kernel panic : Upgraded from 2.4.x to 2.6.17.6
HI
this is my 3rd day ..n i m still stuck with kernel panic ... i hav search n tried every possible solution but nothing worked ... i hav upgraded my kernel from 2.4.X(default in Slackware 10.2) to kernel 2.6.17.6 ... i m using ext3 fs n enabled ext3 fs in the kernel as builtin(no as module) ... now when i compile n install my kernel ... n reset it .. booom .... a kernel panic error... u can see my system detail below .... Code:
VFS: Cannot open root device "303" or unknown_block(3,3) u can see my lilo.conf Code:
# LILO configuration file fstab Code:
/dev/hda4 swap swap defaults 0 0 lspci Code:
00:00.0 Host bridge: Intel Corporation 82845G/GL[Brookdale-G]/GE/PE DRAM Controller/Host-Hub Interface (rev 03) ls -al /dev/hda3 Code:
brw-rw---- 1 root disk 3, 3 2002-06-09 11:27 /dev/hda3 ls -al /boot/ Code:
total 9856 Code:
Filesystem 1M-blocks Used Available Use% Mounted on hope u guyz could help me out :confused: Regards Slack BABY |
This sort of error almost always comes from one of two things, the filesystem or hard disk support. Since you've compiled in the file system, its likely you didn't compile in support for your hard disk drive (in Device Drivers > ATA/ATAP/MFM/RLL support). Again, this needs to be in the kernel and not as a module.
|
thankx Hangdog42 .... i will try it n let u know ...
|
well i hav tried this n its workd .. but unfortunately .. i m starting getting other kernel panic error
Code:
Kernel Panic - not syncing: I/0 error reading memory image |
Hm. Googling on that error was surprisingly uninformative. I'm going to guess that this has something to do with the intial RAM disk support. Check in Device Drivers > Block Devices that you have RAM Disk support and Initial RAM Filesystem are built into the kernel. For RAM disks, I use the defaults for number of RAM disks (16) and RAM disk size (4096). You'll probably want to check on enabling loopback device support as long as your there.
|
thankx Hangdog42 once again ... let me check n will let u know
|
i enabled RAMDisk and RAMSize in a kernel n compile it ... but it didnt sovle da problem .. still getting the same error. ....
|
Quote:
The item I've bolded you should delete. It's linking to the config file for the kernel you're not using. I have made this mistake before. If it doesn't work you can always relink it after. Just in case... make a slackware boot disk in case you get funky problems that won't let you go back and relink that old config... but I've never needed that. Never hurts to be careful :D That's the only reason I can think of why the kernel is having an I/O sync error. With your new kernel you should either compile the .config file into the kernel or put it into /boot with your kernel. The kernels that don't have the config file compiled in will look for a file "config" and if it's different than what you compiled with it will complain like a child. Let us know if that worked... it did for me. |
hey zetabill .. i hav tried wat u said .. but it didnt work .... :cry:
any other solution ? |
okz .. here is update ...
i have notice the two lines about kernel panic Code:
attempt to access beyoud end of device |
It probably isn't a LILO problem since this is specific to this kernel, and given the lilo.conf entry you posted, it would affect ALL kernels.
To be honest, I'm kind of stumped. One thing to try would be to load one of the 2.6.x stock kernels from Slackware (have a look in testing for 2.6.13). If that works, you can swipe the .config from that kernel, then run make oldconfig for your 2.6.17.6. If that works, then you can start adjusting the config until it better meets your needs. |
I agree with Hangdog42...
The only other question I'd have is whether or not you have vfat compiled into the kernel? The only reason I ask is that your hda2 is actually an extended partition which is the sum total of hda5 and hda6 which are logical partitions. If it's complaining about hda2, then it might not know that it's an extended partition or doesn't recognize the type. This is a problem because the logical (hda5, hda6) partitions occur in the space that would be hda2... which exists before your root partition (hda3). If it can't figure out your logical partitions, then it probably can't figure out where hda2 (hda5 + hda6) ends and hda3 begins. Since your Slackware with the 2.4 kernel works fine I can only assume that the kernel your compiling sees the partitions but has no clue what they are and doesn't want to do anything for fear of overwriting data. It's just a thought. I hope it works for you because I'm out of ideas. Good luck. EDIT: Cleaned up grammar. |
yes ..zetabill i hav compiled vfat in a kernel .. infact i hav compiled all fs in kernel .. but still having same the same kernel panic ....
@Hangdog42 thankx 4 ur help... i will try to load it ... |
You need to define a root partition in lilo.conf. try root=/dev/hda<number>
Also, you better write "read-only" in the lilo.conf |
i got something interested ...
http://tastytronic.net/pipermail/poc...ne/000254.html http://tastytronic.net/pipermail/poc...ne/000255.html i guess i m also having the same problem .. where i think in daniel's problem the 01:00 is the device(partition) with major 1 n minor 0 .. in the second link u can see that dainel has solved the problem by setting ram_disk in grub .. i dont know how to do it with lilo .... need suggestiosn... |
All times are GMT -5. The time now is 09:32 AM. |