View previous topic :: View next topic |
Author |
Message |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Mon Mar 26, 2007 5:25 pm Post subject: Problem booting (self made) LiveCD[SOLVED] |
|
|
I've tried posting about this in another thread, but it got buried without any responses and I don't want to pull that thread off track anymore. I made my own LiveCD and everything seems to be in order, but it won't finish booting. Here are the last few lines it outputs before it locks the system up:
Code: | >> Activating mdev
>> Making tmpfs for /newroot
>> Mediafound on /dev/iseries/vcd*
>> Determining root device
>> Determining looptype
!! Invalid loop location: /livecd.squashfs
!! Please export LOOP with a valid location, or reboot and pass a proper loop=
|
I've searched theses forums and google and tried various things people did to resolve their problems. It seems that this kind of problem has been encountered many, many times before, but none of the "solutions" made any difference for me. I'm starting to think that maybe my CD is fine (and the several coasters I've burned before earlier) and that the issue may be BIOS related or related to the fact that I have a SATA hard drive. Any thoughts. Is there more info. I could post? Is /dev/iseries/vcd* a valid location?
Last edited by sirdilznik on Sat Mar 31, 2007 4:41 pm; edited 1 time in total |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Tue Mar 27, 2007 6:56 pm Post subject: |
|
|
Hi
i had this several times with different scenarions... Don't worry, if the Kernel get loaded, you're on the right track and the bios seems fine!
1. Check that the file livecd.squashfs is on the root of the CD itself. Not in a subdir or something. And really check that the root= or real_root= in the bootloader is correct.
2. Check that you have the right drivers compiled into the Kernel (ide, cd-fs, loopback-dev, squashfs etc.)
3. Check the initrd! On "slow" machines i had the issue that the usb-driver (even when compiled into the Kernel) was loaded a little bit after the mount-attempt was made if booting from usb... so i had to spice up /usr/share/genkernel/generic/initrd.scrits with a nice little sleep command!
4. sometimes you can get a little shell to check if the init of the Kernel does everything right (mount the cd manually, and see if you can even mount the squash manually) by simply adding an init=/bin/sh as boot-parameter!
As it seems, the cd-rom gets properly detected... i would say that either the squashfs isn't compiled into the kernel, or the real_root or root variable from the bootloader is wrong....
I hope this will help you a bit further.... otherwise it would be helpful to know your architecture, kernel-version and if you took genkernel or something?
Btw: what kind of iseries do you drive with sata?!? ...... licking my fingers..... |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Wed Mar 28, 2007 4:47 am Post subject: |
|
|
nativemad wrote: | Hi
i had this several times with different scenarions... Don't worry, if the Kernel get loaded, you're on the right track and the bios seems fine!
1. Check that the file livecd.squashfs is on the root of the CD itself. Not in a subdir or something. | Yes, it is on the root. Quote: | And really check that the root= or real_root= in the bootloader is correct. | The root and real_root are copied verbatim from Gentoo Howto build a LiveCD from scratch Code: | root=/dev/ram0
real_root=/dev/loop0 |
Quote: | 2. Check that you have the right drivers compiled into the Kernel (ide, cd-fs, loopback-dev, squashfs etc.) |
Yes, all those are compiled in. I tried with SATA built directly in and with SATA built as modules. No difference
Quote: | 3. Check the initrd! On "slow" machines i had the issue that the usb-driver (even when compiled into the Kernel) was loaded a little bit after the mount-attempt was made if booting from usb... so i had to spice up /usr/share/genkernel/generic/initrd.scrits with a nice little sleep command! | Not sure how to do that.
Quote: | 4. sometimes you can get a little shell to check if the init of the Kernel does everything right (mount the cd manually, and see if you can even mount the squash manually) by simply adding an init=/bin/sh as boot-parameter! | I will try that when I reboot.
Quote: | As it seems, the cd-rom gets properly detected... i would say that either the squashfs isn't compiled into the kernel, or the real_root or root variable from the bootloader is wrong....
I hope this will help you a bit further.... otherwise it would be helpful to know your architecture, kernel-version and if you took genkernel or something? | amd64 2.6.21-rc4-mm1 (patched with squashfs) I did make all;make modules_install;make install (then renamed things to comply with /boot/grub/menu.lst);genkernel initrd. As for the specs of the system i built this on click my sig.
Quote: | Btw: what kind of iseries do you drive with sata?!? ...... licking my fingers..... | I'm not sure what you mean My motherboard is a nvidia nForce4-4x and the SATA controller is a nvidia nForce4-4x controller.
Thanks for taking the time to answer my query. Your help is much appreciated. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Wed Mar 28, 2007 4:18 pm Post subject: |
|
|
Can you mount the squashfs in your (normal booted) system (or chroot)?
mount -t squashfs -o loop /path/to/livecd.squashfs /somewhere/else
Anything (almost a bit) in dependence to access the CDROM should be IN The Kernel, not as Module.... Perhaps you have to mount the squash from the CD and have a look with lsmod which modules have been loaded...
The root= and real_root= seems to be good... also make shure that the loop=/livecd.squashfs is in place!
If you get the tiny shell via init=/bin/sh, you should probably check that real_root=/dev/loop0 is correct! simply type losetup -a and if you see that you got more than one loop mounted, than this would probably mean that you have to do it like real_root=/dev/loop1 or something.
Genkernel is very easy to understand and edit.... If you know a bit bash, simply edit /usr/share/genkernel/generic/initrd.scrits or /usr/share/genkernel/linuxrc according to your needs.... As example you could open /bin/sh when the cdrom gets detected, just before the squash should get mounted...
I was just very confused about ">> Mediafound on /dev/iseries/vcd* "! Is your cdrom at /dev/iseries/vcd*?!?!?!?!?!? -I thougt perhaps you got one of those very nice IBM-Mainframes because of iseries.... but amd64 is also cool! ![Laughing :lol:](images/smiles/icon_lol.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Wed Mar 28, 2007 5:11 pm Post subject: |
|
|
nativemad wrote: | Can you mount the squashfs in your (normal booted) system (or chroot)?
mount -t squashfs -o loop /path/to/livecd.squashfs /somewhere/else |
No because the kernel I'm currently running (not the one on the CD) doesn't have the squashfs support option enabled. Is this part of the problem?
Quote: | Anything (almost a bit) in dependence to access the CDROM should be IN The Kernel, not as Module.... Perhaps you have to mount the squash from the CD and have a look with lsmod which modules have been loaded...
The root= and real_root= seems to be good... also make shure that the loop=/livecd.squashfs is in place!
If you get the tiny shell via init=/bin/sh, you should probably check that real_root=/dev/loop0 is correct! simply type losetup -a and if you see that you got more than one loop mounted, than this would probably mean that you have to do it like real_root=/dev/loop1 or something. | adding init/bin/sh to the kernel line makes no change. I tried real_root=/dev/loop1 and loop2. No difference
Quote: | Genkernel is very easy to understand and edit.... If you know a bit bash, simply edit /usr/share/genkernel/generic/initrd.scrits or /usr/share/genkernel/linuxrc according to your needs.... As example you could open /bin/sh when the cdrom gets detected, just before the squash should get mounted... | Not easy enough for me to completely understand. Would I put just before Code: | check_loop() {
if [ "${LOOP}" = '' -o ! -e "mnt/cdrom/${LOOP}" ]
then
bad_msg "Invalid loop location: ${LOOP}"
bad_msg 'Please export LOOP with a valid location, or reboot and pass a proper loop=...'
bad_msg 'kernel command line!'
run_shell
fi
} | ? (I'm tired of burning coasters )
Quote: | I was just very confused about ">> Mediafound on /dev/iseries/vcd* "! Is your cdrom at /dev/iseries/vcd*?!?!?!?!?!? -I thougt perhaps you got one of those very nice IBM-Mainframes because of iseries.... but amd64 is also cool! ![Laughing :lol:](images/smiles/icon_lol.gif) | My noobishness in this area really shows. I'm a LiveCD virgin. Anyway here is my /usr/share/genkernel/generic/initrd.scripts:
Code: | #!/bin/ash
. /etc/initrd.defaults
backup() {
echo -ne "\033[0G\033[0K"
}
strlen() {
if [ -z "$1" ]
then
echo "usage: strlen <variable_name>"
die
fi
eval echo "\${#${1}}"
}
parse_opt() {
case "$1" in
*\=*)
local key_name="`echo "$1" | cut -f1 -d=`"
local key_len=`strlen key_name`
local value_start=$((key_len+2))
echo "$1" | cut -c ${value_start}-
;;
esac
}
modules_load() {
for module in $*
do
echo ${module} >> /etc/modules/extra_load
done
modules_scan extra_load
}
modules_scan() {
local MODS
[ -d /etc/modules/${1} ] || touch /etc/modules/${1}
[ -f /etc/modules/${1} ] && MODS=`cat /etc/modules/${1}`
for x in ${MODS}
do
MLOAD=`echo ${MLIST} | sed -e "s/.*${x}.*/${x}/"`
if [ "${MLIST}" = "${x}" ] # Only module to no-load
then
echo -e "${BOLD} ::${NORMAL} Skipping ${x}..."
elif [ "${MLOAD}" = "${MLIST}" ] # == No change == No specified no-load
then
[ -n "${DEBUG}" ] && echo -ne "${BOLD} ::${NORMAL} Checking for ${x}..."
# find -name does not work since the return status is always zero
if find /lib/modules | grep /"${x}${KSUFF}" >/dev/null 2>&1
then
echo -ne "${BOLD} ::${NORMAL} Scanning for ${x}..."
modprobe ${x} -n
backup
echo -ne "${NORMAL}"
fi
else
echo -e "${BOLD} ::${NORMAL} Skipping ${x}..."
fi
done
}
findcdmount() {
if [ "$#" -gt "0" ]
then
for x in $*
do
# Check for a block device to mount
if [ -b "${x}" ]
then
good_msg "Attempting to mount media:- ${x}"
if [ -z "${CDROOT_DEV}" ]
then
mount -r -t auto ${x} ${NEW_ROOT}/mnt/cdrom \
> /dev/null 2>&1
else
mount -r -t iso9660 ${x} ${NEW_ROOT}/mnt/cdrom \
> /dev/null 2>&1
fi
if [ "$?" = '0' ]
then
# Check for a LiveCD
if [ -e ${NEW_ROOT}/mnt/cdrom/${SUBDIR}/livecd ]
then
REAL_ROOT="${x}"
break
else
umount ${NEW_ROOT}/mnt/cdrom
fi
fi
fi
done
if [ "${REAL_ROOT}" != '' ]
then
good_msg "Media found on ${x}"
fi
fi
}
cache_cd_contents() {
# Check loop file exists and cache to ramdisk if DO_cache is enabled
if [ "${LOOPTYPE}" != "noloop" ] && [ "${LOOPTYPE}" != "sgimips" ]
then
check_loop
if [ "${DO_cache}" ]
then
# TODO: Check the size of the image versus the size of our tmpfs
# along with the amount of available RAM and increase tmpfs size
# if necessary. (Not having awk sucks...)
# z=0
# for i in $(cat /proc/meminfo | grep -e ^MemFree -e ^Cached | \
# cut -d: -f2 | cut -dk -f1 | sed -e "s/^\s*//") ; do
# z=$(($z + $i)) ; done
# echo $z
good_msg "Copying loop file for caching..."
cp -a ${NEW_ROOT}/mnt/cdrom/${LOOP} ${NEW_ROOT}/mnt/${LOOP}
if [ $? -ne 0 ]
then
bad_msg "Failed to cache the loop file! Lack of space?"
rm -rf ${NEW_ROOT}/mnt/livecd.* 2>/dev/null
rm -rf ${NEW_ROOT}/mnt/zisofs 2>/dev/null
else
LOOPEXT='../'
fi
fi
fi
}
mount_sysfs() {
if [ "${KV_2_6_OR_GREATER}" ]
then
# Udev is semi-broken on non /sys sysfs mount points.
mount -t sysfs /sys /sys >/dev/null 2>&1
ret=$?
# sysfs mount failed .. udev wont work fall back to devfs if available
[ "$ret" -eq '0' ] || USE_UDEV_NORMAL=0
fi
}
# Insert a directory tree $2 to an union specified by $1
# Top-level read-write branch is specified by it's index 0
# $1 = union absolute path (starting with /)
# $2 = path to data directory
#
union_insert_dir() {
# detect branch 0 because newer unionfs doesn't support "0,1,2"
BRANCH_0="/"$(/sbin/unionctl $1 --list | cut -d/ -f2- | cut -d'(' -f1)
/sbin/unionctl $1 --add --after $BRANCH_0 --mode ro $2
if [ $? = '0' ]
then
good_msg "Addition of $2 to $1 after branch $BRANCH_0 successful"
fi
}
findnfsmount() {
if [ "${IP}" != '' ] || busybox udhcpc -R rootpath -n -s /bin/udhcpc.scripts
then
[ -e /rootpath ] && NFSROOT=`cat /rootpath`
if [ "${NFSROOT}" = '' ]
then
# Obtain NFSIP
OPTIONS=`busybox dmesg | grep rootserver | sed -e "s/,/ /g"`
for OPTION in $OPTIONS
do
if [ `echo $OPTION | sed -e "s/=/ /g" | cut -d " " -f 1` = 'rootserver' ]
then
NFSIP=`echo $OPTION | sed -e "s/=/ /g" | cut -d " " -f 2`
fi
done
# Obtain NFSPATH
OPTIONS=`busybox dmesg | grep rootpath | sed -e "s/,/ /g"`
for OPTION in $OPTIONS
do
if [ `echo $OPTION | sed -e "s/=/ /g" | cut -d " " -f 1` = 'rootpath' ]
then
NFSPATH=`echo $OPTION | sed -e "s/=/ /g" | cut -d " " -f 2`
fi
done
# Setup NFSROOT
if [ "${NFSIP}" != '' ] && [ "$NFSPATH" != '' ]
then
NFSROOT="${NFSIP}:${NFSPATH}"
else
bad_msg "The DHCP Server did not send a valid root-path."
bad_msg "Please check your DHCP setup, or provide a nfsroot=<...> parameter."
fi
fi
if [ "${NFSROOT}" != '' ]
then
if [ "${CDROOT}" != '0' ]
then
good_msg "Attempting to mount NFS CD image on ${NFSROOT}"
mount -t nfs -o ro,nolock,rsize=1024,wsize=1024 ${NFSROOT} ${NEW_ROOT}/mnt/cdrom
if [ "$?" = '0' ]
then
REAL_ROOT="/dev/nfs"
else
bad_msg "NFS Mounting failed. Is the path corrent ?"
fi
else
good_msg "Attemping to mount NFS root on ${NFSROOT}"
mount -t nfs -o ro,nolock,rsize=1024,wsize=1024 ${NFSROOT} ${NEW_ROOT}
if [ "$?" = '0' ]
then
REAL_ROOT="/dev/nfs"
else
bad_msg "NFS Mounting failed. Is the path correct ?"
fi
# FIXME: Need to start portmap and the other rpc daemons in order to
# FIXME: remount rw.
fi
fi
fi
}
kill_devfsd() {
killall devfsd > /dev/null 2>&1
}
check_loop() {
if [ "${LOOP}" = '' -o ! -e "mnt/cdrom/${LOOP}" ]
then
bad_msg "Invalid loop location: ${LOOP}"
bad_msg 'Please export LOOP with a valid location, or reboot and pass a proper loop=...'
bad_msg 'kernel command line!'
run_shell
fi
}
run_shell() {
/bin/ash
}
runUdev() {
mount -t tmpfs -o size=100k udev /dev
mkdir /dev/pts
mkdir /dev/shm
echo > /proc/sys/kernel/hotplug
/sbin/udevd --daemon
mkdir -p /dev/.udev/queue
/sbin/udevtrigger
good_msg 'Letting udev process events'
loop=0
while test -d /dev/.udev/queue; do
sleep 1;
test "$loop" -gt 30 && break
loop=$(($loop + 1))
done
good_msg "Udev finished proccsing after $loop iterations"
killall udevd > /dev/null 2>&1
ln -snf /proc/self/fd /dev/fd
ln -snf /proc/self/fd/0 /dev/stdin
ln -snf /proc/self/fd/1 /dev/stdout
ln -snf /proc/self/fd/2 /dev/stderr
ln -snf /proc/kcore /dev/core
}
runmdev() {
# busybox udev replacement
mdev -s
# catch udev hotplug events
echo /sbin/mdev > /proc/sys/kernel/hotplug
}
test_success() {
error_string=$1
error_string="${error_string:-run command}"
# If last command failed send error message and fall back to a shell
if [ "$?" != '0' ]
then
splash 'verbose'
bad_msg 'Failed to $1; failing back to the shell...'
run_shell
fi
}
good_msg() {
msg_string=$1
msg_string="${msg_string:-...}"
echo -e "${GOOD}>>${NORMAL}${BOLD} ${msg_string} ${NORMAL}"
}
bad_msg() {
msg_string=$1
msg_string="${msg_string:-...}"
echo -e "${BAD}!!${NORMAL}${BOLD} ${msg_string} ${NORMAL}"
}
warn_msg() {
msg_string=$1
msg_string="${msg_string:-...}"
echo -e "${WARN}**${NORMAL}${BOLD} ${msg_string} ${NORMAL}"
}
bind_mount_dev() {
# bind-mount /dev/ so that loop devices can be found
mount -o bind ${NEW_ROOT}/dev /dev
}
start_dev_mgr() {
# Check udev is available...
if [ "${KV_2_6_OR_GREATER}" -a ! "${USE_UDEV_NORMAL}" -eq '0' ]
then
USE_UDEV_NORMAL=1
else
USE_UDEV_NORMAL=0
fi
if [ "${USE_UDEV_NORMAL}" -eq '1' ]
then
for dir in /sys/bus/usb/drivers/usb-storage/*
do
if [ -d "${dir}" ]
then
FORCE_slowusb="1"
fi
done
if [ "${DO_slowusb}" ] || [ "${FORCE_slowusb}" ]
then
sleep 10
fi
kill_devfsd
good_msg 'Activating mdev'
runmdev
if [ "${DO_slowusb}" ] || [ "${FORCE_slowusb}" ]
then
sleep 10
fi
else
good_msg 'Activating devfs'
mount -t devfs devfs /dev 2>/dev/null
devfsd /dev -np
fi
}
bootstrapCD() {
# Locate the cdrom device with our media on it.
# CDROM DEVICES
DEVICES="/dev/cdroms/* /dev/ide/cd/* /dev/sr*"
# USB Keychain/Storage
DEVICES="$DEVICES /dev/sd*"
# IDE devices
DEVICES="$DEVICES /dev/hd*"
# USB using the USB Block Driver
DEVICES="$DEVICES /dev/ubd* /dev/ubd/*"
# iSeries devices
DEVICES="$DEVICES /dev/iseries/vcd*"
# The device was specified on the command line. Shold we even be doing a
# scan at this point? I think not.
[ -n "${CDROOT_DEV}" ] && DEVICES="$DEVICES ${CDROOT_DEV}"
findcdmount $DEVICES
}
cmdline_hwopts() {
# Scan CMDLINE for any "doscsi" or "noscsi"-type arguments
local FOUND
local TMP_HWOPTS
for x in $HWOPTS
do
for y in $CMDLINE
do
if [ "${y}" = "do${x}" ]
then
MY_HWOPTS="${MY_HWOPTS} $x"
elif [ "${y}" = "no${x}" ]
then
MY_HWOPTS="`echo ${MY_HWOPTS} | sed -e \"s/${x}//g\" -`"
fi
done
done
# Shouldnt need to sort this as the following loop should figure out the
# duplicates and strip them out
#MY_HWOPTS=`echo ${MY_HWOPTS}| sort`
for x in ${MY_HWOPTS}
do
FOUND=0
for y in ${TMP_HWOPTS}
do
if [ "${y}" = "${x}" ]
then
FOUND=1
fi
done
if [ ! "${FOUND}" = '1' ]
then
TMP_HWOPTS="${TMP_HWOPTS} ${x}"
fi
done
MY_HWOPTS=${TMP_HWOPTS}
}
load_modules() {
# Load modules listed in MY_HWOPTS if /lib/modules exists
if [ -d '/lib/modules' ]
then
good_msg 'Loading modules'
# Load appropriate kernel modules
for modules in $MY_HWOPTS
do
modules_scan $modules
eval DO_`echo $modules | sed 's/-//'`=1
done
else
good_msg 'Skipping module load; no modules in the initrd!'
fi
}
detect_sbp2_devices() {
# http://www.linux1394.org/sbp2.php
# /proc
# /proc/scsi/sbp2/0, /proc/scsi/sbp2/1, etc.
#
# You may manually add/remove SBP-2 devices via the procfs with:
# add-single-device <h> <b> <t> <l> or remove-single-device <h> <b> <t> <l>,
# where:
#
# <h> = host (starting at zero for first SCSI adapter)
# <b> = bus (normally zero)
# <t> = target (starting at zero for first SBP-2 device)
# <l> - lun (normally zero)
# e.g. To manually add/detect a new SBP-2 device
# echo "scsi add-single-device 0 0 0 0" > /proc/scsi/scsi
# e.g. To manually remove a SBP-2 device after it's been unplugged
# echo "scsi remove-single-device 0 0 0 0" > /proc/scsi/scsi
# e.g. To check to see which SBP-2/SCSI devices are currently registered
# cat /proc/scsi/scsi
[ -e /proc/scsi/scsi ] && echo 'scsi add-single-device 0 0 0 0' > /proc/scsi/scsi
}
setup_keymap() {
if [ "${DO_keymap}" ]
then
if [ ! -e /dev/vc/0 -a ! -e /dev/tty0 ]
then
DEVBIND=1
mount -o bind ${NEW_ROOT}/dev /dev
fi
[ ! -e /dev/tty0 ] && ln -s /dev/tty1 /dev/tty0
chooseKeymap
[ "${DEVBIND}" -eq '1' ] && umount /dev
if [ -e /etc/sysconfig/keyboard -a "${CDROOT}" -eq '1' ]
then
mkdir -p ${NEW_ROOT}/etc/sysconfig/
cp /etc/sysconfig/keyboard ${NEW_ROOT}/etc/sysconfig/keyboard
fi
fi
}
chooseKeymap() {
good_msg "Loading keymaps"
cat /lib/keymaps/keymapList
read -t 10 -p '<< Load keymap (Enter for default): ' keymap
case ${keymap} in
1|azerty) keymap=azerty ;;
2|be) keymap=be ;;
3|bg) keymap=bg ;;
4|br-a) keymap=br-a ;;
5|br-l) keymap=br-l ;;
6|by) keymap=by ;;
7|cf) keymap=cf ;;
8|croat) keymap=croat ;;
9|cz) keymap=cz ;;
10|de) keymap=de ;;
11|dk) keymap=dk ;;
12|dvorak) keymap=dvorak ;;
13|es) keymap=es ;;
14|et) keymap=et ;;
15|fi) keymap=fi ;;
16|fr) keymap=fr ;;
17|gr) keymap=gr ;;
18|hu) keymap=hu ;;
19|il) keymap=il ;;
20|is) keymap=is ;;
21|it) keymap=it ;;
22|jp) keymap=jp ;;
23|la) keymap=la ;;
24|lt) keymap=lt ;;
25|mk) keymap=mk ;;
26|nl) keymap=nl ;;
27|no) keymap=no ;;
28|pl) keymap=pl ;;
29|pt) keymap=pt ;;
30|ro) keymap=ro ;;
31|ru) keymap=ru ;;
32|se) keymap=se ;;
33|sg) keymap=sg ;;
34|sk-y) keymap=sk-y ;;
35|sk-z) keymap=sk-z ;;
36|slovene) keymap=slovene ;;
37|trf) keymap=trf ;;
38|trq) keymap=trq ;;
39|ua) keymap=ua ;;
40|uk) keymap=uk ;;
41|us) keymap=us ;;
42|wangbe) keymap=wangbe ;;
esac
if [ -e /lib/keymaps/${keymap}.map ]
then
good_msg "Loading the ''${keymap}'' keymap"
loadkmap < /lib/keymaps/${keymap}.map
# xkeymap=${keymap}
# echo ${keymap} | egrep -e "[0-9]+" >/dev/null 2>&1
# if [ "$?" -eq '0' ]
# then
# xkeymap=`tail -n 7 /lib/keymaps/keymapList | grep ${keymap} | sed -r "s/.*\s+${keymap}\s+([a-z-]+).*/\1/g" | egrep -v 1`
# fi
mkdir -p /etc/sysconfig
# echo "XKEYBOARD=${xkeymap}" > /etc/sysconfig/keyboard
echo "XKEYBOARD=${keymap}" > /etc/sysconfig/keyboard
elif [ "$keymap" = '' ]
then
echo
good_msg "Keeping default keymap"
else
bad_msg "Sorry, but keymap ''${keymap}'' is invalid!"
chooseKeymap
fi
}
startVolumes() {
#good_msg 'Checking if volumes need to be started...'
# Here, we check for /dev/device-mapper, and if it exists, we setup a
# a symlink, which should hopefully fix bug #142775 and bug #147015
if [ -e /dev/device-mapper ] && [ ! -e /dev/mapper/control ]
then
mkdir -p /dev/mapper
ln -sf /dev/device-mapper /dev/mapper/control
fi
if [ "${USE_DMRAID_NORMAL}" -eq '1' ]
then
if [ -e '/sbin/dmraid' ]
then
good_msg "Activating Device-Mapper RAID(s)"
if [ '${DMRAID_OPTS}' = '' ]
then
/sbin/dmraid -ay
else
/sbin/dmraid -ay ${DMRAID_OPTS}
fi
fi
fi
if [ "${USE_LVM2_NORMAL}" -eq '1' ]
then
if [ -e '/bin/vgscan' -a -e '/bin/vgchange' ]
then
for dev in ${RAID_DEVICES}
do
setup_md_device "${dev}"
done
good_msg "Scanning for Volume Groups"
/bin/vgscan --ignorelockingfailure --mknodes 2>/dev/null
good_msg "Activating Volume Groups"
/bin/vgchange -ay --ignorelockingfailure 2>/dev/null
# Disable EVMS since lvm2 is activated and they dont work together.
if [ "${USE_EVMS2_NORMAL}" -eq '1' ]
then
bad_msg "Disabling EVMS Support because LVM2 started"
bad_msg "Do not add dolvm2 to the cmdline if this is not what you want"
bad_msg "LVM2 and EVMS do not work well together"
USE_EVMS2_NORMAL=0
fi
else
bad_msg "vgscan or vgchange not found: skipping LVM2 volume group activation!"
fi
fi
if [ "${USE_EVMS2_NORMAL}" -eq '1' ]
then
if [ -e '/sbin/evms_activate' ]
then
good_msg "Activating EVMS"
evms_activate
fi
fi
}
# Open a LUKS device
# $1 LUKS device
# $2 LUKS name
openLUKS() {
LUKS_DEVICE="$1"
LUKS_NAME="$2"
if [ -e /sbin/cryptsetup ]
then
while [ 1 ]
do
if [ "${LUKS_DEVICE}" = '' ]
then
# LUKS device could not be opened. Prompt user for device.
bad_msg "The LUKS ${LUKS_NAME} block device is not detected."
echo " Please specify a ${LUKS_NAME} LUKS device to open, "q" to skip, or "shell" for a shell..."
echo -n "LUKS ${LUKS_NAME}() :: "
read LUKS_DEVICE
continue
elif [ "${LUKS_DEVICE}" = 'shell' ]
then
run_shell
LUKS_DEVICE=''
continue
elif [ "${LUKS_DEVICE}" = 'q' ]
then
break
else
if cryptsetup isLuks ${LUKS_DEVICE}
then
good_msg "Opening LUKS device ${LUKS_DEVICE}"
cryptsetup luksOpen ${LUKS_DEVICE} ${LUKS_NAME}
if [ ! "$?" -eq '0' ]
then
bad_msg "Failed open LUKS device ${LUKS_DEVICE}"
else
break
fi
else
bad_msg "The LUKS device ${LUKS_DEVICE} does not contain a LUKS header"
fi
fi
LUKS_DEVICE=''
done
else
bad_msg "The initrd does not support LUKS"
fi
}
startLUKS() {
if [ -n "${LUKS_ROOT}" ]
then
openLUKS "${LUKS_ROOT}" "root"
if [ -n "${REAL_ROOT}" ]
then
# Rescan volumes
startVolumes
else
REAL_ROOT="/dev/mapper/root"
fi
fi
if [ -n "${LUKS_SWAP}" ]
then
openLUKS "${LUKS_SWAP}" "swap"
break
fi
}
sdelay() {
# Sleep a specific number of seconds if SDELAY is set otherwise only sleep
# 1 second
if [ -n "${SDELAY}" ]
then
sleep ${SDELAY}
else
sleep 1
fi
}
quiet_kmsg() {
# if QUIET is set make the kernel less chatty
[ -n "$QUIET" ] && echo '0' > /proc/sys/kernel/printk
}
verbose_kmsg() {
# if QUIET is set make the kernel less chatty
[ -n "$QUIET" ] && echo '6' > /proc/sys/kernel/printk
}
cdupdate() {
if [ "${CDROOT}" -eq '1' ]
then
if [ -x /${NEW_ROOT}/mnt/cdrom/cdupdate.sh ]
then
good_msg "Running cdupdate.sh"
${NEW_ROOT}/mnt/cdrom/cdupdate.sh
if [ "$?" != '0' ]
then
splash 'verbose'
bad_msg "Executing cdupdate.sh failed!"
run_shell
fi
else
good_msg 'No cdupdate.sh script found, skipping...'
fi
fi
}
setup_md_device() {
local device
[ -z "$1" ] && device="${REAL_ROOT}" || device="$1"
[ -z "${device}" ] && return # LiveCD
if [ `echo ${device}|sed -e 's#\(luks:\)\?\(/dev/md\)[[:digit:]]\+#\2#'` = "/dev/md" ]
then
good_msg 'Detected real_root as a md device. Setting up the device node...'
MD_NUMBER=`echo ${device}|sed -e 's#\(luks:\)\?/dev/md\([[:digit:]]\+\)#\2#'`
if [ ! -e /dev/md${MD_NUMBER} ]
then
mknod /dev/md${MD_NUMBER} b 9 ${MD_NUMBER} >/dev/null 2>&1
[ "$?" -ne 0 ] && bad_msg "Creation of /dev/md${MD_NUMBER} failed..."
fi
mdstart ${MDPART} /dev/md${MD_NUMBER}
fi
}
rundebugshell() {
if [ -n "$DEBUG" ]
then
good_msg 'Starting debug shell as requested by "debug" option.'
good_msg 'Type "exit" to continue with normal bootup.'
[ -x /bin/sh ] && /bin/sh || /bin/ash
fi
}
setup_unionfs() {
if [ "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
# Directory used for rw changes in union mount filesystem
UNION=/union
MEMORY=/memory
if [ -z "$UID" ]
then
CHANGES=$MEMORY/unionfs_changes/default
else
CHANGES=$MEMORY/unionfs_changes/$UID
fi
mkdir -p ${MEMORY}
mkdir -p ${UNION}
good_msg "Loading unionfs module"
modprobe unionfs > /dev/null 2>&1
if [ -n "${UNIONFS}" ]
then
CHANGESDEV=${UNIONFS}
good_msg "mounting $CHANGESDEV to $MEMORY for unionfs support"
mount -t auto $CHANGESDEV $MEMORY
# mount tmpfs only in the case when changes= boot parameter was
# empty or we were not able to mount the storage device
ret=$?
if [ "${ret}" -ne 0 ]
then
bad_msg "mount of $CHANGESDEV failed falling back to ramdisk based unionfs"
mount -t tmpfs tmpfs $MEMORY
fi
if [ "${CDROOT}" -eq '1' -a ! -f ${MEMORY}/livecd.unionfs ]
then
umount $MEMORY
bad_msg "failed to find livecd.unionfs file on $CHANGESDEV"
bad_msg "create a livecd.unionfs file on this device if you wish to use it for unionfs"
bad_msg "falling back to ramdisk based unionfs for safety"
mount -t tmpfs tmpfs $MEMORY
fi
else
good_msg "Mounting ramdisk to $MEMORY for unionfs support..."
mount -t tmpfs tmpfs $MEMORY
fi
mkdir -p $CHANGES
mount -t unionfs -o dirs=$CHANGES=rw unionfs ${UNION}
ret=$?
if [ "${ret}" -ne 0 ]
then
bad_msg "Can't setup union ${UNION} in directory!"
USE_UNIONFS_NORMAL=0
fi
else
USE_UNIONFS_NORMAL=0
fi
}
suspend_resume() {
[ -x /sbin/resume ] || return 0
/sbin/resume
local ret=$?
if [ "${ret}" -eq 0 ]; then
exit 0
fi
return 0
}
suspend2_resume() {
if [ -d /proc/suspend2 ] || [ -d /sys/power/suspend2 ]; then
local splash_theme
if grep "splash=" /proc/cmdline > /dev/null 2>&1; then
splash_theme=`cat /proc/cmdline | sed 's/.*splash=/splash=/' | sed 's/ .*//' | sed 's/.*theme://' | sed 's/,.*//'`
fi
local suspend2_userui_program="/sys/power/suspend2/user_interface/program"
local suspend2_do_resume="/sys/power/suspend2/do_resume"
#
# Backward compatibility
#
if [ -e /proc/suspend2 ]; then
suspend2_userui_program="/proc/suspend2/userui_program"
suspend2_do_resume="/proc/suspend2/do_resume"
fi
modules_scan suspend2
if which suspend2ui_text > /dev/null 2>&1; then
which suspend2ui_text > "${suspend2_userui_program}"
fi
if [ -n "${splash_theme}" ]; then
ln -s /etc/splash/${splash_theme} /etc/splash/suspend2
if which suspend2ui_fbsplash > /dev/null 2>&1; then
which suspend2ui_fbsplash > "${suspend2_userui_program}"
fi
fi
echo > "${suspend2_do_resume}"
fi
}
|
And /usr/share/genkernel/generic/linuxrc: Code: | #!/bin/sh
# Copyright 2003-2006 Gentoo Foundation
# Distributed under the terms of the GNU General Public License v2
. /etc/initrd.defaults
. /etc/initrd.scripts
splash() {
return 0
}
[ -e /etc/initrd.splash ] && . /etc/initrd.splash
# Clean input/output
exec >${CONSOLE} <${CONSOLE} 2>&1
if [ "$$" != '1' ]
then
echo '/linuxrc has to be run as the init process as the one'
echo 'with a PID of 1. Try adding init="/linuxrc" to the'
echo 'kernel command line or running "exec /linuxrc".'
exit 1
fi
mount -t proc proc /proc >/dev/null 2>&1
mount -o remount,rw / >/dev/null 2>&1
# Set up symlinks
if [ "$0" = '/init' ]
then
/bin/busybox --install -s
[ -e /linuxrc ] && rm /linuxrc
if [ -e /bin/lvm ]
then
ln -s /bin/lvm /bin/vgscan
ln -s /bin/lvm /bin/vgchange
fi
fi
quiet_kmsg
CMDLINE="`cat /proc/cmdline`"
# Scan CMDLINE for any specified real_root= or cdroot arguments
REAL_ROOT=''
FAKE_ROOT=''
REAL_ROOTFLAGS=''
for x in ${CMDLINE}
do
case "${x}" in
real_root\=*)
REAL_ROOT=`parse_opt "${x}"`
;;
root\=*)
FAKE_ROOT=`parse_opt "${x}"`
;;
subdir\=*)
SUBDIR=`parse_opt "${x}"`
;;
real_init\=*)
REAL_INIT=`parse_opt "${x}"`
;;
init_opts\=*)
INIT_OPTS=`parse_opt "${x}"`
;;
# Livecd options
cdroot)
CDROOT=1
;;
cdroot\=*)
CDROOT=1
CDROOT_DEV=`parse_opt "${x}"`
;;
# Start livecd loop and looptype options
loop\=*)
LOOP=`parse_opt "${x}"`
;;
looptype\=*)
LOOPTYPE=`parse_opt "${x}"`
;;
# Start Device Manager options
devfs)
USE_DEVFS_NORMAL=1
USE_UDEV_NORMAL=0
;;
udev)
USE_DEVFS_NORMAL=0
USE_UDEV_NORMAL=1
;;
unionfs)
if [ ! -x /sbin/unionctl ]
then
USE_UNIONFS_NORMAL=0
bad_msg 'Unionctl not found: aborting use of unionfs!'
else
USE_UNIONFS_NORMAL=1
fi
;;
unionfs\=*)
if [ ! -x /sbin/unionctl ]
then
USE_UNIONFS_NORMAL=0
bad_msg 'Unionctl not found: aborting use of unionfs!'
else
USE_UNIONFS_NORMAL=1
CMD_UNIONFS=`parse_opt "${x}"`
echo ${CMD_UNIONFS}|grep , >/dev/null 2>&1
if [ "$?" -eq '0' ]
then
UID=`echo ${CMD_UNIONFS#*,}`
UNIONFS=`echo ${CMD_UNIONFS%,*}`
else
UNIONFS=${CMD_UNIONFS}
fi
fi
;;
# Start Volume manager options
dolvm2)
USE_LVM2_NORMAL=1
;;
dodmraid)
USE_DMRAID_NORMAL=1
;;
dodmraid\=*)
DMRAID_OPTS=`parse_opt "${x}"`
USE_DMRAID_NORMAL=1
;;
doevms2)
USE_EVMS2_NORMAL=1
;;
# Debug Options
debug)
DEBUG='yes'
;;
# Scan delay options
scandelay\=*)
SDELAY=`parse_opt "${x}"`
;;
scandelay)
SDELAY=10
;;
# Module no-loads
doload\=*)
MDOLIST=`parse_opt "${x}"`
MDOLIST="`echo ${MDOLIST} | sed -e \"s/,/ /g\"`"
;;
nodetect)
NODETECT=1
;;
noload\=*)
MLIST=`parse_opt "${x}"`
MLIST="`echo ${MLIST} | sed -e \"s/,/ /g\"`"
export MLIST
;;
# Redirect output to a specific tty
CONSOLE\=*)
CONSOLE=`parse_opt "${x}"`
exec >${CONSOLE} <${CONSOLE} 2>&1
;;
# /dev/md
lvmraid\=*)
RAID_DEVICES="`parse_opt ${x}`"
RAID_DEVICES="`echo ${RAID_DEVICES} | sed -e 's/,/ /g'`"
;;
part\=*)
MDPART=`parse_opt "${x}"`
;;
# NFS
ip\=*)
IP=`parse_opt "${x}"`
;;
nfsroot\=*)
NFSROOT=`parse_opt "${x}"`
;;
crypt_root\=*)
LUKS_ROOT=`parse_opt "${x}"`
;;
crypt_swap\=*)
LUKS_SWAP=`parse_opt "${x}"`
;;
real_rootflags\=*)
REAL_ROOTFLAGS=`parse_opt "${x}"`
;;
esac
done
if [ -z "${REAL_ROOT}" -a \( "${CDROOT}" -eq 0 \) -a \( "${FAKE_ROOT}" != "/dev/ram0" \) ]
then
REAL_ROOT="${FAKE_ROOT}"
fi
splash 'init'
detect_sbp2_devices
cmdline_hwopts
# Load modules listed in MY_HWOPTS if /lib/modules exists
if [ -d '/lib/modules' ]
then
good_msg 'Loading modules'
# Load appropriate kernel modules
if [ "${NODETECT}" -ne '1' ]
then
for modules in $MY_HWOPTS
do
modules_scan $modules
eval DO_`echo $modules | sed 's/-//'`=1
done
fi
# Always eval doload=...
modules_load $MDOLIST
else
for modules in $MY_HWOPTS
do
eval DO_`echo $modules | sed 's/-//'`=1
done
good_msg 'Skipping module load; no modules in the initrd!'
fi
# Mount sysfs
mount_sysfs
# Delay if needed for USB hardware
sdelay
# Start udev/devfs
start_dev_mgr
# Setup md device nodes if they dont exist
setup_md_device
# Scan volumes
startVolumes
# Initialize LUKS root device
startLUKS
# Set up unionfs
mkdir -p ${NEW_ROOT}
setup_unionfs
if [ "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
CHROOT=${UNION}
else
CHROOT=${NEW_ROOT}
fi
# Run debug shell if requested
rundebugshell
suspend_resume
suspend2_resume
if [ "${CDROOT}" -eq '1' ]
then
if [ ! "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
good_msg "Making tmpfs for ${NEW_ROOT}"
mount -t tmpfs tmpfs ${NEW_ROOT}
fi
for i in dev mnt mnt/cdrom mnt/livecd tmp tmp/.initrd mnt/gentoo sys
do
mkdir -p ${NEW_ROOT}/$i
chmod 755 ${NEW_ROOT}/$i
done
[ ! -e ${NEW_ROOT}/dev/null ] && mknod ${NEW_ROOT}/dev/null c 1 3
[ ! -e ${NEW_ROOT}/dev/console ] && mknod ${NEW_ROOT}/dev/console c 5 1
# For SGI LiveCDs ...
if [ "${LOOPTYPE}" = "sgimips" ]
then
[ ! -e ${NEW_ROOT}/dev/sr0 ] && mknod ${NEW_ROOT}/dev/sr0 b 11 0
[ ! -e ${NEW_ROOT}/dev/loop0 ] && mknod ${NEW_ROOT}/dev/loop0 b 7 0
fi
# Required for gensplash to work. Not an issue with the initrd as this
# device isnt created there and is not needed.
if [ -e /dev/tty1 ]
then
[ ! -e ${NEW_ROOT}/dev/tty1 ] && mknod ${NEW_ROOT}/dev/tty1 c 4 1
fi
if [ "${REAL_ROOT}" != "/dev/nfs" ] && [ "${LOOPTYPE}" != "sgimips" ]
then
bootstrapCD
fi
if [ "${REAL_ROOT}" = '' ]
then
echo -n -e "${WARN}>>${NORMAL}${BOLD} No bootable medium found. Waiting for new devices"
COUNTER=0
while [ $COUNTER -lt 3 ]; do
sleep 3
echo -n '.'
let COUNTER=$COUNTER+1
done
sleep 1
echo -e "${NORMAL}"
bootstrapCD
fi
if [ "${REAL_ROOT}" = '' ]
then
# Undo stuff
umount ${NEW_ROOT}/dev 2>/dev/null
umount ${NEW_ROOT}/sys 2>/dev/null
umount /sys 2>/dev/null
umount ${NEW_ROOT}
rm -rf ${NEW_ROOT}/*
bad_msg 'Could not find CD to boot, something else needed!'
CDROOT=0
fi
fi
setup_keymap
# Determine root device
good_msg 'Determining root device...'
while true
do
while [ "${got_good_root}" != '1' ]
do
case "${REAL_ROOT}" in
LABEL\=*|UUID\=*)
ROOT_DEV=""
retval=1
if [ "${retval}" -ne '0' ]; then
ROOT_DEV=`findfs "${REAL_ROOT}" 2>/dev/null`
retval=$?
fi
if [ "$retval" -ne '0' ]; then
ROOT_DEV=`busybox findfs "${REAL_ROOT}" 2>/dev/null`
retval=$?
fi
if [ "${retval}" -ne '0' ]; then
ROOT_DEV=`blkid -t "${REAL_ROOT}" | cut -d ":" -f 1 2>/dev/null`
retval=$?
fi
if [ "${retval}" -eq '0' ] && [ -n "${ROOT_DEV}" ]; then
good_msg "Detected real_root=${ROOT_DEV}"
REAL_ROOT="${ROOT_DEV}"
else
bad_msg "Could not find root block device: ${REAL_ROOT}"
echo ' Please specify a device to boot, or "shell" for a shell...'
echo -n 'boot() :: '
read REAL_ROOT
got_good_root=0
continue
fi
;;
esac
if [ "${REAL_ROOT}" = 'shell' ]
then
run_shell
REAL_ROOT=''
got_good_root=0
continue
elif [ "${REAL_ROOT}" = '' ]
then
# No REAL_ROOT determined/specified. Prompt user for root block device.
bad_msg "The root block device is unspecified or not detected."
echo ' Please specify a device to boot, or "shell" for a shell...'
echo -n 'boot() :: '
read REAL_ROOT
got_good_root=0
# Check for a block device or /dev/nfs
elif [ -b "${REAL_ROOT}" ] || [ "${REAL_ROOT}" = "/dev/nfs" ]
then
got_good_root=1
else
bad_msg "Block device ${REAL_ROOT} is not a valid root device..."
REAL_ROOT=""
got_good_root=0
fi
done
if [ "${CDROOT}" -eq '1' -a "${got_good_root}" = '1' -a "${REAL_ROOT}" != "/dev/nfs" ]
then
# CD already mounted; no further checks necessary
break
elif [ "${LOOPTYPE}" = "sgimips" ]
then
# sgimips mounts the livecd root partition directly
# there is no isofs filesystem to worry about
break
else
good_msg "Mounting root..."
# Try to mount the device as ${NEW_ROOT}
if [ "${REAL_ROOT}" = '/dev/nfs' ]; then
findnfsmount
else
# mount ro so fsck doesn't barf later
# REAL_ROOT_TYPE=`vol_id -t ${REAL_ROOT}`
# mount -t ${REAL_ROOT_TYPE} -o ro ${REAL_ROOT} ${NEW_ROOT}
if [ "${REAL_ROOTFLAGS}" = '' ]; then
mount -o ro ${REAL_ROOT} ${NEW_ROOT}
else
good_msg "Using mount -o ro,${REAL_ROOTFLAGS}"
mount -o ro,${REAL_ROOTFLAGS} ${REAL_ROOT} ${NEW_ROOT}
fi
fi
# If mount is successful break out of the loop
# else not a good root and start over.
if [ "$?" = '0' ]
then
break
else
bad_msg "Could not mount specified ROOT, try again"
got_good_root=0
REAL_ROOT=''
fi
fi
done
# End determine root device
#verbose_kmsg
# If cdroot is set determine the looptype to boot
if [ "${CDROOT}" = '1' ]
then
good_msg 'Determining looptype ...'
cd ${NEW_ROOT}
# Find loop and looptype if we have NFS-mounted a livecd
if [ "${LOOP}" = '' ] && [ "${REAL_ROOT}" = '/dev/nfs' ]
then
if [ -e "${NEW_ROOT}/mnt/cdrom/livecd.loop" ]; then
LOOP='/livecd.loop'
LOOPTYPE='normal'
elif [ -e "${NEW_ROOT}/mnt/cdrom/zisofs" ]; then
LOOP='/zisofs'
LOOPTYPE='zisofs'
elif [ -e "${NEW_ROOT}/mnt/cdrom/livecd.squashfs" ]; then
LOOP='/livecd.squashfs'
LOOPTYPE='squashfs'
elif [ -e "${NEW_ROOT}/mnt/cdrom/image.squashfs" ]; then
LOOP='/image.squashfs'
LOOPTYPE='squashfs'
elif [ -e "${NEW_ROOT}/mnt/cdrom/livecd.gcloop" ]; then
LOOP='/livecd.gcloop'
LOOPTYPE='gcloop'
else
LOOPTYPE='noloop'
fi
fi
# Failsafe the loop type wasn't set
if [ "${LOOPTYPE}" = '' ]
then
warn_msg 'Warning: loop type unspecified!'
if [ "${LOOP}" = '/livecd.loop' ]
then
LOOPTYPE='normal'
elif [ "${LOOP}" = '/zisofs' ]
then
LOOPTYPE='zisofs'
elif [ "${LOOP}" = '/livecd.squashfs' ]
then
LOOPTYPE='squashfs'
elif [ "${LOOP}" = '/image.squashfs' ]
then
LOOPTYPE='squashfs'
elif [ "${LOOP}" = '/livecd.gcloop' ]
then
LOOPTYPE='gcloop'
else
LOOPTYPE='noloop'
fi
fi
cache_cd_contents
# Setup the loopback mounts
if [ "${LOOPTYPE}" = 'normal' ]
then
good_msg 'Mounting loop filesystem'
mount -t ext2 -o loop,ro ${NEW_ROOT}/mnt/cdrom/${LOOPEXT}${LOOP} ${NEW_ROOT}/mnt/livecd
test_success 'Mount filesystem'
FS_LOCATION='mnt/livecd'
elif [ "${LOOPTYPE}" = 'squashfs' ]
then
good_msg 'Mounting squashfs filesystem'
mount -t squashfs -o loop,ro ${NEW_ROOT}/mnt/cdrom/${LOOPEXT}${LOOP} ${NEW_ROOT}/mnt/livecd
test_success 'Mount filesystem'
FS_LOCATION='mnt/livecd'
elif [ "${LOOPTYPE}" = 'gcloop' ]
then
good_msg 'Mounting gcloop filesystem'
echo ' ' | losetup -E 19 -e ucl-0 -p0 ${NEW_ROOT}/dev/loop0 ${NEW_ROOT}/mnt/cdrom/${LOOPEXT}${LOOP}
test_success 'losetup the loop device'
mount -t ext2 -o ro ${NEW_ROOT}/dev/loop0 ${NEW_ROOT}/mnt/livecd
test_success 'Mount the losetup loop device'
FS_LOCATION='mnt/livecd'
elif [ "${LOOPTYPE}" = 'zisofs' ]
then
FS_LOCATION="mnt/cdrom/${LOOPEXT}${LOOP}"
elif [ "${LOOPTYPE}" = 'noloop' ]
then
FS_LOCATION='mnt/cdrom'
elif [ "${LOOPTYPE}" = 'sgimips' ]
then
# getdvhoff finds the starting offset (in bytes) of the squashfs
# partition on the cdrom and returns this offset for losetup
#
# All currently supported SGI Systems use SCSI CD-ROMs, so
# so we know that the CD-ROM is usually going to be /dev/sr0.
#
# We use the value given to losetup to set /dev/loop0 to point
# to the liveCD root partition, and then mount /dev/loop0 as
# the LiveCD rootfs
good_msg 'Locating the SGI LiveCD Root Partition'
echo ' ' | \
losetup -o $(/bin/getdvhoff ${NEW_ROOT}${REAL_ROOT} 0) \
${NEW_ROOT}${CDROOT_DEV} \
${NEW_ROOT}${REAL_ROOT}
test_success 'losetup /dev/sr0 /dev/loop0'
good_msg 'Mounting the Root Partition'
mount -t squashfs -o ro ${NEW_ROOT}${CDROOT_DEV} ${NEW_ROOT}/mnt/livecd
test_success 'mount /dev/loop0 /'
FS_LOCATION='mnt/livecd'
fi
#
# End cdrom looptype determination and mounting if necessary
#
if [ "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
union_insert_dir ${UNION} ${NEW_ROOT}/${FS_LOCATION}
# Make sure fstab notes livecd is mounted ro. Makes system skip remount which fails on unionfs dirs.
sed -e 's|\(.*\s/\s*tmpfs\s*\)defaults\(.*\)|\1defaults,ro\2|' /${UNION}/etc/fstab > /${UNION}/etc/fstab.new
mv /${UNION}/etc/fstab.new /${UNION}/etc/fstab
fi
# Unpacking additional packages from NFS mount
# This is useful for adding kernel modules to /lib
# We do this now, so that additional packages can add whereever they want.
if [ "${REAL_ROOT}" = '/dev/nfs' ]
then
if [ -e "${CHROOT}/mnt/cdrom/add" ]
then
for targz in `ls ${CHROOT}/mnt/cdrom/add/*.tar.gz`
do
tarname=`basename ${targz}`
good_msg "Adding additional package ${tarname}"
(cd ${CHROOT} ; /bin/tar -xzf ${targz})
done
fi
fi
if [ ! "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
good_msg "Copying read-write image contents to tmpfs"
# Copy over stuff that should be writable
(cd ${NEW_ROOT}/${FS_LOCATION}; cp -a ${ROOT_TREES} ${NEW_ROOT})
# Now we do the links.
for x in ${ROOT_LINKS}
do
if [ -L "${NEW_ROOT}/${FS_LOCATION}/${x}" ]
then
ln -s "`readlink ${NEW_ROOT}/${FS_LOCATION}/${x}`" "${x}" 2>/dev/null
else
# List all subdirectories of x
for directory in `find "${NEW_ROOT}/${FS_LOCATION}/${x}" -type d 2>/dev/null`
do
## Strip the prefix of the FS_LOCATION
directory=${directory#${NEW_ROOT}/${FS_LOCATION}/}
## Skip this directory if we already linked a parent directory
if [ "${curent_parrent}" != '' ]; then
var=`echo "${directory}" | grep "^${curent_parrent}"`
if [ "${var}" != '' ]; then
continue
fi
fi
## Test if the directory exists already
if [ -e "/${NEW_ROOT}/${directory}" ]
then
# It does exist, link all the individual files
for file in `ls /${NEW_ROOT}/${FS_LOCATION}/${directory}`
do
if [ ! -d "/${NEW_ROOT}/${FS_LOCATION}/${directory}/${file}" ] && [ ! -e "${NEW_ROOT}/${directory}/${file}" ]; then
ln -s "/${FS_LOCATION}/${directory}/${file}" "${directory}/${file}" 2> /dev/null
fi
done
else
# It does not exist, make a link to the livecd
ln -s "/${FS_LOCATION}/${directory}" "${directory}" 2>/dev/null
current_parent=${directory}
fi
done
fi
done
if [ "${DO_slowusb}" ] || [ "${FORCE_slowusb}" ]
then
sleep 10
fi
mkdir initrd proc tmp sys 2>/dev/null
chmod 1777 tmp
fi
#UML=`cat /proc/cpuinfo|grep UML|sed -e 's|model name.*: ||'`
#if [ "${UML}" = 'UML' ]
#then
# # UML Fixes
# good_msg 'Updating for uml system'
#fi
# Let Init scripts know that we booted from CD
export CDBOOT
CDBOOT=1
if [ "${DO_slowusb}" ] || [ "${FORCE_slowusb}" ]
then
sleep 10
fi
else
if [ "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
union_insert_dir ${UNION} ${NEW_ROOT}
mkdir -p ${UNION}/tmp/.initrd
fi
fi
# Execute script on the cdrom just before boot to update things if necessary
cdupdate
if [ "${SUBDIR}" != '' -a -e "${CHROOT}/${SUBDIR}" ]
then
good_msg "Entering ${SUBDIR} to boot"
CHROOT=${CHROOT}/${SUBDIR}
fi
verbose_kmsg
# Move the /memory mount point to what will be the system root so that
# init scripts will be able to unmount it properly at next reboot
#
# Eventually, all "unions over /" mounts should go in that /.unions/
if [ "${USE_UNIONFS_NORMAL}" -eq '1' ]
then
mkdir -p /${CHROOT}/.unions/memory 2>/dev/null
mount -o move /memory /${CHROOT}/.unions/memory || echo '*: Failed to move unionfs /memory into the system root!'
fi
if [ "$0" = '/linuxrc' ] || [ "$0" = 'linuxrc' ]
then
[ ! -e ${CHROOT}/dev/console ] && mknod ${CHROOT}/dev/console c 5 1
echo -ne "${GOOD}>>${NORMAL}${BOLD} Booting"
cd ${CHROOT}
mkdir ${CHROOT}/proc ${CHROOT}/sys 2>/dev/null
pivot_root . tmp/.initrd
if [ "${DO_slowusb}" ] || [ "${FORCE_slowusb}" ]
then
sleep 10
fi
echo -n '.'
if /tmp/.initrd/bin/[ "${USE_DEVFS_NORMAL}" -eq '1' -a "${CDROOT}" -eq '0' ]
then
umount /tmp/.initrd/proc || echo '*: Failed to unmount the initrd /proc!'
mount -n --move /tmp/.initrd/dev dev || echo '*: Failed to move over the /dev tree!'
rm -rf /tmp/.initrd/dev || echo '*: Failed to remove the initrd /dev!'
elif /tmp/.initrd/bin/[ "${USE_UDEV_NORMAL}" -eq '1' ]
then
/tmp/.initrd/bin/[ -e /tmp/.initrd/dev/fd ] && rm /tmp/.initrd/dev/fd
/tmp/.initrd/bin/[ -e /tmp/.initrd/dev/stdin ] && rm /tmp/.initrd/dev/stdin
/tmp/.initrd/bin/[ -e /tmp/.initrd/dev/stdout ] && rm /tmp/.initrd/dev/stdout
/tmp/.initrd/bin/[ -e /tmp/.initrd/dev/stderr ] && rm /tmp/.initrd/dev/stderr
/tmp/.initrd/bin/[ -e /tmp/.initrd/dev/core ] && rm /tmp/.initrd/dev/core
umount /tmp/.initrd/dev || echo '*: Failed to unmount the initrd /dev!'
umount /tmp/.initrd/proc || echo '*: Failed to unmount the initrd /proc!'
umount /tmp/.initrd/sys || echo '*: Failed to unmount the initrd /sys!'
elif /tmp/.initrd/bin/[ "${CDROOT}" -eq '1' ]
then
umount /tmp/.initrd/proc || echo "*: Failed to unmount the initrd /proc!"
umount /dev 2>/dev/null
mount -n --move /tmp/.initrd/dev dev 2>/dev/null
rm -rf /tmp/.initrd/dev || echo '*: Failed to remove the initrd /dev!'
umount /sys 2>/dev/null
umount /tmp/.initrd/sys 2>/dev/null
fi
echo -n '.'
# /usr/src/linux/Documentation/initrd.txt:
# exec chroot . /sbin/init </dev/console >/dev/console 2>&1
exec <dev/console >dev/console 2>&1
echo '.'
exec chroot . /bin/sh <<- EOF
umount /tmp/.initrd || echo "*: Failed to unmount the initrd!"
/sbin/blockdev --flushbufs /dev/ram0 >/dev/null 2>&1
exec ${REAL_INIT:-/sbin/init} ${INIT_OPTS}
EOF
elif [ "$0" = '/init' ]
then
[ ! -e ${CHROOT}/dev/console ] && mknod ${CHROOT}/dev/console c 5 1
[ ! -e ${CHROOT}/dev/tty1 ] && mknod ${CHROOT}/dev/tty1 c 4 1
echo -ne "${GOOD}>>${NORMAL}${BOLD} Booting (initramfs)"
cd ${CHROOT}
mkdir ${CHROOT}/proc ${CHROOT}/sys 2>/dev/null
echo -n '.'
umount /sys || echo '*: Failed to unmount the initrd /sys!'
umount /proc || echo '*: Failed to unmount the initrd /proc!'
echo -n '.'
exec switch_root -c "/dev/console" "${CHROOT}" ${REAL_INIT:-/sbin/init} ${INIT_OPTS}
fi
splash 'verbose'
echo 'A fatal error has probably occured since /sbin/init did not'
echo 'boot correctly. Trying to open a shell...'
echo
exec /bin/bash
exec /bin/sh
exec /bin/ash
exec sh
| Again thanks for taking the time and excuse my noobishness in this field. I was learning bash, then I got interested in C... then all that got put on the backburner while I was having so much fun playing with edje. That's my problem. Trying to takle 3 things at once instead of concentrating on 1. Anyway I'll be going back to learning C soon, the E17 stuff can wait. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
kernelOfTruth Watchman
![Watchman Watchman](/images/ranks/rank-G-2-watchman.gif)
![](images/avatars/gallery/Star Wars/movie_star_wars_obi-wan_kenobi.gif)
Joined: 20 Dec 2005 Posts: 6111 Location: Vienna, Austria; Germany; hello world :)
|
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 11:38 am Post subject: |
|
|
Quote: | No because the kernel I'm currently running (not the one on the CD) doesn't have the squashfs support option enabled. Is this part of the problem? | It shouldn't be a problem, but would be a bit helpful, if you could look into the squash. Just for checking that everything is in place...
Quote: |
? (I'm tired of burning coasters ) | There is somewhere (the wiki i think) a little howto test just the plain iso with Qemu. Also Vmware is a nice option. And i alway have a bunch of cdrws laying around....
Quote: | Mediafound on /dev/iseries/vcd* | I think this is probably your real problem! Where is your cdrom located with the Kernel you want to run? i've never heard of /dev/iseries/vcd!?! isn't it normally /dev/hd[a-d]1 or /dev/sd[a-d]1!?! I don't have any nforce boards here, probably this could be correct, but then the star behind the device is confusing!?!
Perhaps you should just check something like this in initrd.scripts instead of the original findcdmount() function [i only added the comment at this code-snipptes line 4 and made it a bit more agressive in the following line ]:
Code: | findcdmount() {
if [ "$#" -gt "0" ]
then
#We are now going to take the brute-force variation of finding the right drive...
for x in /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1 $*
do
# Check for a block device to mount
if [ -b "${x}" ]
then
good_msg "Attempting to mount media:- ${x}"
if [ -z "${CDROOT_DEV}" ]
then
mount -r -t auto ${x} ${NEW_ROOT}/mnt/cdrom \
> /dev/null 2>&1
else
mount -r -t iso9660 ${x} ${NEW_ROOT}/mnt/cdrom \
> /dev/null 2>&1
fi
if [ "$?" = '0' ]
then
# Check for a LiveCD
if [ -e ${NEW_ROOT}/mnt/cdrom/${SUBDIR}/livecd ]
then
REAL_ROOT="${x}"
break
else
umount ${NEW_ROOT}/mnt/cdrom
fi
fi
fi
done
if [ "${REAL_ROOT}" != '' ]
then
good_msg "Media found on ${x}"
fi
fi
}
|
Hopefully it is this strange /dev/iseries thing... Otherwise we have to look where we want to place the /bin/sh in!
Nevermind, i am more than a beginner in C, and i think it would be very boring if anybody would have the same interests everywhere...
@kernelOfTruth
I had much problems with recent kernels and genkernel --unionfs-dev! Did it work for you??? |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
wynn Advocate
![Advocate Advocate](/images/ranks/rank-G-1-advocate.gif)
![](images/avatars/46695354144c509f41a088.png)
Joined: 01 Apr 2005 Posts: 2421 Location: UK
|
Posted: Thu Mar 29, 2007 3:55 pm Post subject: |
|
|
nativemad wrote: | sirdilznik wrote: | Mediafound on /dev/iseries/vcd* | I think this is probably your real problem! | You must be right, /dev/iseries/vcd is the fourth virtual disk on an IBM iSeries computer (http://www-03.ibm.com/systems/i/hardware/) â unless, of course, that's what sirdilznik has ! _________________ The avatar is jorma, a "duck" from "Elephants Dream": the film and all the production materials have been made available under a Creative Commons Attribution 2.5 License, see orange.blender.org for details. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Thu Mar 29, 2007 4:03 pm Post subject: |
|
|
wynn wrote: | nativemad wrote: | sirdilznik wrote: | Mediafound on /dev/iseries/vcd* | I think this is probably your real problem! | You must be right, /dev/iseries/vcd is the fourth virtual disk on an IBM iSeries computer (http://www-03.ibm.com/systems/i/hardware/) â unless, of course, that's what sirdilznik has ! | Nope, dual-core Opteron 165. Click my sig for further specs. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 4:28 pm Post subject: |
|
|
He meant, that /dev/iseries/vcd* which seems to get detected by the ramdrive as cdrom cant be your real cdrom! Unless you spent a bit more money to IBM, as an iseries computer is a big Mainframe, which normally runs AIX or OS400! You even can't get a native linux install on that Hardware, as far as i know! -but up to 32 virtual machines...
Also, this is NOT an x86 architecture!! i think iseries are alpha, ppc64 and perhaps the Cell (like ps3) But absoutely unsure, as i prefer Sun mainframes.......
Long story short: Please check that you have choosen the right cpu architecture in your kernelconfig! If this doesn't Help, try the findcdmount() function i posted earlier....
I still wonder why you got /dev/iseries, and the Kernel boots!!!!???!!! Do you manage your /dev manually without udev or what?!? ![Shocked 8O](images/smiles/icon_eek.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Thu Mar 29, 2007 4:34 pm Post subject: |
|
|
nativemad wrote: | He meant, that /dev/iseries/vcd* which seems to get detected by the ramdrive as cdrom cant be your real cdrom! Unless you spent a bit more money to IBM, as an iseries computer is a big Mainframe, which normally runs AIX or OS400! You even can't get a native linux install on that Hardware, as far as i know! -but up to 32 virtual machines...
Also, this is NOT an x86 architecture!! i think iseries are alpha, ppc64 and perhaps the Cell (like ps3) But absoutely unsure, as i prefer Sun mainframes.......
Long story short: Please check that you have choosen the right cpu architecture in your kernelconfig! If this doesn't Help, try the findcdmount() function i posted earlier....
I still wonder why you got /dev/iseries, and the Kernel boots!!!!???!!! Do you manage your /dev manually without udev or what?!? ![Shocked 8O](images/smiles/icon_eek.gif) | Architecture chosen is Generic x86_64
Do I replace the findfsmount() function in the initrd.scripts with your findcdmount() function or just add it in?
Edit: Also I wanted to use Qemu, but can't since i don't have gcc-3.x.x only gcc-4.1.2 and gcc-4.2.0. I guess libdistc++v3 isn't enough to compile Qemu. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 4:42 pm Post subject: |
|
|
You can just edit the line below in your existing function, its just the list of possible cdrom devices.
for x in /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1 $*
instead of
for x in $*
That should help! ![Very Happy :D](images/smiles/icon_biggrin.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Thu Mar 29, 2007 4:51 pm Post subject: |
|
|
nativemad wrote: | You can just edit the line below in your existing function, its just the list of possible cdrom devices.
for x in /dev/hda1 /dev/hdb1 /dev/hdc1 /dev/hdd1 /dev/sda1 /dev/sdb1 /dev/sdc1 /dev/sdd1 $*
instead of
for x in $*
That should help! ![Very Happy :D](images/smiles/icon_biggrin.gif) | Doh! I looked at the script wrong. I'll give it a shot, thanks.
Just to give more info. I have these drives :
/dev/hda = ATA hard drive
/dev/hdc = dvd burner
/dev/hdd = dvd rom
/dev/sda = SATA hard drive (where / is located) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 4:58 pm Post subject: |
|
|
ok, then you also could do it like
for x in /dev/hdc1 $*
for ultra fast booting from the burner on this machine....
Dont forget to rebuild your initrd!!! genkernel initrd should do it. than rename and so on..... |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
wynn Advocate
![Advocate Advocate](/images/ranks/rank-G-1-advocate.gif)
![](images/avatars/46695354144c509f41a088.png)
Joined: 01 Apr 2005 Posts: 2421 Location: UK
|
Posted: Thu Mar 29, 2007 5:54 pm Post subject: |
|
|
nativemad wrote: | ok, then you also could do it like
for x in /dev/hdc1 $*
for ultra fast booting from the burner on this machine....
Dont forget to rebuild your initrd!!! genkernel initrd should do it. than rename and so on..... | Shouldn't it be Code: | for x in /dev/hdd1 /dev/hdc1 $* | for ultra-fast whotsit as /dev/hdd is the reader and /dev/hdc the burner? _________________ The avatar is jorma, a "duck" from "Elephants Dream": the film and all the production materials have been made available under a Creative Commons Attribution 2.5 License, see orange.blender.org for details. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Thu Mar 29, 2007 6:07 pm Post subject: |
|
|
nativemad wrote: | ok, then you also could do it like
for x in /dev/hdc1 $*
for ultra fast booting from the burner on this machine....
Dont forget to rebuild your initrd!!! genkernel initrd should do it. than rename and so on..... | I tried it with that and with the disc in both drives. No difference, just another coaster to add to my growing collection. ![Crying or Very sad :cry:](images/smiles/icon_cry.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 6:09 pm Post subject: |
|
|
Quote: | for ultra-fast whotsit as /dev/hdd is the reader and /dev/hdc the burner? |
Ok, that way it fist checks the reader... If you get really lazy, the /dev/hdc option allows you to just leave the fresh-burned cd in the burner and don't waste the time for an attempt to mount the other drive!
-just joking around!- |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
wynn Advocate
![Advocate Advocate](/images/ranks/rank-G-1-advocate.gif)
![](images/avatars/46695354144c509f41a088.png)
Joined: 01 Apr 2005 Posts: 2421 Location: UK
|
Posted: Thu Mar 29, 2007 6:18 pm Post subject: |
|
|
nativemad wrote: | Quote: | for ultra-fast whotsit as /dev/hdd is the reader and /dev/hdc the burner? |
Ok, that way it fist checks the reader... If you get really lazy, the /dev/hdc option allows you to just leave the fresh-burned cd in the burner and don't waste the time for an attempt to mount the other drive!
-just joking around!- | Aah that's it, it takes me some time to catch up ![Smile :)](images/smiles/icon_smile.gif) _________________ The avatar is jorma, a "duck" from "Elephants Dream": the film and all the production materials have been made available under a Creative Commons Attribution 2.5 License, see orange.blender.org for details. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 6:22 pm Post subject: |
|
|
Quote: | I tried it with that and with the disc in both drives. No difference, just another coaster to add to my growing collection. | Don't you have some cd-rw's?
Ok, i would then try to eliminate that strange device-file in the init! The probably easyest way is this:
Code: |
findcdmount() {
if [ "$#" -gt "0" ]
then
#lets kick that iseries thing!
rm -rf /dev/iseries >/dev/null 2>&1 # sorry i forgot the > before editing!
for x in $*
do
|
or you also could try it manually...
Code: | findcdmount() {
if [ "$#" -gt "0" ]
then
#lets openup a shell to have a look whats going on!
/bin/sh
for x in $*
do
|
EDIT: forgot the > the first time in code nr. 1!!!!!!
Last edited by nativemad on Thu Mar 29, 2007 6:31 pm; edited 2 times in total |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Thu Mar 29, 2007 6:25 pm Post subject: |
|
|
nativemad wrote: | Quote: | I tried it with that and with the disc in both drives. No difference, just another coaster to add to my growing collection. | Don't you have some cd-rw's?
Ok, i would then try to eliminate that strange device-file in the init! The probably easyest way is this:
Code: |
findcdmount() {
if [ "$#" -gt "0" ]
then
#lets kick that iseries thing!
[b] rm -rf /dev/iseries /dev/null 2>&1[/b]
for x in $*
do
|
or you also could try it manually...
Code: | findcdmount() {
if [ "$#" -gt "0" ]
then
#lets openup a shell to have a look whats going on!
[b]/bin/sh[/b]
for x in $*
do
|
| No CD-RWs, but I have plenty of CD-Rs (and they're cheap), like 50 more. I have to go to work now, but I will try your suggestion when I get back. Also I will recompile my current kernel to include squashfs so I can mount the disc in my current system. Again thanks for all the help. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
kernelOfTruth Watchman
![Watchman Watchman](/images/ranks/rank-G-2-watchman.gif)
![](images/avatars/gallery/Star Wars/movie_star_wars_obi-wan_kenobi.gif)
Joined: 20 Dec 2005 Posts: 6111 Location: Vienna, Austria; Germany; hello world :)
|
Posted: Thu Mar 29, 2007 6:35 pm Post subject: |
|
|
nativemad wrote: |
@kernelOfTruth
I had much problems with recent kernels and genkernel --unionfs-dev! Did it work for you??? |
I just omitted this setting & used the unionfs-version from the kernel (I compiled it in), perhaps therefore I'm experiencing some weird errors when trying to set up cupsd & installing additional printer drivers on the livecd, I think -> it runs out of space, although there's still 1 GB remaining
(I'm on "small gentoo X [internal test build ] with xorg & printer-support) _________________ https://github.com/kernelOfTruth/ZFS-for-SystemRescueCD/tree/ZFS-for-SysRescCD-4.9.0
https://github.com/kernelOfTruth/pulseaudio-equalizer-ladspa
Hardcore Gentoo Linux user since 2004 ![Very Happy :D](images/smiles/icon_biggrin.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
nativemad Developer
![Developer Developer](/images/ranks/rank-dev.gif)
![](images/avatars/gallery/Simpsons/Simpsons_-_Krusty.jpg)
Joined: 30 Aug 2004 Posts: 918 Location: Switzerland
|
Posted: Thu Mar 29, 2007 6:55 pm Post subject: |
|
|
Quote: | I just omitted this setting & used the unionfs-version from the kernel (I compiled it in), perhaps therefore I'm experiencing some weird errors when trying to set up cupsd & installing additional printer drivers on the livecd, I think -> it runs out of space, although there's still 1 GB remaining |
I also compiled it in the kernel and leaved the genkernel-option away. But to get unionfs really working i have it like this, somewhere in linuxrc (this is just a snippet, as my files are really, really modded ):
Code: | mkdir ${NEW_ROOT}/root
for x in usr lib opt bin sbin ${ROOT_TREES}
do
if [ ! -e ${NEW_ROOT}/${x} ] || [ ! -e ${NEW_ROOT}/mnt/livecd/${x} ]
then
mkdir ${NEW_ROOT}/${x}
mkdir ${NEW_ROOT}/root/${x}
mount -t unionfs -o rw,dirs=${NEW_ROOT}/root/${x}=rw:${NEW_ROOT}/mnt/livecd/${x}=ro writeloop${x} ${NEW_ROOT}/${x}
fi
done
|
You should be able to see some union-mounts with "mount" without parameters. |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
sirdilznik l33t
![l33t l33t](/images/ranks/rank_rect_4.gif)
![](images/avatars/141045994047641242b3744.jpg)
Joined: 28 Apr 2005 Posts: 731
|
Posted: Sat Mar 31, 2007 4:59 pm Post subject: |
|
|
YES!!!!!!!! Success!! sirdilznix 0.1 is alive!!!!
I started over from scratch and decided to make this LiveCD specific to my particular computer and not "universal" like I was trying to make before. This time I basically copied the kernel config from my current kernel to the kernel on the LiveCD, then made sure all the necessary options were enabled (they were) as opposed to running make allmodconfig, then building certain options directly in. My goal for this LiveCD was to make a disc that would be a minimal LiveCD proper for installing or rescuing a reiser4 partition with the new cryptcompress plugin (I'll be experimenting with that soon *rubs hands together*), thus I needed a very recent mm kernel and reiser4progs-1.0.6. I also added support for all other major filesystems (including ext4) and the necessary tools. I used mm-sources-2.6.21-rc5-mm3. I added links and mc for the bare minimum for installation along with mpg123 and cdplay so I could listen to music, and angband and moon-buggy for entertainment purposes while stuff is compiling, etc... It works!
Now that i have a functioning disc for myself I can work on making a more universal one for the masses (an ~amd64 one first, since that is what I have now, and if there is demand I'll make a ~x86 one too). I'm convinced that there was probably some kind of conflict between drivers with too many modules being loaded and conflicting with each other, or too many options built directly into the kernel conflicting with each other. I think that is what was causing the disc to have faulty detection. Now that I have a working disc for myself I can take my time making the universal ones. I will probably include a stage3 tarball and a portage snapshot tarball for those that won't have a working internet connection to start with (not an issue on my machine, dhcp works just fine).
Thank you everybody for all the help. I really appreciate it. I learned a bunch during this whole process and will learn a lot more during the making of the universal versions. Hopefully sirilznix 0.2 ~amd64 and ~x86 will be available soon. ![Cool 8)](images/smiles/icon_cool.gif) |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
naclosagc n00b
![n00b n00b](/images/ranks/rank_rect_0.gif)
Joined: 26 Jan 2006 Posts: 12 Location: St Louis, MO
|
Posted: Tue Apr 03, 2007 2:56 pm Post subject: Creating a LiveCD |
|
|
I am attempting to create a livecd using the instructions fromhttp://gentoo-wiki.com/HOWTO_build_a_LiveCD_from_scratch , and it is going pretty well. The problem, is, that when it boots, it tells me my loops is not found. I'm gonna include everything I can think of without making the post totally humungous:
~/livecd/source/boot/grub/menu.lst:
default 0
timeout 30
splashimage=/boot/grub/splash.xpm.gz
title=LiveCD
kernel /boot/vmlinuz root=/dev/ram0 real_root=/dev/loop0 looptype=squashfs loop=/livecd.squashfs udev nodevfs cdroot
initrd /boot/initrd
title=Memtest86+
kernel /boot/memtest86plus/memtest.binmksquashfs ~/livecd/target/source/ ~/livecd/target/holding/livecd.squashfs -noappend
mksquash and mkisofs commands:
mksquashfs ~/livecd/target/source/ ~/livecd/target/holding/livecd.squashfs -noappend
touch ~/livecd/target/holding/livecd
mkisofs -R -b boot/grub/stage2_eltorito -no-emul-boot -boot-load-size 4 -boot-info-table -iso-level 4 -o ~/livecd.iso ~/livecd/target/holding
directory Structure:
/root - - - source #where I build the system
|
- - - target
|
- - - source #the reduced source, without such things as portage, etc - created using rsync
|
- - - holding
|
- - - boot
|
- - - livecd.squashfs
Messages:
>> Attempting to mount media:- /dev/hda
>> Attempting to mount media:- /dev/hda2
>> Attempting to mount media:- /dev/hda3
>> Media found on /dev/iseries/vcd*
>> Deterimining root device. . .
>>Determining looptype . . .
!! Invalid loop location: /livecd.squiashfs
!! Please export LOOP with a valid location, or reboot and pass a proper loop=. . .
!! kernel command line
/dev/hda is a hard disk, and I am not sure what /dev/iseries/vcd* is. Thanks for any input on this. _________________ Andy Carlson |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
vibrokatana Guru
![Guru Guru](/images/ranks/rank_rect_3.gif)
![](images/avatars/16290473954616d80d58b91.jpg)
Joined: 09 Feb 2006 Posts: 328 Location: o0o0oo
|
Posted: Tue Apr 03, 2007 3:39 pm Post subject: |
|
|
Make sure "loop=/livecd.squashfs" is in the proper place on the livecd, it may not be there and complains. _________________ My Systems - "I suggest the whole thing be coded in whitespace. Henceforth the code will be obscure and functional at the same time." |
|
Back to top |
|
![](templates/gentoo/images/spacer.gif) |
|
|
You cannot post new topics in this forum You cannot reply to topics in this forum You cannot edit your posts in this forum You cannot delete your posts in this forum You cannot vote in polls in this forum
|
|