View previous topic :: View next topic |
Author |
Message |
ashaver n00b
Joined: 27 Jul 2003 Posts: 2
|
Posted: Sun Jul 27, 2003 9:31 pm Post subject: after install, reproducible crash on reboot |
|
|
I have just installed gentoo on my P4 system, dual-booting with WinXP Pro. I have two IDE hard drives: hda is 40 gigs, hda1 is the XP NTFS part and hda2-4 are the boot, swap, and root parts respectively. hdb is 120 gigs; hdb1 is a Win95 FAT32 part I plan to use to swap between linux and XP, and hdb2 is a large NTFS part where I keep all my Win data.
Grub is installed on hda2, which is the bootable part; it dual-boots into XP without a problem and starts the gentoo boot seemingly normally.
The gentoo boot always stops at the same place:
Code: | Partition check:
hda: hda1 hda2 hda3 hda4
hdb: kernel BUG at sched.c:1141!
invalid operand: 0000
|
It then proceeds to dump a bunch of what I assume is register data to the screen and then gives
Code: | <0> Kernel panic: Aiee, killing interrupt handler!
In interrupt handler - not syncing
|
...after which the comp locks up.
My fstab has entries for the hda parts but not for the hdb parts. Both of the hdb parts look fine when I check them out with fdisk using the LiveCD and both work acceptably under Windows.
Where have I gone wrong? Or is my error hidden somewhere else and it just becomes visible at this point?
Thanks. |
|
Back to top |
|
|
miunk Apprentice
Joined: 24 Sep 2002 Posts: 199
|
Posted: Mon Jul 28, 2003 1:42 pm Post subject: Chipset support in the kernel |
|
|
Perhaps you did not compile the correct chipset support into the kernel? I think that could cause problems with your IDE disks. What motherboard do you have? If everything seems to work fine on the LiveCD you could try to use the kernel config from there (directions on that in the install docs) and remove only support for things that you are sure you do not have. |
|
Back to top |
|
|
ashaver n00b
Joined: 27 Jul 2003 Posts: 2
|
Posted: Mon Jul 28, 2003 4:15 pm Post subject: |
|
|
Thanks for the suggestion...
I included the correct chipset support in the kernel -- I'm using an Asus p4SDX mobo with the sis 655 chipset, and I included the sis family support in the kernel.
To be sure, I went through and selected all of the IDE chipset options in the kernel and recompiled, but still got the same error.
Also, immediately above the part of the boot process I quoted above, gentoo detected and correctly identified hdb, along with both of my IDE cd-roms.
Is the problem perhaps with my partitions on hdb? I haven't yet tried booting with hdb disconnected, but since that's where all my data is, that's not really a good long-term solution.
Thanks again... |
|
Back to top |
|
|
BradN Advocate
Joined: 19 Apr 2002 Posts: 2391 Location: Wisconsin (USA)
|
Posted: Tue Jul 29, 2003 3:20 am Post subject: |
|
|
One thing to possibly try disabling is preemptable kernel support - I've found it causes some strange problems in general so maybe it's affecting this too.
Otherwise you could try using a different kernel sources (vanilla-sources, etc) |
|
Back to top |
|
|
Stu_28 Tux's lil' helper
Joined: 12 Jun 2003 Posts: 129
|
Posted: Tue Jul 29, 2003 7:40 am Post subject: |
|
|
Wow, I had this same thing happen to me last week, and all the Google-ing didn't help at all! I fixed the problem by adding the kernel option root=/dev/hde3 into grub.conf. So, your grub.conf should look something like this (except yours would have hda4 not hde3):
Code: |
default=0
timeout=30
splashimage=(hd0,0)/grub/splash.xpm.gz
# Default kernel
title Gentoo Linux (2.4.20-r5)
root (hd0,0)
kernel (hd0,0)/bzImage-2.4.20-gentoo-r5 root=/dev/hde3
|
Hope that helps... |
|
Back to top |
|
|
spindizy n00b
Joined: 13 Apr 2003 Posts: 10 Location: nextothecpu
|
Posted: Thu Aug 14, 2003 4:17 am Post subject: ive had something similar to this |
|
|
I get the kernel panic to say
kernel BUG at sched.c:1141! but at a diffrent place than the origional poster.. I first got this dump at the same point he did but then found a problem in my /etc/fstab that allows it to boot until it reaches the IPV4 stuff, and immediatly where ipv6 over ipv4 should load i get that kernel error and it panics and dumps..
Im installing on an IBM 704 server, quad PPro200, 1GB ram, and full 12 scsi disks. The first kernel i compiled dumped out with a similar bug in scsi.c so i took out scsi support, to try and elimnate the problem, but now i get the bug in sched.c
Also with large ammount of googleing for no help, this is the first thread ive come across with the same problem as mine...
As this machine is not the machine im using now, its hard to paste errors, altho i'll retype them if it help, also, im using LILO not grub, and it doesnt seem to make much diffrence which bootloader i use..
I think im going to try the vanilla-sources....... Let you know how that goes. _________________ Eagles may soar but weasels dont get stuck in jet engines. |
|
Back to top |
|
|
spindizy n00b
Joined: 13 Apr 2003 Posts: 10 Location: nextothecpu
|
Posted: Thu Aug 14, 2003 7:57 am Post subject: Re: ive had something similar to this |
|
|
spindizy wrote: | I think im going to try the vanilla-sources....... Let you know how that goes. |
Worked like a charm, first time, scsi and all. Something in the gentoo sources is skrewy. _________________ Eagles may soar but weasels dont get stuck in jet engines. |
|
Back to top |
|
|
|