[SOLVED] Kernel panic - not syncing

[melodie]

The listing in /kernel/drivers is the same as yours:

is it full in all 3 subdirectories under /lib/modules ?

It would seem that everything is indeed where it ought to be

No sir, the symlinks in the /boot directory are messed up.

config -> config-2.6.32.24-pclos1.a64
initrd.img -> initrd-2.6.33.7-pclos6.a64.img

kernel.h -> /boot/kernel.h-2.6.32.24-pclos1.a64      
System.map -> System.map-2.6.32.24-pclos1.a64
vmlinuz -> vmlinuz-2.6.33.7-pclos6.a64

:-/

kernel config file from 2.6.32.24-pclos1.a64 and try to boot to vmlinuz-2.6.33.7-pclos6.a64 no wonder it doesn't boot.

I won't explain how to change symlinks now, you must first check **the 3** kernel subdirectories, because there might be one (or 2 over 3) where the modules are missing. Then you must choose a kernel version that has it's modules, then modify the symlinks that need to be modified (this is not hard to do, a simple command line).

[IrishGuy]

IrishGuy:

Post the contents of your /etc/fstab.

[root@localhost ~]# cat /etc/fstab                     <Enter>

here it is:

Code: [Select]

[root@localhost drivers]# cat /etc/fstab
/dev/root / rootfs defaults 0 0
none /proc proc defaults 0 0
none /dev/pts devpts defaults 0 0


Melodie: I'll check the contents of the /lib/modules now

[Old-Polack]

The listing in /kernel/drivers is the same as yours:

is it full in all 3 subdirectories under /lib/modules ?

It would seem that everything is indeed where it ought to be

No sir, the symlinks in the /boot directory are messed up.

config -> config-2.6.32.24-pclos1.a64
initrd.img -> initrd-2.6.33.7-pclos6.a64.img

kernel.h -> /boot/kernel.h-2.6.32.24-pclos1.a64      
System.map -> System.map-2.6.32.24-pclos1.a64
vmlinuz -> vmlinuz-2.6.33.7-pclos6.a64

:-/

kernel config file from 2.6.32.24-pclos1.a64 and try to boot to vmlinuz-2.6.33.7-pclos6.a64 no wonder it doesn't boot.

I won't explain how to change symlinks now, you must first check **the 3** kernel subdirectories, because there might be one (or 2 over 3) where the modules are missing. Then you must choose a kernel version that has it's modules, then modify the symlinks that need to be modified (this is not hard to do, a simple command line).

The symlinks, other than vmlinuz and initrd.img, don't matter during boot. Grub only uses those two during a default boot, and the kernel only uses the proper, similarly named versions of those files, not the symlink.

I have a number of installations where the other symlinks didn't get updated with new kernel installs, and none failed to boot because of it, including the first openbox installation.

[root@littleboy ~]# ls -l /mnt/boot/openbox

Code: [Select]

total 17821
-rw-r--r-- 1 root root     440 2010-04-29 09:29 boot.backup.sdc
lrwxrwxrwx 1 root root      27 2010-04-30 06:09 config -> config-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root  109905 2010-04-07 13:38 config-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root  112948 2010-04-03 12:54 config-2.6.33.2-pclos1.bfs
-rwxr-xr-x 1 root root  502272 2010-04-29 09:29 gfxmenu*
drwxr-xr-x 2 root root    1024 2010-04-30 06:09 grub/
-rw------- 1 root root 5133299 2010-04-29 09:29 initrd-2.6.32.11-pclos2.bfs.img
-rw------- 1 root root 5138187 2010-04-30 05:23 initrd-2.6.33.2-pclos1.bfs.img
lrwxrwxrwx 1 root root      30 2010-04-30 06:09 initrd.img -> initrd-2.6.33.2-pclos1.bfs.img
lrwxrwxrwx 1 root root      35 2010-04-30 06:09 kernel.h -> /boot/kernel.h-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root    1493 2010-04-30 05:01 kernel.h-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root  254766 2006-11-05 23:23 message-graphic
lrwxrwxrwx 1 root root      31 2010-04-30 06:09 System.map -> System.map-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root 1454342 2010-04-07 13:38 System.map-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root 1401098 2010-04-03 12:54 System.map-2.6.33.2-pclos1.bfs
-rw-rw-r-- 1 root root     256 2010-04-29 09:29 us.klt
lrwxrwxrwx 1 root root      27 2010-04-30 06:09 vmlinuz -> vmlinuz-2.6.33.2-pclos1.bfs
-rw-r--r-- 1 root root 2056848 2010-04-07 13:38 vmlinuz-2.6.32.11-pclos2.bfs
-rw-r--r-- 1 root root 1992736 2010-04-03 12:54 vmlinuz-2.6.33.2-pclos1.bfs

This was only recently fixed.

[melodie]

IrishGuy:

Post the contents of your /etc/fstab.

[root@localhost ~]# cat /etc/fstab                     <Enter>

here it is:

Code: [Select]

[root@localhost drivers]# cat /etc/fstab
/dev/root / rootfs defaults 0 0
none /proc proc defaults 0 0
none /dev/pts devpts defaults 0 0


Melodie: I'll check the contents of the /lib/modules now

if the fstab you just gave is from the live cd it's not good. You must mount the / partition on a mountpoint (/media ?) then do :

Code: [Select]

cat /media/etc/fstab
Keep courage !

The symlinks, other than vmlinuz and initrd.img, don't matter during boot. Grub only uses those two during a default boot, and the kernel only uses the proper, similarly named versions of those files, not the symlink.

You know for me your word is gold.

Good to know. I leave IrishGuy in your hands, and go sleep now.

regards,
Mélodie

[Old-Polack]

IrishGuy:

Post the contents of your /etc/fstab.

[root@localhost ~]# cat /etc/fstab                     <Enter>

here it is:

Code: [Select]

[root@localhost drivers]# cat /etc/fstab
/dev/root / rootfs defaults 0 0
none /proc proc defaults 0 0
none /dev/pts devpts defaults 0 0


Melodie: I'll check the contents of the /lib/modules now

This appears to be the /etc/fstab from the liveCD, not the installed system. We need the one from the installed system.

[IrishGuy]

IrishGuy:

Post the contents of your /etc/fstab.

[root@localhost ~]# cat /etc/fstab                     <Enter>

here it is:

Code: [Select]

[root@localhost drivers]# cat /etc/fstab
/dev/root / rootfs defaults 0 0
none /proc proc defaults 0 0
none /dev/pts devpts defaults 0 0


Melodie: I'll check the contents of the /lib/modules now

This appears to be the /etc/fstab from the liveCD, not the installed system. We need the one from the installed system.

My bad, here is the one from the hard drive:

Code: [Select]

[root@localhost media]# cat etc/fstab
# Entry for /dev/sda1 :
UUID=d361e038-dbb4-48db-9e65-09342edf7c36 / ext4 defaults 1 1
# Entry for /dev/sda6 :
UUID=447b3452-8e8f-40cb-8b05-8753f3bb2d3f /home ext4 defaults 1 2
none /proc proc defaults 0 0
# Entry for /dev/sda5 :
UUID=cc59aaf8-e77b-4f24-95ea-4e7ee4a9e380 swap swap defaults 0 0
none /dev/pts devpts defaults 0 0
tmpfs /dev/shm tmpfs defaults 0 0


[Old-Polack]

My bad, here is the one from the hard drive:

Code: [Select]

[root@localhost media]# cat etc/fstab
# Entry for /dev/sda1 :
UUID=d361e038-dbb4-48db-9e65-09342edf7c36 / ext4 defaults 1 1
# Entry for /dev/sda6 :
UUID=447b3452-8e8f-40cb-8b05-8753f3bb2d3f /home ext4 defaults 1 2
none /proc proc defaults 0 0
# Entry for /dev/sda5 :
UUID=cc59aaf8-e77b-4f24-95ea-4e7ee4a9e380 swap swap defaults 0 0
none /dev/pts devpts defaults 0 0
tmpfs /dev/shm tmpfs defaults 0 0


That appears to also be correct. When the reboot seems to stall, how long are you waiting before you decide it won't boot, or is the kernel panicking and stopping the process?

[IrishGuy]

It tries to find sda1 and sda5 before going into a kernel panic, however, it does sometimes give me this:

For which I have no explanation whatsoever. 

[Old-Polack]

It tries to find sda1 and sda5 before going into a kernel panic, however, it does sometimes give me this:

For which I have no explanation whatsoever. 

Before changing the MB, was the system running properly? I'm thinking there may be a bad connection to the hard drive in either the power or data cable. If you have spares, try a different data cable. If not try disconnecting from both ends and reconnecting. Also disconnect and reconnect the power cable to the hard drive. The files are there, but the kernel can't seem to read the data.

The other thing to try is a forced fsck on the / partition. From the terminal on the liveCD, with the / partition unmounted, as root;

[root@localhost ~]# umount /dev/sda1                 <Enter>

[root@localhost ~]# fsck -f /dev/sda1                     <Enter>

Post the result of the second command.

-------------------------------------------------------------------------------------

FYI:

Most of my installations are on external hard drives, connected by either eSATA when available, or USB2. I move these from one computer to another, some my own, some belonging to friends. Obviously, different MBs are involved with each move, yet none of the installations fail to boot properly. At worst I get notified that new hardware was found, and I'm then asked if I wish to configure that new hardware. I say yes, the job gets done, and I'm good to go. You should be getting a single, similar experience, from the MB change, and nothing more, assuming all connections are proper, the memory cards have full contact, and the video card is fully seated, if it's an add-in card.

[IrishGuy]

Yup, it was running just fine prior to the MB going bad.

I tried a different data cable (i.e. the old one) with the same results. I also checked the power cable and its OK too. I've never had any problems accessing the HDD from the LiveCD either.

Here is the fsck output:

Code: [Select]

[root@localhost guest]# fsck -f /dev/sda1   
fsck from util-linux-ng 2.16.2
e2fsck 1.41.9 (22-Aug-2009)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/dev/sda1: 183546/512064 files (0.1% non-contiguous), 1417431/2046271 blocks


[Old-Polack]

Yup, it was running just fine prior to the MB going bad.

I tried a different data cable (i.e. the old one) with the same results. I also checked the power cable and its OK too. I've never had any problems accessing the HDD from the LiveCD either.

Here is the fsck output:

Code: [Select]

[root@localhost guest]# fsck -f /dev/sda1   
fsck from util-linux-ng 2.16.2
e2fsck 1.41.9 (22-Aug-2009)
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/dev/sda1: 183546/512064 files (0.1% non-contiguous), 1417431/2046271 blocks


That went well; e2fsck had no problem accessing the partition, and the results are as they should be. Have you tried rebooting to the installed OS after running the fsck? If not, do so.

[IrishGuy]

That went well; e2fsck had no problem accessing the partition, and the results are as they should be. Have you tried rebooting to the installed OS after running the fsck? If not, do so.

I gave that a shot but still got the same bunch of errors as I've gotten before

[Old-Polack]

That went well; e2fsck had no problem accessing the partition, and the results are as they should be. Have you tried rebooting to the installed OS after running the fsck? If not, do so.

I gave that a shot but still got the same bunch of errors as I've gotten before

I think you're at the point of reinstalling, keeping your old /home partition and just installing the OS to the / partition, unless someone else has any other ideas to try. I'm fresh out.

Actually I can think of a number of things to try, but they're all very time consuming, and in the meantime you're without a running OS. A reinstall would only take about 20 minutes, plus a system upgrade and the installation of the specific apps you use. For me, the whole process would take about an hour; your time would vary depending on connection speed and which apps you use. From a practical standpoint, that seems the quickest option.

[IrishGuy]

I think you're at the point of reinstalling, keeping your old /home partition and just installing the OS to the / partition, unless someone else has any other ideas to try. I'm fresh out.

Actually I can think of a number of things to try, but they're all very time consuming, and in the meantime you're without a running OS. A reinstall would only take about 20 minutes, plus a system upgrade and the installation of the specific apps you use. For me, the whole process would take about an hour; your time would vary depending on connection speed and which apps you use. From a practical standpoint, that seems the quickest option.

Well I certainly appreciate all the help and hard work from both yourself and Melodie, especially as this seems to be more of a hardware issue than a software one. Being a loyal user for almost 3 years now, I figured PCLOS wouldn't screw with me in that kind of way

Since a mobo replacement isn't all that common, I'm hoping that if I ever run into a situation like this in the future, I'll be much better prepared