source: postlfs/config/bootdisk.xml@ 9a9a0fec

<sect1 id="postlfs-config-bootdisk">
<?dbhtml filename="bootdisk.html" dir="postlfs"?>
<title>Creating a Custom Bootdisk</title>

<sect2>
<title>Decent Rescue Bootdisk Needs</title>
<para>This section is really about creating a <emphasis>rescue</emphasis>
diskette.  As the name <emphasis>rescue</emphasis> implies, the host
system has a problem, often lost partition information or corrupted file
systems, that prevents it from booting and/or operating normally.  For
this reason, you <emphasis>must not</emphasis> depend on resources from
the host being "rescued".  To presume that any given partition or hard
drive <emphasis>will</emphasis> be available is a risky presumption.</para>

<para>Heeding the warning, the rescue disk created here has no
dependency on the host system's resources, other than basic bootability
and hardware soundness.  At a minimum, the most common sorts of failures
requiring a rescue bootdisk should be addressed by the contents of the
bootdisk.  This would include the common loss of partitioning (master
boot record is lost or corrupted), file system corruption, and the need
to allow creation and editing of files that may have been lost or
corrupted, possibly as an effect of the other two problems.</para>

<para>Additional utilities should be available to search for text or
files, copy, move and remove files, and many other normal operations
that might be expected to be needed when reconstructing.</para>
</sect2>

<sect2>
<title>This Minimal Decent Rescue Disk</title>

<para>The intent here is to create a "rescue bootdisk" that will support
the common operations listed above.  These functions are provided by
including selected executables from <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application> and <application><ulink
url="http://freshmeat.net/projects/e2fsprogs/">e2fsprogs</ulink></application>.
A basic editor and rudimentary disk partitioning utility may also be
optionally included.</para>

<para>This, however, is not the limit.  A minimal disk is described
here, but you can add anything you can fit on the floppy.  Furthermore,
if one floppy is not enough to meet your needs, you can make a
multi-diskette rescue set that means, essentially, the sky is the limit.
This is discussed below.  The number of other possible variations are
just too numerous to mention here.</para>
</sect2>

<sect2>
<title>Build the Rescue Bootdisk</title>
<sect3>
<title>Prerequisites</title>

<para>You should have known-good floppy diskettes available.  Some
people prefer to use the <command>fdformat</command> command to prepare
these because it also does a verification. See the man page for more
details.  Another good idea is to always prepare duplicates of the
rescue diskette.  Media does deteriorate.</para>

<para>These instructions presume a base <acronym>LFS</acronym> install
using ext2/ext3 file systems.</para>

<para>You need to have loopback device support enabled in your host's
kernel to use this procedure.</para>

<para>You should make a custom kernel that includes only those features
needed to rescue your system, so it will have the least size.  No
sense in building in support for things like <application>xfree86</application>,
<application>dri</application>, etc, as most rescues are performed from the
command prompt.  Along the same lines, if you have <xref linkend="gcc2"/>, it 
is known to produce smaller kernels.  So you might want to use that
compiler for this kernel.  If you do so, don't overlook any loadable
modules (which are not addressed here) you might need - they need to be
compiled with same compiler used to make the kernel.</para>

<para>The rescueimage must include support for the file system of your
choice (we presume ext2/3 here), ramdisk and initial ramdisk (initrd).
Disable everything that you can in the kernel configuration. You should
keep support for the proc file system and tempfs file system enabled
because of their general utility. The proc file system is needed for
the <command>mount</command> to report properly.</para>

<para>If you install <emphasis>only</emphasis> the minimal set of
components shown in this document, you will need a kernel that is 643 or
fewer blocks in size.  If you want the optional programs - a very basic
editor, like <command>ed</command>, and rudimentary disk partitioning, like 
<command>sfdisk</command> - the kernel will need to be 595 or fewer blocks in 
size.  This should not be a major problem unless your needs are fairly 
esoteric.  On the system used to develop this version of the procedures, using 
only ext2 file systems and not using networking or <acronym>CD</acronym>s for 
recovery, the kernel image is only 481 blocks.  And there may be more to gain - it has not been closely examined for additional gains.</para>

<para>This kernel image will be called "rescueimage" hereinafter. You
can actually name your image anything you like and just use its name
instead in any commands that include "rescueimage".</para>

<para>If you can not get your kernel down to the size needed to allow
all you need on the ramdisk image, don't fret.  You can always build a
two diskette set, one boot and one root diskette.  The kernel will prompt
you to insert the root file system diskette.  This will allow room for a
compressed ramdisk image of 1440 blocks and a kernel of the same
size.</para>

<para>The kernel size limits given above are likely to vary as
local system-specific configurations change.  Use them only as a
guideline and not as gospel.  The size of the kernel image as shown by
<command>ls -sk</command> is only an approximation because of some
"overhead".  On the system used to develop this version of these
procedures, that command shows 488 blocks but the actual number of
blocks written is only 480 and a fraction, which means that 481 blocks
are actually used.</para>
</sect3>

<sect3>
<title>Rescue Disk Build Process</title>

<para>The basic process will be:</para>
<itemizedlist>
<listitem><para>make a mount point for a file system</para></listitem>
<listitem><para>make an empty file to hold the file system</para></listitem>
<listitem><para>bind the empty file to a loopback device</para></listitem>
<listitem><para>make a 4MB file system</para></listitem>
<listitem><para>mount the file system</para></listitem>
<listitem><para>add components to the file system</para></listitem>
<listitem><para>make the compressed initrd</para></listitem>
<listitem><para>join a kernel image and initrd onto a diskette</para></listitem>
</itemizedlist>

<para>The initial ramdisk will be automatically loaded at boot time if
setup is done correctly.</para>

<para><emphasis>Make a mount point and an empty file to hold a file
system</emphasis></para>

<screen><userinput><command>mkdir -p /mnt/loop1
dd if=/dev/zero of=/tmp/rfloppy bs=1k count=4096</command></userinput></screen>

<para><emphasis>Command explanations</emphasis></para>

<para><command>dd</command>: This is a generalized input-to-output copy
utility that also has many transformation capabilities.</para>

<para><command>if=/dev/zero</command>: This parameter assigns
<command>dd</command>'s input file to a device that returns an infinite
stream of zeroes.</para>

<para><command>of=/tmp/rfloppy</command>: This parameter directs
<command>dd</command>'s output to <filename>/tmp/rfloppy</filename>.</para>

<para><command>bs=1k count=4096</command>: These parameters tell
<filename>dd</filename> to read and write in "chunks" of 1024 bytes and
process 4096 "chunks".</para>

<para><emphasis>Bind the file to a loopback device, make a file system and mount it.</emphasis></para>

<para>The reason these commands are used is that they work regardless of
the version of <command>mount</command> (older ones don't have the
<userinput>-o loop</userinput> option) or if <filename>/etc/mtab</filename>
is symlinked to <filename>/proc</filename> (which causes mount to be
unable to properly "unbind" a loop device, due to "lost" information).
An alternate set of commands is provided, after these three commands,
that you can use if you don't have either of these situations.</para>

<screen><userinput><command>losetup /dev/loop1 /tmp/rfloppy
mke2fs -m 0 -N 504 /dev/loop1
mount -t ext2 /dev/loop1 /mnt/loop1</command></userinput></screen>

<para><emphasis>Command explanations</emphasis></para>

<para><command>losetup /dev/loop1 /tmp/rfloppy</command>: This
command "binds" a loopback device to the empty file.</para>

<para><command>mke2fs -m 0 -N 504 /dev/loop1</command>: This
command makes an ext2 file system on the loopback device (which really
means it is created in the file to which the loopback device is bound)
and reserves no blocks.  The <userinput>-N 504</userinput> parameter causes 
only 504 inodes to be allocated, leaving more space for other things needed 
in the file system.</para>

<para><command>mount -t ext2 /dev/loop1 /mnt/loop1</command>: This
mounts the file system just created, just as if it were a real device,
like a hard drive or diskette.  This allows all the normal system I/O
commands to operate as if a real device were present.</para>

<para>If your <command>mount</command> supports the
<userinput>-o loop</userinput> option <emphasis>and</emphasis> your
<filename>/etc/mtab</filename> is a real file, rather than a symlink to
<filename>/proc</filename>, the three above commands can be replaced
by these next two commands.</para>

<screen><userinput><command>mke2fs -F -m 0 -N 504 /tmp/rfloppy
mount -o loop /tmp/rfloppy /mnt/loop1</command></userinput></screen>

<para><emphasis>Command explanations</emphasis></para>

<para><command>mke2fs -F -m 0 -N 504 /tmp/rfloppy</command>: As before,
a file system is made, with only 504 inodes and no reserved blocks, that
will be bound to a loopback device.  The <userinput>-F</userinput> parameter
just suppresses an irritating question issued when
<command>mke2fs</command> realizes that you are not accessing a
device.</para>

<para><command>mount -o loop /tmp/rfloppy /mnt/loop1</command>: This
command tells <command>mount</command> to bind the named file to a
loopback device it automatically selects (the first available) and mount
the device on <filename>/mnt/loop1</filename>.</para>

<para><emphasis>Add components to the file system</emphasis></para>

<para><emphasis>A cautionary note:</emphasis> if you are not running in a
<command>chroot</command> environment, be sure that you do not accidently
omit the <filename>/mnt/loop1</filename> reference in the commands. If
you do so, you might replace the equivalent components on your host with
the components that are installed by these procedures.  Even if you are
in a chroot environment, you may need to be careful if the environment
is your freshly built <acronym>LFS</acronym> system which you intend to
use as a host in the future.</para>

<para>First, to have as much free space as possible, remove the
<filename>lost+found</filename> directory, which is not needed because
it is only used by <command>fsck</command>.  Since <command>fsck</command> 
will never be run on this file system, it is unneeded.</para>

<screen><userinput><command>rmdir /mnt/loop1/lost+found/</command></userinput></screen>

<para>Now make a minimal set of directories.</para>

<screen><userinput><command>mkdir /mnt/loop1/{dev,proc,etc,sbin,bin,lib,mnt,usr,var}</command></userinput></screen>

<para>Add needed device files to the initrd image.  If you use devfs,
the following command works well, as you only have the devices you use
anyway.</para>

<screen><userinput><command>cp -dpR /dev/* /mnt/loop1/dev</command></userinput></screen>

<para>If you used <command>MAKEDEV</command> to create your devices on
your host, you'll want to use something similar to this longer command,
to minimize wasting space with unneeded inodes.</para>

<para><emphasis>You must modify this to suit your kernel configuration and
other needs.</emphasis>  For example, you may need scsi devices and may not need
frame buffer devices or the pseudo-terminal directory.  Also, the number
of hard drives and partitions that you include should be the minimal
that you need.  Extensive analysis has not been done on the list below,
so there are more inodes and space to be gained by "fine tuning" this
set.</para>

<screen><userinput><command>mkdir /mnt/loop1/dev/pts
cp -a \ 
&nbsp;&nbsp;&nbsp;&nbsp;/dev/null /dev/console \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/fb[0-7] /dev/fd /dev/fd0 /dev/fd0h1440 /dev/full \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/hda* /dev/hdb* /dev/hdc* /dev/hdd* /dev/initctl /dev/kmem \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/loop[0-3] /dev/lp0 /dev/mem /dev/port \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/psaux /dev/ram \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/ram0 /dev/ram1 /dev/ram2 /dev/ram3 /dev/random /dev/rtc \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/shm /dev/stderr /dev/stdin /dev/stdout /dev/tty \
&nbsp;&nbsp;&nbsp;&nbsp;/dev/tty[0-9] /dev/ttyS0 /dev/ttyS1 /dev/urandom /dev/zero \
&nbsp;&nbsp;/mnt/loop1/dev</command></userinput></screen>

<para><emphasis>What is needed in the <filename>/etc</filename>
directory</emphasis></para>

<para>If you choose, you can copy all or selected parts of your
<filename>/etc/passwd</filename> and <filename>/etc/group</filename>
files.  But even if each is less than 1024 bytes, you will lose two
inodes and two blocks of space on the initial ramdisk.  This only really
matters because of trying to squeeze everything onto a 1.44MB
diskette.  Every little bit helps.  The strategy taken here is to create
these two files as part of the rescue boot and initialization process.
The commands that make the two files will be imbedded inside the
<filename>rcS</filename> script that <filename>linuxrc</filename>
(really <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application>) invokes after the initrd is
loaded.  This way no more inodes or blocks are used on the
diskette to carry these files.</para>

<para>Some might like to copy their <filename>/etc/rc*</filename>
directory into the ramdisk image, but this may have no value, other
than archival use, in a worst-case recovery scenario.  If you want
automatic initialization of the system after repair, they may have some
value. But few people need or want this to happen. If the file system
on the hard drives are corrupted, what good will mount scripts do? Some
scripts may be useful, like access to a network to copy over backup data
when the hard drive's file systems are usable again.  The point is that
you should copy only the parts that you can use because space is at a
premium on the diskette.</para>

<para>Here, only the <filename>fstab</filename> will be included.  This
is handy because it eases mounting of partitions that may be useful and
also can be examined and used as a guide as to what is available and
what may need reconstruction.  Because it may be larger than needed, you
should edit it to remove any useless entries and minimize commentary.
No other editing is needed because the boot scripts are not included and
no automatic mounting will be done using the <filename>fstab</filename>.
If you decide to include some bootscripts that might try to mount
things, change the <filename>fstab</filename>'s entries to
<command>noauto</command> in the options field so they don't cause an
attempt to mount a potentially corrupt partition.  Copy it to
<filename>/tmp</filename>, edit it as desired and then:</para>

<screen><userinput><command>cp -a /tmp/fstab /mnt/loop1/etc</command></userinput></screen>

<para>Now the initialization script will be added.  As mentioned above,
<command>linuxrc</command> is symlinked to <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application>.
After the kernel and initial ramdisk have been loaded, the kernel gives
control to <command>linuxrc</command> (<application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application>).  It wants to run an <filename>/etc/init.d/rcS</filename>
script to do any initial setup.</para>

<para>If you use devfsd, you will need to set up the
<filename>rcS</filename> script to handle the devfsd startup.  Put the
following commands in <filename>/mnt/loop1/etc/init.d/rcS</filename>.
You may also want to add some of the processes shown in the non-devfs
version that follows.</para>

<screen><userinput>#!/bin/sh
mount -t devfs devfs /dev
/sbin/devfsd /dev</userinput></screen>

<para>If you don't use devfsd, but created a static <filename>/dev</filename> 
directory using <command>MAKEDEV</command>, or any similar process, the
<filename>rcS</filename> script will do slightly different things.
Also, don't forget that it is creating the
<filename>/etc/passwd</filename> and <filename>/etc/group</filename>
files, thus saving space on the diskette.</para>

<para>The script made next will mount <filename>/proc</filename>, turn
on swap (no harm is done if it fails), make the
<filename>/etc/passwd</filename> and <filename>/etc/group</filename>
files, create a log directory and turn on swapping.  Create the script
with:</para>

<screen><userinput><command>mkdir -p /mnt/loop1/etc/init.d
cat &gt;/mnt/loop1/etc/init.d/rcS &lt;&lt; EOD</command>
#!/bin/sh
mount -t proc proc /proc
swapon -a

echo "root:x:0:0:root:/root:/bin/bash" &gt; /etc/passwd

<command>cat &gt; /etc/group &lt;&lt;EOF</command>
root:x:0:
bin:x:1:
sys:x:2:
kmem:x:3:
tty:x:4:
tape:x:5:
daemon:x:6:
floppy:x:7:
disk:x:8:
lp:x:9:
dialout:x:10:
audio:x:11:
<command>EOF
chmod 644 /etc/passwd /etc/group</command>

mkdir /var/log

<command>EOD
chmod u+x /mnt/loop1/etc/init.d/rcS</command></userinput></screen>

<para>Unless you add a lot to this script, which <emphasis>is</emphasis>
encouraged, the above should be reasonably close to what you need.</para>

<para><emphasis>Install packages</emphasis></para>

<para>There are two packages that must be installed.  The <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox</ulink></application>
package incorporates the core functions that provide a shell and many
basic utilities.  A file system package, like <application><ulink
url="http://freshmeat.net/projects/e2fsprogs/">e2fsprogs</ulink></application>, or
a package for the file system you are using will provide a minimal
set of utilities for file system checking and reconstruction.  The whole
package will not be installed, but only certain needed components.</para>

<para>If you use devfsd, you will also need to install that software.</para>

<para>Install <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox</ulink></application>
into the initial ramdisk image.  Busybox incorporates many UNIX utility
program functions into a single small executable file.</para>

<screen><userinput><command>make &amp;&amp;
make PREFIX=/mnt/loop1 install &amp;&amp;
&gt; /mnt/loop1/var/utmp</command></userinput></screen>

<para>A <filename>var/utmp</filename> is made because <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox</ulink></application>
needs it for the reboot command to work properly. If this file doesn't
exist when <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox</ulink></application>
is started, the reboot command will not work. This would be a bad thing
for people that have no reset button available to them.</para>

<para>If you use devfs to create devices on the fly and free up precious
inodes on the floppy, you'll also install devfsd to facilitate the
devices that <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox</ulink></application>
expects to find. Use the following commands to do the install.</para>

<screen><userinput><command>mv GNUmakefile Makefile &amp;&amp;
make &amp;&amp;
make PREFIX=/mnt/loop1 install &amp;&amp;</command></userinput></screen>

<para><emphasis>Install part of <application>e2fsprogs</application></emphasis></para>

<para>If you use the ext2 or ext3 (journaling) file system, you can use
the commands below to install the minimal functionality that should
allow you to get your hard drives usable again.  If you use ext3, keep in
mind that it is a part of the <application>e2fsprogs</application>
package and you can get the components, which are mostly hard links,
from the same places shown below.  If you use some other file system,
such as reiserfs, you should apply the <emphasis>principals</emphasis>
you see here to install parts of that package instead.</para>

<screen><userinput><command>LDFLAGS='-s'
mkdir build &amp;&amp;
cd build &amp;&amp;
../configure --prefix=/mnt/loop1/usr --with-root-prefix="" \
    --disable-swapfs --disable-debugfs \
    --enable-dynamic-e2fsck --disable-nls --disable-evms \
    --disable-rpath &amp;&amp;
make LDFLAGS="$LDFLAGS" &amp;&amp;
strip -p --strip-unneeded --remove-section=.comment \
    -o /mnt/loop1/sbin/mke2fs misc/mke2fs &amp;&amp;
strip -p --strip-unneeded --remove-section=.comment \
    -o /mnt/loop1/sbin/e2fsck e2fsck/e2fsck &amp;&amp;
chmod 555 /mnt/loop1/sbin/{mke2f,e2fsck}
</command></userinput></screen>

<para><emphasis>Two useful utilities</emphasis></para>

<para>There are two very useful utilities that any rescue disk should
have to help in faster and more accurate recovery.  The first is a
partitioning utility.  The <command>sfdisk</command> program is
used here because of its small size and great power.  Be warned though -
it is not what is considered to be "user friendly".  But the
<command>fdisk</command> and <command>cfdisk</command> programs are
substantially larger or require more shared objects, like <application>ncurses
</application>.</para>

<para>The second utility is an editor.  Most graphical editors are
inherently too large and also require additional shared objects.  For
this reason, <command>ed</command> is used here.  It is small, requires
no additional shared objects and is a regex-based editor that is the
ancestor to almost all subsequent editors that support regex-based
editing, whether graphical or not.  It is a "context editor" and offers
powerful, but non-graphical, editing features. There are many other
editors that may be suitable - feel free to use one of them instead.</para>

<para>Read the <application>busybox</application>
<filename>INSTALL</filename> and <filename>README</filename> files to
see how to include a <command>vi</command> editor. It has not been
investigated here yet, so it may or may not easily fit onto a single
diskette image such as is made here.</para>

<para>You can install these or not, but it is important for you to have
some capability such as these offer.  Exactly how you would install the
utilities you choose will have to be determined by you.</para>

<para><command>Sfdisk</command> and <command>ed</command> are installed
by, essentially, copying them from your host.  Strip is used, just to
assure that they carry no "excess baggage", even though the base
<acronym>LFS</acronym> install should have stripped them already.  Use
the following commands:</para>

<screen><userinput><command>strip -p --strip-unneeded --remove-section=.comment \
    -o /mnt/loop1/sbin/sfdisk /sbin/sfdisk
strip -p --strip-unneeded --remove-section=.comment \
    -o /mnt/loop1/bin/ed /bin/ed
chmod 555 /mnt/loop1/sbin/sfdisk /mnt/loop1/bin/ed</command></userinput></screen>

<para>Also, keeping in mind your space limitations, copy any other
binaries and libraries you need to the image.  Use the
<command>ldd</command> command to see which libraries you will need to
copy over for any executables.  Don't forget to also strip them
<emphasis>before</emphasis> copying them to the ramdisk image or use the
<command>strip</command>, as above, to "copy" them.</para>

<para><emphasis>Set up the lib directory</emphasis></para>

<para>Once you have installed all the utilities from above and any
additional ones you want, use the <command>ldd</command> command, as
mentioned above, on those that were not listed in this document.  If
any additional libraries are needed, add them into the setup commands
shown next.</para>

<para>If you installed only those things shown above, the shared objects
needed will be minimal.  You can add them to the ramdisk image with:</para>

<screen><userinput><command>strip -p --strip-unneeded --remove-section=.comment \
&nbsp;&nbsp;&nbsp;&nbsp;-o /mnt/loop1/lib/libc.so.6 /lib/libc-2.3.2.so &amp;&amp;
strip -p --strip-unneeded --remove-section=.comment \
&nbsp;&nbsp;&nbsp;&nbsp;-o /mnt/loop1/lib/ld-linux.so.2 /lib/ld-2.3.2.so &amp;&amp;
strip -p --strip-unneeded --remove-section=.comment \
&nbsp;&nbsp;&nbsp;&nbsp;-o /mnt/loop1/lib/libdl.so.2 /lib/libdl-2.3.2.so &amp;&amp;
chmod 555 /mnt/loop1/lib/{libc.so.6,ld-linux.so.2,libdl.so.2}</command></userinput></screen>  

<para>Note that the above commands change the names of the libraries,
eliminating the need for the usual symlinks. If you add any additional
shared objects, be alert for similar opportunities and also the pitfalls
that may be present.</para>

<para><emphasis>Make the compressed initrd</emphasis></para>

<para>Unmount the loopback file.  If you used <command>mount</command>'s
<userinput>-o loop</userinput> option, the "bond" between the loop device
and the file will be removed when the unmount is done.  Just omit the
<command>losetup -d /dev/loop1</command> from the following
sequence.  The <userinput>-9</userinput> parameter is used with
<command>gzip</command> to make the smallest possible compressed image.  To 
make sure it will fit on the diskette, list the file's size.</para>

<screen><userinput><command>umount /mnt/loop1 &amp;&amp;
losetup -d /dev/loop1 &amp;&amp;  # Omit if mount's -o loop was used
gzip -9 &lt; /tmp/rfloppy &gt; /tmp/rootfs.gz
ls -l /tmp/rootfs.gz</command></userinput></screen>  

<para><emphasis>Join a kernel image and initrd onto a diskette</emphasis></para>

<para>Now the kernel image and initial ramdisk image will be written to
the boot diskette.  Before doing this, calculate the number of blocks
needed for the kernel and for the initrd, individually, by dividing each
size by 1024 and adding one if there is any remainder.  Add these two
results together.  They must total 1,440 or fewer blocks.  If they total
more than this, don't worry too much.  Changes to make a two-diskette
set are presented later.  Of course, you could reexamine your choices and
try to shrink either the kernel or the initial ramdisk image.</para>

<para>To make a single-floppy rescue, using devfs, use the following
commands. If you use the static <filename>/dev</filename> setup, use
<filename>/dev/fd0</filename> instead of the /dev/floppy/0.</para>

<screen><userinput><command>dd if=rescueimage of=/dev/floppy/0 bs=1k
rdev /dev/floppy/0 0,0
rdev -R /dev/floppy/0 0</command></userinput></screen>

<para><emphasis>Command explanations</emphasis></para>

<para><command>rdev /dev/floppy/0 0,0</command>: sets the root file system
the kernel will use when it boots. Because it loads an initrd, it will
automatically set that as the root device, initially. So, the
<userinput>0,0</userinput> gives it "no value", telling the kernel to not
mount any other device. Some folks give <filename>/dev/fd0</filename> or
something similar. But this has effect <emphasis>only</emphasis> when
<command>linuxrc</command> (really <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application>) exits and the normal <command>init</command>
processes get invoked. Since this is not being done here, and the floppy
is <emphasis>not</emphasis> a valid file system, it would be useless
here. A hard drive would be a better choice if you are looking to
automatically bring the system up after repair. Since <application><ulink
url="http://www.busybox.net/downloads/busybox-0.60.4.tar.bz2">busybox
</ulink></application> provides the <command>reboot</command> command,
automatic initialization is not needed.</para>

<para>The <command>rdev -R /dev/floppy/0 0</command> will set the
"root flags" to zero. They have no use in this application.</para>

<para>The <command>dd</command> from above showed some results, like</para>

<screen>        480+1 records in
        480+1 records out</screen>

<para>In this example the rescueimage (kernel) was 480+1 blocks in size.
Make sure that this number, which may be different for you, matches your
calculations from above.  You need to calculate a "magic number" now
that will be inserted into the kernel image.  The value consists of three
significant parts.  Two are discussed here.  The third is touched upon
later.</para>

<para>Bits 0 - 10 will contain the size of the kernel image, in blocks,
that you calculated above, and which should match the results from the
dd above. Bit 14 (the 15th bit, which is 2 to the 14th power, or 16,384)
is a flag that, when set to 1, tells the kernel an initial ramdisk is to
be loaded. So for the single-floppy rescue diskette, the two numbers
16,384 and 481 (or whatever number is right for your kernel size) are
added together to produce a decimal value, like 16865. This value is
inserted into the proper place in the kernel image by the
<command>rdev</command> command done next.</para>

<para>Insert the "magic number" into the kernel image and then write the
root file system right after the kernel on the floppy by executing the
following commands, with the proper numbers inserted. Notice that the
<command>seek</command> parameter's number must be the size, in blocks,
of your kernel image. If you use the static <filename>/dev</filename>
setup, use <filename>/dev/fd0</filename> in the commands below, instead
of <filename>/dev/floppy/0</filename>.</para>

<screen><userinput><command>rdev -r /dev/floppy/0 16865
dd if=/tmp/rootfs.gz of=/dev/floppy/0 bs=1k seek=481</command></userinput></screen>

<para>In this command, <command>seek</command> was used to position to
the block following the end of the kernel (480+1) and begin writing the
root file system to the floppy.</para>
</sect3>
</sect2>

<sect2>
<title>A Two-diskette Rescue Setup</title>

<para>If you just can't live with a single-diskette rescue system, here
is what to do to make a simple two-diskette system. Note that the
endless possibilities presented by the availability of
<command>linuxrc</command> and other components are not addressed
here. Here you will just use the kernel's ability to prompt for a second
diskette that contains the initrd image and load it.</para>

<para>Modify the above instructions as follows. First a different magic
number is needed. The 15th bit (bit 14) still needs to be set, but the
size of the kernel image, in blocks, is replaced with a zero.  The third
component, which was not discussed above, is now used. This is the 16th
bit (bit 15) of the "magic number". When set, it tells the kernel to ask
the user to insert the "root" floppy. It then loads the initrd image
from that diskette.  Because the size of the kernel image was replaced
by zero, the kernel starts loading from the "zero'th" block (the first
one) on the second diskette.</para>

<para>The 16th bit (bit 15) represents 2 raised to the 15th power, or
32,768. So the new magic number is 32,768 + 16384, which is 49,152. This
value tells the kernel to prompt for, and then load, an initial ramdisk
image from the first block on the inserted floppy.  So your first
modification is to the command to write the "magic number" to the kernel
image on the diskette.</para>

<screen><userinput><command>rdev -r /dev/floppy/0 49152</command></userinput></screen>

<para>Note that the initrd image is <emphasis>not</emphasis> copied to
the diskette yet. Remove the boot diskette and insert another diskette
that will hold your root file system. Run this modified command (don't
forget to use <filename>/dev/fd0</filename> if you don't use devfs).
Note that no <command>seek</command> parameter is used.</para>

<screen><userinput><command>dd if=/tmp/rootfs.gz of=/dev/floppy/0 bs=1k</command></userinput></screen>

<para>That's all there is to it. The possibilities from here are limited only
by your imagination and tenacity in pursueing enhancements. And your
willingness to research available documentation. A good starting point
is the "Documentation" directory in your kernel source tree. More help
may be gained at
<ulink url="http://linuxfromscratch.org/hints/news.html"><acronym>LFS </acronym> Hints</ulink>
(please use a mirror site that is suitable) and
<ulink url="http://www.tldp.org">TLDP</ulink>.</para>

</sect2>
</sect1>