[SOLVED] No bootup after slackpkg update with kernel upgrade

sattvakarma · 01-25-2017, 12:49 AM

Hello,

I ran slackpkg update today on my PC and after the update which also included a kernel upgrade among other things, the PC has stopped booting up with an error msg:

Quote:

Loading Linux EBDA is big; Kernel setup stack overlaps LILO second stage

System is not responsive after this...i just have to hard reset my PC after that.

PC is dual boot with Windows, and this msg i am posting from my Windows setup.

Tips/assistance to resolve this problem will be appreciated!

Regards,
Sattva

allend · 01-25-2017, 01:15 AM

As you have not given details of your Slackware version, whether you are running the -huge kernel or the -generic kernel with an initrd, whether you are actually using the LILO bootloader, what your /etc/slackpkg/blacklist contains, whether you ran lilo after the upgrade, whether you needed to rebuild your mkinitrd, then I will guess that you failed to run lilo.
For recovery, see http://docs.slackware.com/howtos:sla...oot_from_media

sattvakarma · 01-25-2017, 01:35 AM

Allend,

Thank you for your response!

The details for my PC setup are:

- Slackware 14.2 x64
- Generic kernel with initrd
- LILO is used
- /etc/slackpkg/blacklist has never been modified (never saw it actually)
- did run lilo as suggested by slackpkg update

I understand that after a kernel upgrade, i use the huge kernel to boot up and work my way to a generic version for the new kernel??

I also used my Slackware DVD to boot up my system and run

Quote:

lilo -v

Still, its the same situation as in my original post.

aragorn2101 · 01-25-2017, 02:47 AM

Hi,

Maybe lilo was not run after the kernel upgrade.

I see you used a Slackware DVD to boot, but did you chroot into your system before running "lilo -v"?

If not, then boot into a Slackware DVD again and follow with these:
(assuming your Slackware install in at /dev/sda3 with ext4 filesystem)

Code:

mkdir /mnt/Slackware
mount -t ext4 /dev/sda3 /mnt/Slackware

mount -t proc /proc /mnt/Slackware/proc
mount --rbind /sys /mnt/Slackware/sys
mount --rbind /dev /mnt/Slackware/dev
mount --rbind /run /mnt/Slackware/run
chroot /mnt/Slackware env -i HOME=/root TERM=$TERM PS1='\u@\h:\w# ' PATH=/usr/bin:/usr/sbin:/bin:/sbin bash --login +h

lilo -v

Replace with your system's details.

All the best.

allend · 01-25-2017, 03:32 AM

Quote:

I understand that after a kernel upgrade, i use the huge kernel to boot up and work my way to a generic version for the new kernel??

When running the generic kernel with an initrd, after upgrading the kernel using slackpkg, you need to build a new initrd. Then you run lilo before rebooting.

Personally, I have this in my /etc/slackpkg/blacklist

Code:

# Automated upgrade of kernel packages aren't a good idea (and you need to
# run "lilo" after upgrade). If you think the same, uncomment the lines
# below 
#
#kernel-firmware
#kernel-generic
kernel-generic-smp
#kernel-headers
kernel-huge
kernel-huge-smp
#kernel-modules
kernel-modules-smp
#kernel-source

so that only the generic kernel is upgraded.

I also use a custom function so that I am prompted to build a new initrd when required.

sattvakarma · 01-25-2017, 11:21 PM

aragorn2101,

My Slackware installation partition setup:

/dev/sda5 is /boot
/dev/sda6 is /
/dev/sda7 is swap
/dev/sda8 is /home

Is there a change in the partition mounts to the suggestion that you have provided? Kindly provide a detailed post if there is one.

When slackpkg was updating, it did prompt for lilo and i went ahead with it followed by a re-boot which resulted in the error msg and my system not booting.

I then booted using the Slackware DVD and did NOT chroot into my system to run lilo -v. Just logged in as root and ran lilo -v from the prompt.

aragorn2101 · 01-25-2017, 11:28 PM

Quote:

Originally Posted by sattvakarma

My Slackware installation partition setup:

/dev/sda5 is /boot
/dev/sda6 is /
/dev/sda7 is swap
/dev/sda8 is /home

Is there a change in the partition mounts to the suggestion that you have provided? Kindly provide a detailed post if there is one.

Thank you for providing the details; we can help you better this way. Yes, a little detail will change as your /boot is on a different drive. Here's what you should do:

Boot with the Slackware DVD and then,

Code:

mkdir /mnt/Slackware
mount -t ext4 /dev/sda6 /mnt/Slackware
mount -t ext4 /dev/sda5 /mnt/Slackware/boot
mount -t proc /proc /mnt/Slackware/proc
mount --rbind /sys /mnt/Slackware/sys
mount --rbind /dev /mnt/Slackware/dev
mount --rbind /run /mnt/Slackware/run
chroot /mnt/Slackware env -i HOME=/root TERM=$TERM PS1='\u@\h:\w# ' PATH=/usr/bin:/usr/sbin:/bin:/sbin bash --login +h

lilo -v

In the above I assumed /dev/sda5 has an ext4 filesystem, you can change it if it is not the case.

sattvakarma · 01-26-2017, 12:34 AM

aragorn2101,

Thank you so much for the assistance provided. It is very much appreciated!

I followed the steps outlined by you (including changes to the filesystem type for /boot as it is ext2) and ran /sbin/lilo -v and i was still unable to boot up the system. Same error msg.

I then edited /etc/lilo.conf and removed the entry for the generic kernel (there was a warning every time for its existence) and ran /sbin/lilo -v

Now i am able to boot into the system :-)

I will now work out the procedure for a generic kernel.

Though i still fail to understand why this problem occurred in the first place? I have updated the system earlier also with no problems.

sattvakarma · 01-26-2017, 01:25 AM

Thank you for your suggestion allend.

I will read further and put it to good use!

aragorn2101 · 01-26-2017, 02:34 AM

Great to know you reached a solution. We are very happy to help.

Good Slacking.

iNC0GNi70 · 03-05-2018, 02:06 PM

Quote:

Originally Posted by aragorn2101

Hi,

Maybe lilo was not run after the kernel upgrade.

I see you used a Slackware DVD to boot, but did you chroot into your system before running "lilo -v"?

If not, then boot into a Slackware DVD again and follow with these:
(assuming your Slackware install in at /dev/sda3 with ext4 filesystem)

Code:

mkdir /mnt/Slackware
mount -t ext4 /dev/sda3 /mnt/Slackware

mount -t proc /proc /mnt/Slackware/proc
mount --rbind /sys /mnt/Slackware/sys
mount --rbind /dev /mnt/Slackware/dev
mount --rbind /run /mnt/Slackware/run
chroot /mnt/Slackware env -i HOME=/root TERM=$TERM PS1='\u@\h:\w# ' PATH=/usr/bin:/usr/sbin:/bin:/sbin bash --login +h

lilo -v

Replace with your system's details.

All the best.

Hi, I happen to have the same problem, I have followed the instructions in your post, changing the command for sda4, but it gives me a segmentation fault when I run the chroot line, any idea on what may cause it ?
Also I remember I have added a security line to a file some time ago to prevent direct login as root, so maybe it is what that causes that issue ? I'm pretty desperate on this one.
Heeeelp !
14.2 after an update

FlinchX · 03-06-2018, 12:09 AM

Quote:

Originally Posted by allend

whether you ran lilo after the upgrade

Is that message (the one slackpkg displays after updating kernel packages) a legacy one (since AFAIK in UEFI mode one has to deal with elilo instead)?

iNC0GNi70 · 03-06-2018, 12:53 AM

I dunno. I'm still using BIOS on a 2010 laptop.

cesarsj · 09-04-2019, 11:56 AM

Quote:

Originally Posted by aragorn2101

Hi,

Maybe lilo was not run after the kernel upgrade.

I see you used a Slackware DVD to boot, but did you chroot into your system before running "lilo -v"?

If not, then boot into a Slackware DVD again and follow with these:
(assuming your Slackware install in at /dev/sda3 with ext4 filesystem)

Code:

mkdir /mnt/Slackware
mount -t ext4 /dev/sda3 /mnt/Slackware

mount -t proc /proc /mnt/Slackware/proc
mount --rbind /sys /mnt/Slackware/sys
mount --rbind /dev /mnt/Slackware/dev
mount --rbind /run /mnt/Slackware/run
chroot /mnt/Slackware env -i HOME=/root TERM=$TERM PS1='\u@\h:\w# ' PATH=/usr/bin:/usr/sbin:/bin:/sbin bash --login +h

lilo -v

Replace with your system's details.

All the best.

I followed these steps, but upon restarting the messages came up:

[ end kernel panic - not syncing: No working init found. Try passing init= option to kernel.
hrtime: interrupt took
random: nonbloking pool is initialized