |  only /boot, filesystem needn't be journaled   | form a mirrored (RAID1)
  +------------------------------------------------+  +------------------------------------------------+ volume
  |  rest of disk in second partition              |  |  rest of disk in second partition              |
  | T================================================================================================T | These 2 partitions form
  | T +--------------------------------------------------------------------------------------------+ T | a second mirrored
  | T | cryptsetup volume                                                                          | T | (RAID1) volume
  | T |  +---------------------------------------------------------------------------------------+ | T |
  | T |  | LVM PV, containing single VG,                                                         | | T |
  | T |  | containing multiple LVs, holding                                                      | | T |
  | T |  | /, /home, swap, etc                                                                   | | T |
  | T |  +---------------------------------------------------------------------------------------+ | T |
  | T +--------------------------------------------------------------------------------------------+ T |
  | T================================================================================================T |
  |                                                |  |                                                |
  +------------------------------------------------+  +------------------------------------------------+

</pre>

In a three-disk system, the {{Path|/boot}} would still be RAID1, but the larger partition might in that case be RAID5.


=== RAID ===

<code>setup-disk</code> will automatically build a RAID array if you supply the '''-r''' switch, or if you specify more than one device.

If you instead want to build your RAID array manually, see [[Setting up a software RAID array]]. Then you can add additional layers of encryption and/or LVM, or just assemble the RAID array, and supply the {{Path|/dev/md<i>i</i>}} device directly to [[setup-disk]]. When you're finished, be sure to disassemble the RAID array before rebooting.

If <code>setup-disk</code> sees that you're using RAID---either because you gave it the <code>-r</code> switch, or multiple devices, or a {{Path|/dev/md<i>i</i>}} device---then it will setup your initramfs and extlinux.conf file properly. However, in other cases, such as when you're also using encryption, or you invoke <code>setup-disk</code> with a mounted directory argument, these might not be properly setup for RAID. In that case, you may need to manually edit/rebuild them. The following assumes that <code>$MNT</code> holds the root directory you're installing into:

{{Cmd|1=echo "/sbin/mdadm" > $MNT/etc/mkinitfs/files.d/raid
echo "/etc/mdadm.conf" >> $MNT/etc/mkinitfs/files.d/raid
&#35; edit $MNT/etc/mkinitfs/mkinitfs.conf to make sure features="..."
&#35; includes raid (this field is space-separated and quoted)
mkinitfs -c $MNT/etc/mkinitfs/mkinitfs.conf -b $MNT
&#35; edit $MNT/etc/update-extlinux.conf to make sure modules=... contains
&#35; raid1 or raid456 (whichever your / is on; this field is comma-separated)
&#35; also check the root= setting
extlinux --raid --install $MNT/boot --update
}}

{{Todo|Does adding the <code>--update</code> option to <code>extlinux ...</code> suffice to make {{Path|/boot/extlinux.conf}} be regenerated? Or do we need to manually tweak that file, or run <code>update-extlinux</code>, as well?}}

You might also need to manually tweak {{Path|$MNT/etc/fstab}}. And you might need to copy {{Path|/usr/share/syslinux/mbr.bin}} to your disk's MBR.

=== Encryption ===

See [[Setting up encrypted volumes with LUKS]]. Then you can add an additional layer of LVM, or just unlock the volume you've created (using <code>cryptsetup luksOpen ...</code>), and supply the {{Path|/dev/mapper/<i>something</i>}} device directly to [[setup-disk]]. When you're finished, be sure to relock the volume (using <code>cryptsetup luksClose ...</code>) before rebooting.

If you install your {{Path|/}} (root) on an encrypted volume, you'll need to manually edit/rebuild your initram and your extlinux.conf file. The following assumes that <code>$MNT</code> holds the root directory you're installing into, that you've created the cryptvolume on the device {{Path|/dev/md2}}, and that you want to unlock the encrypted volume into a virtual volume named "crypt":

{{Cmd|1=&#35; edit $MNT/etc/mkinitfs/mkinitfs.conf to make sure features="..."
&#35; includes cryptsetup (this field is space-separated and quoted)
mkinitfs -c $MNT/etc/mkinitfs/mkinitfs.conf -b $MNT
&#35; edit $MNT/etc/update-extlinux.conf to make sure default_kernel_opts="..."
&#35; contains cryptroot=/dev/md1 and cryptdm=crypt (this field is also space-separated and quoted)
&#35; also check the root= setting
extlinux --install $MNT/boot --update
}}

{{Todo|Does adding the <code>--update</code> option to <code>extlinux ...</code> suffice to make {{Path|/boot/extlinux.conf}} be regenerated? Or do we need to manually tweak that file, or run <code>update-extlinux</code>, as well?}}

You might also need to manually tweak {{Path|$MNT/etc/fstab}}.



=== LVM ===

<code>setup-disk</code> will automatically build and use volumes in a LVM group if you supply the '''-L''' switch.

If you instead want to build your LVM system manually, see [[Setting up Logical Volumes with LVM]]. Then <code>vgchange -ay</code>, format and mount your volumes, and supply the root mountpoint to [[setup-disk]]. When you're finished, be sure to
{{Cmd|umount ...
vgchange -an}}
before rebooting.


If <code>setup-disk</code> sees that you're using LVM---perhaps because you gave it the <code>-L</code> switch---then it will setup your initram and extlinux.conf file properly. However, in other cases, these might not be properly setup. In that case, you may need to manually edit/rebuild them. The following assumes that <code>$MNT</code> holds the root directory you're installing into:

{{Cmd|1=&#35; edit $MNT/etc/mkinitfs/mkinitfs.conf to make sure features="..."
&#35; includes lvm (this field is space-separated and quoted)
mkinitfs -c $MNT/etc/mkinitfs/mkinitfs.conf -b $MNT
&#35; edit $MNT/etc/update-extlinux.conf to make sure root= is set correctly
extlinux --install $MNT/boot --update
}}

{{Todo|Does adding the <code>--update</code> option to <code>extlinux ...</code> suffice to make {{Path|/boot/extlinux.conf}} be regenerated? Or do we need to manually tweak that file, or run <code>update-extlinux</code>, as well?}}

You might also need to manually tweak {{Path|$MNT/etc/fstab}}.




= Manual partitioning =


== For "diskless" and "data" disk mode installs ==

The enviroment of the installation images may only have the very basic busybox build-in <code>fdisk</code> command available. However, alpine allows to install more complete partitioning tool packages like <code>sfdisk</code> (stardard fdisk), <code>gptfdisk</code>, <code>parted</code> (CLI), <code>cfdisk</code> (text menus), or even <code>gparted</code> (after [[Installation#Post-Install|setting up a graphical environment]]).

The following example installs <code>cfdisk</code>, looks for existing partitions, and starts to partition <code>/dev/sdX</code>:
 # apk add cfdisk
 # blkid
 /dev/sdX1: LABEL="some" UUID="..." TYPE="vfat"
 /dev/sdX2: LABEL="other" UUID="..." TYPE="ext4"
 # cfdisk /dev/sdX


* Be aware of this Bug: [https://gitlab.alpinelinux.org/alpine/aports/-/issues/11589 The APKOVL loading of diskless setups doesn't work on btrfs and xfs filesystems, or nvme-based devices] => So, for the moment, use eg. only ext4 filesystem partitions on classic drives to store diskless mode states.

The following creates an ext4 partition with disabled journaling, to reduce write operations and allow the disk to spin down after the .apkovl and the packages have been read from the partition during the boot.
 # apk add e2fsprogs
 # mkfs.ext4 -O ^has_journal /dev/sdXY   # the "-O ^has_journal" DISABLES journaling ("^" meaning "not")

{{Note|IMPORTANT! There may still be further steps necessary to make proper use of the partition, see
[[Alpine_local_backup#Saving_and_loading_ISO_image_customizations|Saving_and_loading_ISO_image_customizations]].
}}

== For "sys" disk mode installs ==

<code>setup-disk</code> will by default set up a root partition, a separate /boot partition and a swap. If you want a different layout you can manually create the partitions, filesystems and mount them up on {{Path|/mnt}} (or any other mount point) and then run:

{{Cmd|setup-disk /mnt}}

<code>setup-disk</code> will install your running system on the mounted root, detect your file system layout and generate a fstab.

If you're going for MBR, you are responsible for making the proper partition bootable and make sure the MBR is OK for extlinux. If you're going for EFI, make sure you set <code>BOOTLOADER=grub</code> and <code>USE_EFI=1</code> and {{Pkg|grub-efi}} and {{Pkg|efibootmgr}} are installed before running <code>setup-disk</code>.

See also [https://github.com/itoffshore/alpine-linux-scripts setup-partitions]

=== Dual-booting ===
See [[Installing Alpine on HDD dualbooting|Install to HDD with dual-boot]]

=== Other needs ===
* [[Installing Alpine Linux in a chroot]]
* [[Replacing non-Alpine Linux with Alpine remotely]]


<!--
 Create partition with with type "Linux" (83).
 apk_add e2fsprogs rsync
 mkfs.ext3 /dev/hda1
 mount -t ext3 /dev/hda1 /mnt
 ROOT=/mnt apk_add uclibc busybox apk-tools alpine-baselayout alpine-conf
 # Install busybox links
 mkdir /mnt/proc && mount --bind /proc /mnt/proc && chroot /mnt /bin/busybox --install -s && umount /mnt/proc
 # Copy the apk repository
 rsync -ruav /media/cdrom/apks /mnt
 mkdir /mnt/etc/apk && echo "APK_PATH=file://apks" > /mnt/etc/apk/apk.conf
 # Copy the hd/ext3 initramfs image, kernel and kernel modules
 rsync -ruav /media/cdrom/kernel/generic/hd-ext3.gz /media/cdrom/kernel/generic/bzImage /mnt
 rsync -ruav /lib/modules/* /mnt/lib/modules/
-->


<!--
== Setting up the RAID ==
Set up a raid array as described [[Setting up a software RAID1 array|here]].
In this document two raid arrays are configured: md0 for swap (512MB) and md1 for /var. 

== Create filesystem ==
We need to install the software to create the filesystem ("format" the partition).
 apk_add e2fsprogs

If you use an Alpine release older than 1.3.8 you will need to manually create a link to /etc/mtab.
 ln -fs /proc/mounts /etc/mtab

Create the filesystem. The -j option makes it ext'''3'''. Without the -j option it will become non-journaling ext'''2'''. This step might take some time if your partition is big.
 mke2fs -j /dev/md1


Now edit /etc/fstab and add your new partitions. Mine looks like this:
 none            /proc           proc    defaults 0 0
 none            /sys            sysfs   defaults 0 0
 udev            /dev            tmpfs   size=100k 0 0
 none            /dev/pts        devpts  defaults 0 0
 tmpfs           /dev/shm        tmpfs   defaults 0 0
 /dev/cdrom      /media/cdrom    iso9660 ro 0 0
 /dev/fd0        /media/floppy   vfat    noauto  0 0
 /dev/usba1      /media/usb      vfat    noauto  0 0
 none            /proc/bus/usb   usbfs noauto 0 0
  
 /dev/md0        swap            swap    defaults 0 0
 /dev/md1        /var            ext3    defaults 0 0

== Move the data ==
Now you should stop all services running that put anything in /var (syslog for example). If you have booted on a clean installation and not run setup-alpine, then no services should be running. However, some packages might have created dirs in /var so we need to backup /var mount the new and move all backed up dirs back to the raided /var.

 mv /var /var.tmp
 mkdir /var
 mount /var
 mv /var.tmp/* /var
 rmdir /var.tmp

Verify that everyting looks ok with the ''df'' utility.
 ~ $ df
 Filesystem           1k-blocks      Used Available Use% Mounted on
 none                    255172     23544    231628   9% /
 udev                       100         0       100   0% /dev
 /dev/cdrom              142276    142276         0 100% /media/cdrom
 /dev/md1              37977060    181056  35866876   1% /var

== Survive reboots ==
Now we have everything up and running. We need to make sure that everything will be restored during next reboot.

Create an initscript that will mount /var for you during boot. I call it /etc/init.d/mountdisk and it looks like this:
 #!/sbin/runscript
 
 start() {
         ebegin "Mounting /var"
         mount /var
         eend $?
 }
 
 stop() {
         ebegin "Unmounting /var"
         umount /var
         eend $?
 }

Make it exectutable:
 chmod +x /etc/init.d/mountdisk

'''NOTE:''' Since Alpine-1.7.3 there is a ''localmount'' script shipped so you will not need to create your own ''mountdisk'' script.

And that /var is mounted *after* raid is created. The -k option will make alpine to unmount the /Var partition during boot. Also add start of swap too boot
 rc_add -k -s 06 mountdisk
 rc_add -k -s 06 swap

The /dev/md* device nodes will not be created automatically so we need to put the on floppy too.
 lbu include /dev/md*

If you have users on the server and want /home to be permanent, you can create a directory /var/home and create links to /var/home.
 mkdir /var/home
 mv /home/* /var/home/
 ln -s /var/home/* /home/

'''NOTE:''' You cannot just replace /home with a link that points to /var/home since the base has a /home directory. When the boot tries to copy the config from floppy it will fail because of the already existing /home directory.

Make sure the links are stored to floppy:
 lbu include /home/*

Also remember to move any newly created users to /var/home and create a link:
 adduser bob
 mv /home/bob /var/home/
 ln -s /var/home/bob /home/bob
 lbu include /home/bob

Save to floppy:
 lbu commit floppy

== Test it works ==
Reboot computer. Now should the raid start and /var should be mounted. Check with df:
 ~ $ df
 Filesystem           1k-blocks      Used Available Use% Mounted on
 none                    255172     23976    231196   9% /
 mdev                       100         0       100   0% /dev
 /dev/cdrom              140932    140932         0 100% /media/cdrom
 /dev/md1              37977060    180984  35866948   1% /var

== Upgrades ==
Since the package database is placed on disk, you cannot update by simply replacing the CDROM. You will have to either run the upgrade on the new CDROM or run ''apk_add -u ... && update-conf'' manually.
-->

== Setting up swap ==

# create partition with type "linux swap" (82) (If you're going to use an LVM logical volume for swap, skip this step and <code>lvcreate</code> that instead.)
# <code>mkswap /dev/sda2</code>
# <code>echo -e "/dev/sda2 none swap sw 0 0" >> /mnt/etc/fstab</code>
# <code>swapon /dev/sda2</code> (or <code>rc-service swap start</code>)

Then {{Cmd|free -m}} will show how much swap space is available (in MB).

If you prefer maximum speed, you don't need configure any raid devices for swap. Just add 2 swap partitions on different disks and linux will stripe them automatically. The downside is that at the moment one disk fails, the system will go down. For better reliability, put swap on RAID1. 

{{Todo|Instructions for cryptswap?}}


[[Category:Installation]]
[[Category:Storage]]