Kernel panic due to Ctrl+C interrupted dom0 update and reboot

27 views
Skip to first unread message

simon...@gmail.com

unread,
Jan 12, 2019, 3:43:21 PM1/12/19
to qubes-users
Hi,

Error message :

[3.680649] Call Trace:
[3.680707] dump_stack+0x5c/0x85
[3.680764] panic+0xe4/0x252
[3.681184] mount_block_root+0x35b/0x52c
[3.681601] ? do_early_param+0x16e/0x16e
[3.682018] prepare_namespace+0x278/0x2d5
[3.682434] ? do_early_param+0x16e/0x16e
[3.682851] kernel_init_freeable+0x2e8/0x32d
[3.683270] ? rest_init+0xb0/0xb0
[3.683685] kernel_init+0xa/0x101
[3.684101] ret_from_fork+0x35/0x40
[3.684536] Kernel Offset: disabled.

So, I booted into recovery mode prompt from a USB key (Troubleshooting and Rescue a Qubes system (read-write choice).

It seems that I can’t boot an older kernel version.

# ls /boot
initramfs-vmlinuz-4.14.74-1.pvops.qubes.x86_64
#

I don't really know what to do to fix it.

Your help will be greatly appreciated!

Thank you a lot for your valued assistance

awokd

unread,
Jan 12, 2019, 4:17:09 PM1/12/19
to qubes...@googlegroups.com
simon...@gmail.com wrote on 1/12/19 8:43 PM:
> Hi,
>
> Error message :
>
> [3.680649] Call Trace:
> [3.680707] dump_stack+0x5c/0x85
> [3.680764] panic+0xe4/0x252
> [3.681184] mount_block_root+0x35b/0x52c
> [3.681601] ? do_early_param+0x16e/0x16e
> [3.682018] prepare_namespace+0x278/0x2d5
> [3.682434] ? do_early_param+0x16e/0x16e
> [3.682851] kernel_init_freeable+0x2e8/0x32d
> [3.683270] ? rest_init+0xb0/0xb0
> [3.683685] kernel_init+0xa/0x101
> [3.684101] ret_from_fork+0x35/0x40
> [3.684536] Kernel Offset: disabled.
>
> So, I booted into recovery mode prompt from a USB key (Troubleshooting and Rescue a Qubes system (read-write choice).
>
> It seems that I can’t boot an older kernel version.
>
> # ls /boot
> initramfs-vmlinuz-4.14.74-1.pvops.qubes.x86_64
> #

I think you might be looking at the USB key's /boot there. Your fixed
disk should be somewhere under /mnt- it tells you where it's mounting it
but I don't remember exactly. Then look for grub.cfg or xen.cfg under
that and edit it so it boots an older kernel.

simon...@gmail.com

unread,
Jan 12, 2019, 5:31:10 PM1/12/19
to qubes-users
Thanks awokd. The system is mounted under /mnt/sysimage directory.

In /mnt/sysimage/boot/grub2/grub.cfg, I changed :

module /vmlinuz-4.14.74-1.pvops.qubes.x86_64 by module /vmlinuz-4.14.57-2.pvops.qubes.x86_64

but no change, same error when booting :-(

awokd

unread,
Jan 12, 2019, 6:21:30 PM1/12/19
to qubes...@googlegroups.com
simon...@gmail.com wrote on 1/12/19 10:31 PM:
That looks better. Double check /mnt/sysimage/boot/efi/EFI/qubes/xen.cfg
too in case you are booting UEFI maybe. Not sure what else it could be...

simon...@gmail.com

unread,
Jan 13, 2019, 5:30:02 AM1/13/19
to qubes-users
Le samedi 12 janvier 2019 21:43:21 UTC+1, simon...@gmail.com a écrit :
file is empty.

sh-4.3# pwd
sh-4.3# /mnt/sysimage/boot/efi/EFI/qubes
sh-4.3# ls -al xen.cfg
-rw-r--r-- 1 root root 0 Jan 6 11:14 xen.cfg

It seems that the same errors that :

https://github.com/QubesOS/qubes-issues/issues/3915

I don't know how to handle it without reinstalling.

simon...@gmail.com

unread,
Jan 13, 2019, 6:39:57 AM1/13/19
to qubes-users
Le samedi 12 janvier 2019 21:43:21 UTC+1, simon...@gmail.com a écrit :
I tried it without success : https://deniszanin.com/bad-grub-fixing-qubes-boot-missing-on-uefi/

simon...@gmail.com

unread,
Jan 13, 2019, 8:39:28 AM1/13/19
to qubes-users
Le samedi 12 janvier 2019 21:43:21 UTC+1, simon...@gmail.com a écrit :
Fixed :-)

grub.cfg corrupted. Missing two lines :

echo 'Loading initial ramdisk ...'
module --nounzip /initramfs-4.14. 74-1.pvops.qubes.x86_64.img

I found a backup of the grub.cfg file in user's home directory (dated Sept. 26 2018). Fortunately I make backups.

Reboot with an older kernel version and :

sudo dracut /boot/initramfs-4.14.74-1.pvops.qubes.x86_64.img 4.14. 74-1.pvops.qubes.x86_64
sudo grub2-mkconfig -o /boot/grub2/grub.cfg

It is now fully operationnal.
Reply all
Reply to author
Forward
0 new messages