LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Kernel (http://www.linuxquestions.org/questions/linux-kernel-70/)
-   -   Kernel panic error - unable to mount root (http://www.linuxquestions.org/questions/linux-kernel-70/kernel-panic-error-unable-to-mount-root-600151/)

css 11-16-2007 11:02 AM

Kernel panic error - unable to mount root
 
Hello:

I'm completely unfamiliar with Linux but do need some help.
A Nortel BCM switch is running Linux ver 2.4.22-ngcl-10.12.1.0. After programming the BCM, which included a number of reboots and shutdowns, the system will now not boot.

Power was disconnected from the unit for about 15 hours. While watching the boot process, we see the following on the last two lines of output:

kmod: failed to exec /sbin/modprobe -s -k nls_iso8859-1, errno = 2
Kernel panic: VFS: Unable to mount root fs on 03:05

i have no idea what this means, but am hoping someone can provide input.
Thank you in advance...

ken

kav 11-16-2007 03:12 PM

I think that means one of the kernel modules died and as a result could not mount the root file system. But I am not an expert.

jschiwal 11-16-2007 03:18 PM

That error usually indicates a media error. For a router, that could be the memory or flash memory.

b0uncer 11-16-2007 03:21 PM

I usually bump into that panic message after an installation of a Linux operating system where root partition (/) is formatted to something that the kernel doesn't have built-in support for, and there is no initrd for it. An example is to install Slackware and format root partition to ext3, and try to use a stock kernel without initrd - the kernel doesn't know how to mount ext3 partition, and therefore panics.

So make sure you know what filesystem your root partition has, that your kernel supports that (and if it's not built directly into the kernel, but as a module, you'll need to make initrd too to be able to mount root filesystem before the fs module is loaded that enables the kernel to mount the real root partition) and that your bootloader settings are correct.

Willy Gommel 11-16-2007 05:29 PM

Modprobe is a very important central component of your OS. It sounds as though either /sbin is not where the kernel expects it to be -- in / -- or possibly as though modprobe itself has gotten moved, renamed, or deleted.

Tell me: was there any mention of /etc/fstab in your configuration of the switch? If so, I'm betting that it has redefined the partition used for mounting the root filesystem [/]. Of course, since /etc is in the root filesystem, the moment the wrong partition gets mounted as root, /etc/fstab ceases to be available as such.

As to the last line, that too contains some suggestive information. "Unable to mount root filesystem on 03:05" ... what is your device #03:05? If it's a CD-ROM, a floppy drive, or a HDD partition other than the one your boot loader contains as being the root filesystem, then of course the answer is obvious. Even more so if it turns out that it's not a data storage device at all!

reddazz 11-17-2007 05:45 AM

I've merged two similar threads.

tmick 12-03-2007 06:55 PM

Quote:

Originally Posted by css (Post 2961120)
Hello:

I'm completely unfamiliar with Linux but do need some help.
A Nortel BCM switch is running Linux ver 2.4.22-ngcl-10.12.1.0. After programming the BCM, which included a number of reboots and shutdowns, the system will now not boot.

Power was disconnected from the unit for about 15 hours. While watching the boot process, we see the following on the last two lines of output:

kmod: failed to exec /sbin/modprobe -s -k nls_iso8859-1, errno = 2
Kernel panic: VFS: Unable to mount root fs on 03:05

i have no idea what this means, but am hoping someone can provide input.
Thank you in advance...

ken

kmod is a Kernel Module and /sbin/modprobe is what autolaunches the kernel module. VFS should be your virtual files system.
Judging by what I see nls_iso8859-1 is a boot image. Try running fsck on the boot disk/partition and see if it recovers it, if not re-install the OS.


All times are GMT -5. The time now is 12:33 PM.