New kernel install had issues. (6.0.5-200.fc36.x86_64)

[Doug H.]

I do dnf upgrade any morning where something shows up as new. Today was
the big kernel jump to 6.x kernel. Was not worried I guess since I just
ran the update like normal and rebooted to it.

Ended up at:

error: ../../grub-core/kern/dl.c:429:symbol `grub_debug_is_enabled' not
found.
Entering rescue mode...
gruub rescue> _


Booted to the F36 install media and did the rescue option. It found my
system and setup my chroot. I simply ran:

chroot /mnt/sysroot
grub2-install /dev/sda
(no errors)
exit

And rebooted to that. All seems great.

Note that this is a very "old" system, having been dnf upgraded from
F19 all the way to today. I notice that the grub menu looks a bit
different, so it could be an issue of having an old grub on the disk
that needed an update.

Also note that I am pretty sure I have not rebooted since doing a disk
clone from my laptop (not this system) where I mounted old and new
laptop on this system and did a variety of dd, rsync, grub2-install,
reset UUIDs, etc. If I accidentally touch my sda instead of the sdd and
sde that I was working on then I might have created a bomb that went
off at reboot. I don't *think* that happened but not 100% sure.

Everything seems to be working, so I don't need any help, but wanted to
report this just in case.

-- 
Doug H.

[Tom Horsley]

On Sun, 30 Oct 2022 09:54:28 -0700
Doug H. wrote:

>  Everything seems to be working, so I don't need any help, but
> wanted to
>  report this just in case. 
As an additional report, I did a dnf update this morning, got the same
kernel, and had no problems at all. Must be one of those things mankind
was not meant to know :-).

[Go Canes]

On Sun, Oct 30, 2022 at 1:00 PM Tom Horsley <horsley1953(a)gmail.com&gt; wrote:
>  On Sun, 30 Oct 2022 09:54:28 -0700
>  Doug H. wrote:
>
>  > Everything seems to be working, so I don't need any help, but wanted to
>  > report this just in case.
>
>  As an additional report, I did a dnf update this morning, got the same
>  kernel, and had no problems at all. Must be one of those things mankind
>  was not meant to know :-). 
And an additional report - updated 7 systems this morning - no
apparent issues so far.

[Leander Hutton]

On 10/30/22 13:00, Tom Horsley wrote:
>  As an additional report, I did a dnf update this morning, got the
> same
>  kernel, and had no problems at all. Must be one of those things mankind
>  was not meant to know 
Another two reports: one on a Ryzen TR 2950X/6900XT with no issues, the second is a 12th
Gen Intel 12900K mobile (Dell XPS 9720) that has been hard locking since 6.0 came out on
Fedora workstation. Pretty normal default install, I am running in X11 to placate the
nVidia driver gods but that's about it. UEFI install as well, BIOS support turned to
disabled.

Granted I've had a lot of issues out of 12th Gen Intel mobile and Linux in general,
particularly with audio and the Intel Xe graphics but the hard lock requiring the power
button press is a new one. I'll see if I can figure out what triggers it.

Leander

[Doug Herr]

On Sun, Oct 30, 2022, at 9:54 AM, Doug H. wrote:
>  I do dnf upgrade any morning where something shows up as new. Today
> was
>  the big kernel jump to 6.x kernel. Was not worried I guess since I just
>  ran the update like normal and rebooted to it.
>
>  Ended up at:
>
>  error: ../../grub-core/kern/dl.c:429:symbol `grub_debug_is_enabled' not
>  found.
>  Entering rescue mode...
>  gruub rescue> _ 
[snip]

>  Note that this is a very "old" system, having been dnf
> upgraded from
>  F19 all the way to today. I notice that the grub menu looks a bit
>  different, so it could be an issue of having an old grub on the disk
>  that needed an update. 
Posting the above got me thinking. I have a 3 drive raided system, so I can boot from any
of sda, sdb or sdc.

I ran a grub2-install on sdb and rebooted, stopping at bios to reboot from sdc and then
sdb. Yup, each drive needed a new grub install or else it had that issue. So I am 99% sure
that this is just a case of me not having updated my grub version on disk for so long.

Let this be a reminder to people upgrading over many version of Fedora. It does not hurt
to update your grub boot sector every now and then. Maybe worth doing after each major
Fedora version update.

P.S. Thanks to Tom for confirming that this is rare.

[Samuel Sieb]

On 10/30/22 10:06, Doug Herr wrote:
>  Let this be a reminder to people upgrading over many version of
> Fedora. It does not hurt to update your grub boot sector every now and then. Maybe worth
> doing after each major Fedora version update. 
This only applies if you are still using legacy BIOS boot, which is 
quite rare now.

[Stephen Morris]

On 31/10/22 09:14, Samuel Sieb wrote:
>  On 10/30/22 10:06, Doug Herr wrote:
> > Let this be a reminder to people upgrading over many version of 
> > Fedora. It does not hurt to update your grub boot sector every now 
> > and then. Maybe worth doing after each major Fedora version update.
>
>  This only applies if you are still using legacy BIOS boot, which is 
>  quite rare now. 
This might be common place is Fedora is installed in a VM. I was
running 
that way until recently via Vmware Player and it did not support UEFI. I 
hadn't tried the processing after support was added for Win 11 as I 
upgraded my motherboard, cpu, memory, graphics card and power supply and 
moved away from the raid setup I was using. I was only using a VM 
because Fedora Workstation didn't support raid and the motherboard bios 
only provided Windows drivers.

Just relative to the other thread I had with the nouveau driver not 
being blacklisted in the kernel parameters in /etc/default/grub, could 
that be a change in the functionality in the 6.05 kernel which I got 
either yesterday or the day before, as the reinstall processing I was 
doing to get the nvidia driver signed for secureboot, was being done 
after the upgrade to the 6.05 kernel?

regards,
Steve
>  _______________________________________________
>  users mailing list -- users(a)lists.fedoraproject.org
>  To unsubscribe send an email to users-leave(a)lists.fedoraproject.org
>  Fedora Code of Conduct: 
>  https://docs.fedoraproject.org/en-US/project/code-of-conduct/
>  List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>  List Archives: 
>  https://lists.fedoraproject.org/archives/list/[e-mail]
>  Do not reply to spam, report it: 
>  https://pagure.io/fedora-infrastructure/new_issue

[Richard Shaw]

Just a "me too" here, but I ran 6.0 fine when there was a call for testing.
When I updated to 6.0.5 my computer (B550 w/ Ryzen 5 5700X) paused at the
BIOS post for a long time and then kind of tried to load but stalled.

I forced it off and turned off the PS for 10 seconds and rebooted, then it
seemed to come up fine. I didn't think anything of it at the time..

Thanks,
Richard

[Tom Horsley]

Slight weirdness for me too. Probably the first time since I installed
6.0.5 I tried to play a youtube video and got no sound (HDMI via
rpmfusion nvidia driver).

I tried to reboot to go back one kernel version and see if sound worked
there, and the reboot blanked the screen, but then was stuck. I eventually
power cycled the computer and it came back up OK and sound did work on
the previous kernel.

At this point I rebooted (which worked) and went back to 6.0.5 and sound
was working fine again.

So something about a complete power down seems to have made it better.

[Tom Horsley]

On Mon, 31 Oct 2022 10:26:16 -0400
Tom Horsley wrote:

>  So something about a complete power down seems to have made it
> better. 
And I tried rebooting again, and that is working now as well. So the
power cycle does seem to have made it all better. I guess I'll find
out in a few days if there is some creeping corruption that screws
up the system over time.

[Stephen Morris]

On 1/11/22 01:45, Tom Horsley wrote:
>  On Mon, 31 Oct 2022 10:26:16 -0400
>  Tom Horsley wrote:
>
> > So something about a complete power down seems to have made it better.
>  And I tried rebooting again, and that is working now as well. So the
>  power cycle does seem to have made it all better. I guess I'll find
>  out in a few days if there is some creeping corruption that screws
>  up the system over time. 
Just following on from my previous thread, ignoring the
grub issues and 
the secureboot signing issue I had with the nvidia drivers, and the fact 
that on the 6.0.5 kernel the boot did a fail over to the nouveau driver 
when it couldn't load the nvidia driver even though the nouveau driver 
was blacklisted in the kernel parameters (albeit with the blacklisting 
occurring before the "quiet rhgb" kernel parameters because that is 
where a fresh install of F36 from a live CD placed them with the 5.19 
kernel), the 6.0.5 kernel seems to be working okay with an AMD Ryzen 9 
5950X cpu and a Nvidia RTX 3080 graphics card.

regards,
Steve

>  _______________________________________________
>  users mailing list -- users(a)lists.fedoraproject.org
>  To unsubscribe send an email to users-leave(a)lists.fedoraproject.org
>  Fedora Code of Conduct: https://docs.fedoraproject.org/en-US/project/code-of-conduct/
>  List Guidelines: https://fedoraproject.org/wiki/Mailing_list_guidelines
>  List Archives:
> https://lists.fedoraproject.org/archives/list/[e-mail]
>  Do not reply to spam, report it: https://pagure.io/fedora-infrastructure/new_issue

[Gabriel Ramirez]

On 10/31/22 08:15, Richard Shaw wrote:
>  Just a "me too" here, but I ran 6.0 fine when there was a
> call for 
>  testing. When I updated to 6.0.5 my computer (B550 w/ Ryzen 5 5700X) 
>  paused at the BIOS post for a long time and then kind of tried to load 
>  but stalled.
>
>  I forced it off and turned off the PS for 10 seconds and rebooted, 
>  then it seemed to come up fine. I didn't think anything of it at the 
>  time..
>
>  Thanks,
>  Richard
>
>  
The above happened to two machines with 6.0.4 and 6.05:

- a ryzen 7 3700X
- a thinkcentre with a i5 2400

the common denominator with the two, is have the partitions in linux 
raid1, /boot inclusive, booting from EFI, the efi partition is sda1, and 
are remote machines, my only access is via ssh.

I updated my laptop, ryzen, no raid, to 6.0.5, but it rebooted just fine 
with the new kernel

regards,

Gabriel

[Barry Scott]

>  On 30 Oct 2022, at 16:54, Doug H.
> <fedoraproject.org(a)wombatz.com&gt; wrote:
>  
>  Today was
>  the big kernel jump to 6.x kernel. 
No it is not a big jump it the one after 5.19.
After 20 minors Linux runs out of fingers and toes and bumps the major.
So after 6.19 with me 7.0 for example.

Barry

[Slade Watkins]

On 10/30/22 4:33 PM, Barry Scott wrote:
>  
> > On 30 Oct 2022, at 16:54, Doug H. <fedoraproject.org(a)wombatz.com&gt; wrote:
> >
> > Today was
> > the big kernel jump to 6.x kernel. 
As someone who tests 6.0 stable regularly, I can *assure you* it's
nothing major.

>  No it is not a big jump it the one after 5.19.
>  After 20 minors Linux runs out of fingers and toes and bumps the major.
>  So after 6.19 with me 7.0 for example. 
Exactly!
The "major number" means *nothing*, even Linus says that himself _a lot_.

-srw