initrd / pivot_root + boot problems

Jeremy Booker (JerMe@nc.rr.com)
Wed, 12 Mar 2003 17:54:02 -0500


Hi,

While looking for support on the egigma-list@redhat.com list, someone
recommended that I ask here.

Here's the deal...

After forgetting my root password and trying to boot into single user mode
under Red Hat 7.2 Pro (boxed retail set) I have run into a kernel panic. In
grub I edited the boot commands to include "kernel /boot/.... single".
Immediately after doing this the system failed to boot with the error
message "kernel panic: no init found. Try passing init= to kernel".
(Previously the machine had run just fine and had been up for over 150
days). After the first failure I tried adding any combination of "single",
"-single", "s", and "-s" I could think of, and changed their position in the
command (first/last parameter, appended "single" to other commands including
initrd line). Nothing could get me into single user mode, I never got past
the kernel panic. I also tried passing init= as a kernel argument with the
path listed on the "initrd" command. I eventually booted to rescue mode from
a RH 7.2 install disk and got my password changed. However, I am still
unable to boot from the hard drives. (In a software RAID 1 config)

My grub.conf file contains the following:
root (hd0,0)
kernel /boot/vmlinuz-2.4.7-10 ro root=/dev/sda1
initrd /boot/initrd-2.4.7-10.img

( I set the file back to exactly how it looked when I started, any changes I
made were always lost after having to reboot)

Also of interest is an error that occurred just before the kernel panic. It
read "pivotroot: pivot_root(/sysroot, /sysroot/initd) failed:2". Does anyone
know what the "2" error code indicates? I cannot find this anywhere in any
documentation of pivot_root. It seems to be a rather obscure function in the
first place. Maybe if someone could shed some light on this?

The machine in question is a single CPU (Intel PIII 800mhz) with 2 SCSI
drives in RAID 1 (mirroring). I'm guessing from the 'kernel' line that my
kernel version is "vmlinuz-2.4.7-10". It is the one that came with RH 7.2
Pro retail box. I haven't upgraded or modified anything.

I'm not entirely sure of my file system layout. (I'm a relative newbie,
unsure about devices and file systems still) I'm using RAID 1 on two SCSI
drives. It was setup by the installer during the install process and I
assume it uses all the defaults. I don't know the layout of fstab. The
machine is remote (co-located in a data center), and I have no way of
copying/pasting or saving that information from the server.

I also took a look in /var/log/boot and all the recent files were blank.
Nothing ever got written to them.

exact problem. However, none of them have any solutions. It was suggested
that this may be due to file system corruption. I doubt this is the case as
I can mount my software RAID 1 drives from rescue mode and read/write all
the files just fine. Everything seems to be intact. I have not run fsck but
I will do that ASAP.

I have gotten my hands on the latest RH 8.0 ISOs. Could upgrading from 7.2
to 8.0 cure my woes? Would this just preserve the hardware/device
misconfiguration?

My problem looks to described in more detail in this post:
https://listman.redhat.com/pipermail/valhalla-list/2003-January/thread.html#
23350
and here: http://lists.ethernal.org/cantlug-0208/msg00472.html
and here:
http://archive.linuxfromscratch.org/mail-archives/lfs-support/2003/02/0018.h
tml
and here: http://www.gollatz.net/troubleshooting

Any help or ideas would be greatly appreciated.

Regards,
Jeremy

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Jeremy Booker - Owner / Webmaster
JTech Web Systems
www.JTechWebSystems.com
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
"Therefore do not worry about tomorrow, for tomorrow will worry about
itself. Each day has enough trouble of its own." -Mathew 5:34

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/