For starters, yes I have looked at similar threads and nothing seemed quite the same nor did it seem to be at all useful or make any sense to me, of coure I am new to linux so not much makes sense to me, anyway here is the deal.

I am dual booting Windows XP and SUSE Linux 10.2. I did the "Online Update" which is listed in Yast. It updated the kernel, kernel sources and a host of other things. It told me to run /sbin/lilo when done update. I did and got an error of "cannot find kernel-default" or something like that. I reboot and got a kernel panic. I ran the installation from the DVD, went to "other options" and tried the "automatic repair", received dome errors, most of which it said it fixed. Reboot again and got the same thing. I really don't want to have to back everything up and reinstall from scratch again. It's taken me over 2 weeks to get most things working the way I want and don't really have time to spend another two weeks getting things installed and configured.

Everything looks fine on reboot until I get the below messages.

RAMDISK: Couldn't find valid RAM disk image starting at 0. <---prior to this there is a message about it setting up a ramdisk which seemed like it was successful.
VFS: Cannot open root device "303" or unknown-block(3,3)
Please append correct "root=" boot option
Kernel panic - not syncing" VFS: Unable to mount root fs on unknoen-block(3,3)

I have it setup using LILO as a boot loader ang gnome as a desktop manager. hda1 is windows boot drive, hda2 is swap, hda3 is root, hdb and hdd are additional drives, all drives are ide and all are ext3 except the windows boot drive is of course NTFS.

I obviously have no access to linux now and am currently in windows writing this message, I can access the ext3 partitions in windows as I have the "Ext2 Installable File System for Windows" from fs-driver.org installed.

Given these circumstances, how can I fix my linux installation so I can boot to it again. Instructions should assume I know nothing. I know a lot about computers and windows but not much about linux aside from how to install it, add rpms, and have done the occasional make/make install from tar.gz when necessary.

Thanks in advance for any help, especially if it enables me to fix this problem.

Update:
Additional info which may be useful.

Here is my /etc/lilo.conf

# Modified by YaST2. Last modification on Tue May 1 01:23:13 ADT 2007
menu-scheme = Wb:kw:Wb:Wb
timeout = 80
lba32
change-rules
reset
read-only
prompt
default = oS_10.2
message = /boot/message
boot = /dev/hda3

image = /boot/vmlinuz
###Don't change this comment - YaST2 identifier: Original name: linux###
label = oS_10.2
append = " resume=/dev/hda2 splash=silent showopts"
vga = 0x317
initrd = /boot/initrd
root = /dev/hda3

image = /boot/vmlinuz
###Don't change this comment - YaST2 identifier: Original name: failsafe###
label = Failsafe
append = "showopts ide=nodma apm=off acpi=off noresume nosmp noapic maxcpus=0 edd=off 3"
vga = normal
initrd = /boot/initrd
root = /dev/hda3

other = /dev/hda1
label = Windows_XP

This is a listing of my /boot folder, I included the file sizes in case this helps. Some of them seem small to me but maybe that is normal.

<DIR> grub
1 boot
379,904 message
88,195 symsets-2.6.18.8-ccj45-bigsmp.tar.gz
85,935 symsets-2.6.18.2-34-xenpae.tar.gz
29 initrd
30 vmlinuz
498,176 map
88,307 symsets-2.6.18.8-ccj45-desktop.tar.gz
88,521 symsets-2.6.18.8-ccj45-debug.tar.gz
87,897 symsets-2.6.18.8-ccj45-kdump.tar.gz
446 backup_mbr
512 boot.0303
94,632 memtest.bin
88,261 symsets-2.6.18.8-ccj45-default.tar.gz
908,044 System.map-2.6.18.8-ccj45-default
74,250 config-2.6.18.8-ccj45-default
3,240,830 initrd-2.6.18.8-ccj45-default
344,836 symtypes-2.6.18.8-ccj45-default.gz
101,445 symvers-2.6.18.8-ccj45-default.gz
1,830,869 vmlinux-2.6.18.8-ccj45-default.gz
1,545,223 vmlinuz-2.6.18.8-ccj45-default
87,301 symsets-2.6.18.8-ccj45-xen.tar.gz

In other posting on similar problem ther were a lot of suggestions which one poster compiled into a single post. Here it is as well as some information on what I have done or can do etc. My comments in red.

Here are all the solutions I found on this page (as of 2005-11-22)

in kernel: Not much I can do in the kernel as this was a yast update and it should have taken care of what it needed to. If not, I cannot boot linux and therefore cannot rebuild the kernel unless there is some other way to do it, like from the installation DVD or something. Furthermore, if I could get into linux to rebuild the kernel and do what is suggested below, I wouldn't have the first clue as to how to do this.
- make sure you have IDE devices built into the kernel (not a module)
- if you're booting from SCSI or SATA ditto for SCSI support
- same goes for your filesystem... make it built-in not modules
- advanced partitioning support, ditch it if you don't need it
- try increasing the size of RAMDISK support, if it's loading to RAMDISK to boot, you'll need that to be bigger than your kernel
- Disable root plug option (security options) if not needed

in your bootmanager (probably lilo or grub) :
- vga=normal will usually fix the "black screen" problem, if not vga=ask No black screen so this obviously is not a problem.
- try taking out the root=/dev/xxxx parameterTried this and it didn't help.
- add/edit lilo.conf with this: label=/dev/hda This didn't help either.
- get rid of lba32 in lilo.conf And this did not help either.

No idea what the below block means or how I could accomplish this if I can't boot into linux to run any commands.
Try booting from an image (I believe this is so you can load the IDE or SCSI modules. If you have a problem with built-in support for things, then this would presumably be an alternative... but it didn't work for me)
sudo mkinitrd -o /boot/initrd-2.6.0.img 2.6.0
If lilo, add this line to lilo.conf: initrd=/boot/initrd-2.6.0.img
if grub, add this to menu.lst: initrd /boot/initrd-2.6.0.img

Oh, and here's my fstab if that is of interest, need anything else, let me know.

/dev/hda3 / ext3 acl,user_xattr 1 1
/dev/hda2 swap swap defaults 0 0
proc /proc proc defaults 0 0
sysfs /sys sysfs noauto 0 0
debugfs /sys/kernel/debug debugfs noauto 0 0
usbfs /proc/bus/usb usbfs noauto 0 0
devpts /dev/pts devpts mode=0620,gid=5 0 0
/dev/fd0 /media/floppy auto noauto,user,sync 0 0
/dev/hda1 /home/sandra/WindowsC ntfs ro,users,gid=users,umask=0002,nls=utf8 0 0
/dev/hdb1 /home/sandra/WindowsD ext3 acl,user_xattr 1 2
/dev/hdd1 /home/sandra/WindowsE ext3 acl,user_xattr 1 2


Maybe someone out there can makes heads or tails out of all this. Thanks again.

(quote omitted)

Sorry, but there was too much info to quote (but thanks for all of it!).

I had a similar issue previously, though I didn't get the error specified. If you can't boot to Linux, though, your problem sounds similar.

Mine turned out to be a boot partition failure, or rather mbr problem. I had to boot to the installer and re-config the boot options. There is an issue with this, though: you'll lose your Windows config line in Lilo. I'm not sure how to resolve that other than manually adding it later, but this solution fixed the issue for me (I'm running OpenSUSE 10.2, 2.6.18; I also had the problem with the update).

Gryyphyn, out.

Hi, I hesitate to jump in here because I am by no means an expert with Linux myself, so please bear with me if my replies are not perfect. You may need to use some of your own savvy to fill in the gaps. I have a dual boot system (Suse 10.1 & WinXP) and have had a similar if not identical experience. After a Kernel Update. You stated you had LILO and the Gnome Desktop installed, please don't take this the wrong way; but are you sure it's 'LILO' installed, because Suse installs the 'GRUB' loader by default. The reason I ask this is because I use the default GRUB Loader & KDE so anything I tell you will defer to this. Sometimes during Kernel updates the Boot Loader seems to be demolished, and then can't find the Partitions, my first way to fix this was to use the Suse CD/DVD > Start Install > Accept License etc. And when you get to the install page, click the 'Other' button at bottom right, then use the Repair modules or YAST to fix the Bootloader, (something which sometimes scrubs the Windows boot so be careful with that method. The other way is to use the Suse CD to start Linux as before then use YAST to modify the Boot Loader manually. There is actually a section which allows you to restore your Bootloader, and I'm sure someone out there could tell you what lines are in their own LILO dual boot Loader to help you if you want to do it manually, (mine are for Grub & KDE).

I hope this is of some help.
J7

As an added note to my previous post, Novel have just today (May 4th '07) issued a Kernel Update
for Suse and what's really interesting a Perl patch for the Boot Loader, presumably to prevent the
Kernel Updates from scrapping it when it is installed It's a bit like closing the stable door etc. But at least it should help from now on.
Good Luck, J7

Hmmm- I did an upgrade to 10.2, and yesterday (5/4/07) downloaded all of the on-line updates, including a kernel update. After the kernel patch(es), system wouldn't boot. If I've diagnosed it right, it's because the boot-time ram-disk isn't getting the right modules copied over; I'm trying to boot the "bigsmp" kernel, but I'm getting the "default" modules loaded into the ramdisk. Changing the vmlinuz sym-link (in /boot) to the "default" kernel got me up; in single CPU mode. But, I wonder if finding & loading this patch would fix me up.

My online update sources are:
http://ftp.skynet.be/ftp.opensuse.or.../repo/non-oss/
http://ftp.skynet.be/ftp.suse.com/suse/update/10.2/
http://ftp.skynet.be/packman/suse/10.2/
http://ftp.skynet.be/suser-guru/rpm/10.2/
http://ftp.skynet.be/ftp.opensuse.or.../10.2/repo/oss
http://download.nvidia.com/opensuse/10.2/

And as of today (5/5/07), there are "no patches available" (I've got them all).

I use Suse 10.1 with a single core 64bit CPU and it wasn't me that actually received the patch it was a friend who uses Open Suse 10.2. on a Dual Core 64bit machine, I asked him for details of the patch but the only thing he can remember is that it came with a Kernel update and it part of it's name was Perl Bootloader Patch one other thing that may be relevant both of us use the Smart Package Manager as opposed to Zen which only seems to work when it feels like it. I went to Yast > Software Management and typed Bootloader in Search with quite interesting results. My Perl libraries are installed but are highlighted in red and have a problem. I next checked with my friend and asked for the version number of his Perl Bootloader Libraries and his are 0.4.14-1.2 whereas mine are 0.2.27-0.4 . I Googled for 'Perl Bootloader Patch for Suse 64 bit' and it found among others a Recommended update for Perl Bootloader at: support.novell.com/techcenter/psdb/d86260152157c019a26338a6081490a3.html . I am neither experienced enough or clever enough to mess about with alternative Kernels etc. so I only ever try to fix something if a default stops working for some reason, so I couldn't really tell you if this patch/update will help but it seems like it could.
J7

P.S. type this string into Google 'patch-perl-bootloader' and you will find a whole flock of Perl patching links which fix the Bootloader after a Kernel Update, including one which mentions non-standard boot blocks. Sorry I didn't paste the actual links because some of them link straight to an XML file and just show as dead but the name of the file which seems most interesting is:
patch-perl-Bootloader-1928.xml

I hope this is useful.
J7

ok, here's an update, I tried doing the update on a different system and had the same problem, not sure what date that was on, will have to check, I do still plan to attempt a fix on that one just for learning purposes, meanwhile I did reinstall this system from scratch and did the update during the install which worked fine. This was on May 5th so it may have been because of the fix or it just works better doing it during an install. The other difference being that during the install this time I chose grub instead of lilo which may have made a difference as well. I'll have to find some time to play with it some more and see what I can figure out.