kernel panic

Les Mikesell lesmikesell at gmail.com
Wed Sep 5 22:31:19 UTC 2007


Karl Larsen wrote:
> Les Mikesell wrote:
>> Karl Larsen wrote:
>>> Les Mikesell wrote:
>>>> Karl Larsen wrote:
>>>>>    I found this on Google but it doesn't help me much:
>>>>>
>>>>> echo "Loading scsi_mod module"
>>>>> insmod /lib/scsi_mod.o echo "Loading sd_mod module"
>>>>> insmod /lib/sd_mod.o echo "Loading cpqarray module"
>>>>> insmod /lib/cpqarray.o echo "Loading jbd module"
>>>>> insmod /lib/jbd.o echo "Loading ext3 module"
>>>>> insmod /lib/ext3.o mount -t proc /proc /proc
>>>>> echo Mounting /proc filesystem
>>>>> echo Creating root device
>>>>> mkrootdev /dev/root
>>>>>
>>>>> This seems to be where /dev/root comes from. And my kernel can't 
>>>>> find it for some reason.
>>>>>
>>>>>
>>>>>
>>>>> echo 0x0100 > /proc/sys/kernel/real-root-dev
>>>>> umount /proc
>>>>> echo Mounting root filesystem
>>>>> mount --ro -t ext3 /dev/root /sysroot
>>>>> pivot_root /sysroot /sysroot/initrd
>>>>
>>>> There is a point in the boot sequence where the bios-loaded kernel 
>>>> trades the initrd RAM disk image (also bios loaded) for the real 
>>>> root partion mount point.  Several things can go wrong here. Grub 
>>>> may have told the kernel to look in the wrong place for the root 
>>>> file system, the file system might be unreadable, or you may have 
>>>> moved the system onto a machine with a different type of disk 
>>>> controller that needs a different driver module included on the 
>>>> initrd.  The main reason you need the initrd is to load drivers for 
>>>> the root filesystem if they aren't compiled into the kernel but it 
>>>> will only include ones for the the machine where the system was 
>>>> originally installed. If you can access the disk, somewhere in the 
>>>> boot messages you should see it detecting the device and 
>>>> partitions.  If you don't see that, the kernel can't see the disk 
>>>> and you'll have to rebuild the initrd with the right module.
>>>>
>>>    Initrd has to be the problem Les. I have had the old hard drive on 
>>> this new box and I was surprised that FC6 would boot up but F7 did 
>>> what it has done all along. Now I did load a new F7 on the old hard 
>>> drive and it booted up fine. So this info backs up my belief that 
>>> initrd is the problem.
>>>
>>>    I will now try and find out HowTo rebuild initrd from a Rescue cd :-)
>>
>> If your rescue boot mounts the system drives and suggests a chroot 
>> command, do that, then add the right alias command to 
>> /etc/modprobe.conf (will depend on hardware) and then run mkinitrd.
>>
>    What is an alias command? How do I find it with a rescue cd?

If you have a working install for this disk controller, look in 
/etc/modprobe.conf to see what the line should be.  Here's a sample 
modprobe.conf from a machine with an intel NIC and 2 different scsi 
controllers:
# cat modprobe.conf
alias eth0 e1000
alias scsi_hostadapter mptscsibase
alias scsi_hostadapter1 mptscsih
alias scsi_hostadapter2 aic7xxx

But the modules need to match your exact hardware.  If you have a normal 
  old IDE controller, you shouldn't need this to boot.

If you boot the install CD/DVD with 'linux rescue' at the command prompt 
and your fstab file is correct, it should mount the system partitions 
for you and suggest a chroot command for troubleshooting.  After you do 
the chroot, you'll have all the installed system facilities at your 
disposal to edit the /etc/modprobe.conf file and run mkinitrd, including 
the man pages.  When you are done, type exit twice to unmount cleanly 
and reboot (once for the chroot shell, once for the rescue shell).  If 
the rescue mode boot doesn't mount the partitions for you, then you have 
some other problem to fix first.

-- 
   Les Mikesell
     lesmikesell at gmail.com




More information about the fedora-list mailing list