Unable to boot from multimedia kernel

[emh]

I'm attempting to use ther multimedia kernel 2.6.17.13-3 on Mandriva 2007.

 

I get the following message:

 

"BIG FAT WARNING: failed to translate "/dev/hde6" into a device ID (note: /dev/hde6 is my swap partition)

 

"If you want to use the current suspend image, reboot and try again with the same kernel that you suspended from. If you want to forget that image, continue and the image will be erased."

 

It gives me the option to continue booting or to reboot. Unfortunately, continuing to boot ends with a kernel panic, so I have to reboot and use the default kernel.

 

I really need the multimedia kernel, because I do occasional music creation with my computer. Anybody that knows how to fix this would be greatly appreciated. Thank you.

 

 

[moved from Software by spinynorman]

[Steve Scrimpshire]

Where does the kernel panic happen? Immediately after continuing? What is the last line you see?

[emh]

When I select the option to continue, these are the messages that pop up:

 

Suspend2 2.2.7.3: missing or invalid storage location (resume2= parameter) Please correct and rerun LILO (or equivalent) before suspending.

 

Mounting root filesystem /dev/root with flags notail

mount: error 6 mounting reiserfs flags notail

well, retrying without the option flags

mount: error 6 mounting reiserfs

well, retrying read-only without any flag

mount: error 6 mounting reiserfs

Switching to new root

ERROR opening /dev/console!!!!: 2

unmounting old /proc

unmounting old /sys

switchroot: mount failed: 22

Initrd finished

Kernel Panic - not syncing: Attempted to kill init!!

[ianw1974]

The resume parameter in /etc/lilo.conf *must* be pointing to a swap partition? Does it?

 

Post your /etc/fstab, and also the output from:

 

fdisk -l

 

so we can see the partition table. That's a lowercase L.

[emh]

Here is my /etc/lilo.conf:

 

# File generated by DrakX/drakboot

# WARNING: do not forget to run lilo after modifying this file

 

default="desktop_2.6.17.13-mm-3"

boot=/dev/hde

prompt

nowarn

timeout=100

message=/boot/message

image=/boot/vmlinuz

label="linux"

root=/dev/hde1

initrd=/boot/initrd.img

append="resume=/dev/hde6 splash=silent"

vga=788

image=/boot/vmlinuz

label="linux-nonfb"

root=/dev/hde1

initrd=/boot/initrd.img

append="resume=/dev/hde6"

image=/boot/vmlinuz

label="failsafe"

root=/dev/hde1

initrd=/boot/initrd.img

append="failsafe resume=/dev/hde6"

image=/boot/vmlinuz-2.6.17.13-mm-desktop-3mdv

label="desktop_2.6.17.13-mm-3"

root=/dev/hde1

initrd=/boot/initrd-2.6.17.13-mm-desktop-3mdv.img

append="resume=/dev/hde6 splash=silent resume2=swap:/dev/hde6"

vga=788

 

Here is my /etc/fstab:

 

/dev/hde1 / reiserfs notail 1 1

/dev/hdh /mnt/cdrom auto umask=0022,users,iocharset=utf8,noauto,ro,exec 0 0

/dev/sdb1 /mnt/data reiserfs user,defaults 1 2

none /mnt/floppy supermount dev=/dev/fd0,fs=ext2:vfat,--,umask=0022,iocharset=utf8,sync 0 0

none /proc proc defaults 0 0

/dev/hde6 swap swap defaults 0 0

 

My output of "fdisk -l"

 

Disk /dev/hde: 41.1 GB, 41110142976 bytes

255 heads, 63 sectors/track, 4998 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Device Boot Start End Blocks Id System

/dev/hde1 * 1 3571 28684026 83 Linux

/dev/hde2 3572 4998 11462377+ 5 Extended

/dev/hde5 3572 4851 10281568+ 83 Linux

/dev/hde6 4852 4998 1180746 82 Linux swap / Solaris

 

Disk /dev/sdb: 82.3 GB, 82348277760 bytes

255 heads, 63 sectors/track, 10011 cylinders

Units = cylinders of 16065 * 512 = 8225280 bytes

 

Device Boot Start End Blocks Id System

/dev/sdb1 * 1 10011 80413326 83 Linux

 

Thanks in advance for any help.

[Steve Scrimpshire]

/dev/hde1 / reiserfs notail 1 1

/dev/hdh /mnt/cdrom auto umask=0022,users,iocharset=utf8,noauto,ro,exec 0 0

/dev/sdb1 /mnt/data reiserfs user,defaults 1 2

none /mnt/floppy supermount dev=/dev/fd0,fs=ext2:vfat,--,umask=0022,iocharset=utf8,sync 0 0

none /proc proc defaults 0 0

/dev/hde6 swap swap defaults 0 0

 

Device Boot Start End Blocks Id System

/dev/hde1 * 1 3571 28684026 83 Linux

/dev/hde2 3572 4998 11462377+ 5 Extended

/dev/hde5 3572 4851 10281568+ 83 Linux

/dev/hde6 4852 4998 1180746 82 Linux swap / Solaris

 

I'm a little confused by why your HD is hde, but I think the real problem is that /dev/hde5 (your /home partition?) is not in your fstab.

Edited October 25, 2006 by Steve Scrimpshire

[emh]

I'm a little confused by why your HD is hde, but I think the real problem is that /dev/hde5 (your /home partition?) is not in your fstab.

 

My hard drive has been "hde" since I got this computer, back with Mandrake 9.1.

 

My /dev/hde5 is actually a separate partition that I use in case I wanted to experiment with other distros or install a separate version of Mandriva to do something potentially dangerous, so that I don't screw up my main installation. I don't actually have a separate /home partition

 

I'll try adding /dev/hde5 to my fstab, but it doesn't explain why my regular kernel boots just fine while my multimedia kernel doesn't. As far as I know, they use the same fstab (although I could be wrong).

[pmpatrick]

Given the designation of your hard drive as hde, I assume you have an addon pci ide controller card that your hard drive connects to. The first drive connected to a pci ide controller is usually hde with the onboard drives taking up hda thru hdd.

 

Those kernel messages indicate that hde is not being detected at all, i.e. it can't find the swap partition on hde6 and can't find the root partition on hde1. This could indicate some kernel problem with that addon ide controller card, most likely an oversight in failing to compile in support for the card. Please post what, if any, ide controller card you are using and your general hardware setup(motherboard, processor, ide and sata drive configuration). With your good kernel, run as root and post the output of:

 

# lspci

 

You could also try passing some of the usual boot parameters with the problem kernel by hitting the Esc key when you see the lilo boot selection screen which will take you to a boot prompt. At the prompt run:

 

desktop_2.6.17.13-mm-3 noapic nolapic acpi=off

 

Those are the first boot parameters tried when having boot problems as acpi can cause all kinds of trouble with certain hardware/kernel combinations.

Edited October 25, 2006 by pmpatrick

[ianw1974]

initrd=/boot/initrd-2.6.17.13-mm-desktop-3mdv.img

append="resume=/dev/hde6 splash=silent resume2=swap:/dev/hde6"

 

This is why, you have two resumes in your append line, remove the resume2 line and then rerun lilo to rewrite lilo to the mbr. Just type:

 

lilo

 

and it will take care of it. Boot your first CD, and choose rescue mode, then mount all partitions and exit to command prompt, then type:

 

chroot /mnt /bin/bash
source /etc/profile

 

then run the lilo command after editing /etc/lilo.conf

 

I think /dev/hde1 is your root partition, if that is what you're fstab says, /dev/hde5 could be your home partition, but we don't know as you didn't say what you installed where. If you could let us know, that will help with the configuring of your /etc/fstab file.

 

But see how you get on after this.

[emh]

This is why, you have two resumes in your append line, remove the resume2 line and then rerun lilo to rewrite lilo to the mbr. Just type:

 

I tried that, still the same kernel errors, except it also tells me that the extra resume2 line is necessary and to put it in.

 

Also, my /dev/hde5 partition is just a junk partition that I use to mess around with. I didn't get around to having it set to be mounted every time, but I went ahead and edited fstab to mount /dev/hde5 every time, but that didn't fix the problem with my multimedia kernel.

 

Given the designation of your hard drive as hde, I assume you have an addon pci ide controller card that your hard drive connects to. The first drive connected to a pci ide controller is usually hde with the onboard drives taking up hda thru hdd.

 

Actually, I don't. The main hard drive (hde) is connected to the primary IDE channel.

 

Output of lspci:

 

00:00.0 Host bridge: VIA Technologies, Inc. VT8366/A/7 [Apollo KT266/A/333]

00:01.0 PCI bridge: VIA Technologies, Inc. VT8366/A/7 [Apollo KT266/A/333 AGP]

00:09.0 SCSI storage controller: Adaptec AHA-2940U/UW/D / AIC-7881U

00:0a.0 RAID bus controller: Silicon Image, Inc. SiI 3112 [sATALink/SATARaid] Serial ATA Controller (rev 02)

00:0c.0 Multimedia audio controller: Creative Labs SB Live! EMU10k1 (rev 0a)

00:0c.1 Input device controller: Creative Labs SB Live! Game Port (rev 0a)

00:0d.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)

00:0e.0 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 50)

00:0e.1 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 50)

00:0e.2 USB Controller: VIA Technologies, Inc. USB 2.0 (rev 51)

00:0f.0 RAID bus controller: Triones Technologies, Inc. HPT366/368/370/370A/372/372N (rev 05)

00:11.0 ISA bridge: VIA Technologies, Inc. VT8233A ISA Bridge

00:11.1 IDE interface: VIA Technologies, Inc. VT82C586A/B/VT82C686/A/B/VT823x/A/C PIPC Bus Master IDE (rev 06)

00:11.2 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 23)

00:11.3 USB Controller: VIA Technologies, Inc. VT82xxxxx UHCI USB 1.1 Controller (rev 23)

01:00.0 VGA compatible controller: nVidia Corporation NV11 [GeForce2 MX/MX 400] (rev b2)

 

The PCI cards in my system are a SCSI controller, a SATA controller (/dev/sdb), and aa Soundblaster Live Card.

 

I'll also try booting with parameters suggested earlier another day. I'm tired of messing with this tonight.

 

P.S. disabling acpi in the boot menu of MCC had no effect on the error.

Edited October 26, 2006 by emh

[pmpatrick]

I've never run into a situation where the onboard primary ide master drive was designated hde; that is very weird. However, I still believe it is a kernel issue with the ide controller. My guess is the new kernel sees that drive as hda instead of hde. Have you ever seen the drive designated hda with any live cd or any other distro you might have tried? Also, how is the drive listed in the bios setup?

Edited October 26, 2006 by pmpatrick

[emh]

While I'm not at home right now to check for sure, I'm fairly certain that when I use Knoppix Live-CD, it also sees the primary hard drive as hde. I do know that every version of Mandrake/Mandriva I've used since I've had that computer has always seen my hard drive as hde, and I've never had a problem with a multimedia kernel booting before.

[pmpatrick]

Take a look in /proc/ide. I'm curious as to whether you have an ide0 and ide1 in there. You should also have an ide2 where you will find hde inside and an ide3 where you will find hdd.

 

It might be your Silicon Image SATA controller causing the reassignment of hda to hde. It's the only thing that I can think of. If the kernels you have used consistently pick up the drive as hde, that's probably just the way it is, but that is very unusual. I'm fairly certain that your hde drive is not being picked up by the mm kernel; the error messages you are getting are consistent with that theory.

 

Have you tried passing any other boot options with the mm kernel?

Edited October 26, 2006 by pmpatrick

[scoonma]

Do you actually need suspend/resume functions from the kernel? If your main purpose is running the multimedia kernel, I'd just (temporarily) disable these options from lilo. You could simply make a backup and try a different lilo.conf without them.

 

I'm not sure how suspend/resume is working internally. Maybe the function tries to boot from the first visible partion (which is not hdeX, but residing within hdaX-hddX space)? Are you able to insert a small /swap partition in the "lower" IDE area and go on with that one? You can use more than one /swap part.

[emh]

I finally got booted into the multimedia kernel. But it was a bit of a roundabout way of getting there...

 

I had a theory that it was my onboard RAID controller that was taking up hda, hdb, hdc, and hdd. But I didn't have any drives plugged into it (they are labeled "IDE3" and IDE4 on my motherboard). I plugged into what I thought was my primary and secondary IDE controllers (IDE1 and IDE2). I guess for some reason, any linux kept assuming that the RAID controllers were the first drives, and the regular IDE controllers were the second ones, hence, why my main hard drive, plugged into IDE1, was being read as hde instead of hda.

 

I'm guessing this particular multimedia kernel was, for some reason, hard-wired into looking specifically for hda, and would complain when it couldn't find it.

 

So I went into my BIOS and disabled my onboard RAID controller. This wasn't without consequences, though. My system wouldn't boot at all after that. I ended up running the Mandriva install routine, but I picked the "upgrade" option, which doesn't erase anything, but creates correct configuration files.

 

So all was well when I booted into the regular kernel, then I booted into the multimedia kernel, and it had no problems at all.

 

Probably a small oversight in compiling the kernel, but not really a problem most people would face, so it was probably overlooked.