Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.debian.kernel > #85219 > unrolled thread

Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64

Back to article view | Back to linux.debian.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Salvatore Bonaccorso <carnil@debian.org> - 2025-01-19 21:00 +0100
    Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Mark Mandel <mark.mandel@gmail.com> - 2025-01-22 00:30 +0100
      Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Mark Mandel <mark.mandel@gmail.com> - 2025-01-26 18:50 +0100
        Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Mark Mandel <mark.mandel@gmail.com> - 2025-01-26 19:40 +0100
          Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Mark Mandel <mark.mandel@gmail.com> - 2025-01-26 20:30 +0100
            Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Salvatore Bonaccorso <carnil@debian.org> - 2025-01-27 19:00 +0100
              Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64 Mark Mandel <mark.mandel@gmail.com> - 2025-01-27 19:10 +0100

#85219 — Bug#1093435: linux-image-6.12.9-amd64: Unable to boot on linux-image-6.12.9-amd64

Hi Mark,

Thanks for the reply.

On Sun, Jan 19, 2025 at 11:24:15AM -0800, Mark Mandel wrote:
> This is a system I use regularly, so I am unable to test without Nvidia.

Yes I understand, we need though a way to determine if the proprietary
modules tainting the kernel are responsible (for an upstream report
thee is need for the kernel not to be tainted, otherwise there is
unlikely it is going to be looked at).
> 
> That being said, all nvidia packages came from Debian repositories.

FWIW, it does less "matter" here if they are from Debian or externally
installed, the problem I would like to exclude is if tainting modules
have an influence. Let's try to tackle it from another angle: You can
have a look at #1089513.

Assuming you can access the machine remotely as well, when the system
seems to hang, can you still login remotely to the machine? Is just
the graphical stack not coming up? If you can login, can you provide
the full boot log and check if things matches to the bug #1089513?

Secondly, we have in meanwhile 6.12.10 in unstable, can you upgrade to
this kernel. Is the problem still reproducible with 6.12.10?

Is the system booting with quiet in the kernel command line? If Yes,
can you drop it so you get more verbose boot log, where does it stops?

Regards,
Salvatore

[toc] | [next] | [standalone]

#85246

[Multipart message — attachments visible in raw view] — view raw

>
>
>
> Yes I understand, we need though a way to determine if the proprietary
> modules tainting the kernel are responsible (for an upstream report
> thee is need for the kernel not to be tainted, otherwise there is
> unlikely it is going to be looked at).
>

Totally fair - thanks for explaining. Would a boot from a liveUSB suffice?
I did have similar issues off a liveUSB when installing - but it was
relatively random, and less frequent.

I tried to SSH In on a hang, but unfortunately, SSH was unresponsive.



>
> Assuming you can access the machine remotely as well, when the system
> seems to hang, can you still login remotely to the machine? Is just
> the graphical stack not coming up? If you can login, can you provide
> the full boot log and check if things matches to the bug #1089513?
>

Right now I don't use a graphical login like sddm/gdm/etc -- been starting
from scratch, so my system boots directly into a terminal. But yes, I also
have the 1089513 issue, i.e. Nvidia doesn't work with 6.12.x (but that's a
separate issue! 😆)


>
> Secondly, we have in meanwhile 6.12.10 in unstable, can you upgrade to
> this kernel. Is the problem still reproducible with 6.12.10?
>
> Is the system booting with quiet in the kernel command line? If Yes,
> can you drop it so you get more verbose boot log, where does it stops?
>

Ooh I can try this!

Recently I've noticed it will boot into 6.12.9 -- it doesn't seem to fail
as often (I have been trying it after every apt full-upgrade I run).
I've been keeping everything very up to date, in hopes of a fix ☺️🤞🏻 so,
maybe something else got updated recenty?

I'll keep you updated as I dig in further!

Mark

[toc] | [prev] | [next] | [standalone]

#85294

[Multipart message — attachments visible in raw view] — view raw

So good news! I removed Nvidia packages, and it seems like I can't
reproduce this on 6.12.10 anymore. I've rebooted several times.

So it's quite possible this is more to do with Nvidia issues than the
kernel. Appreciate you taking the time to walk me through this!

Mark


>

[toc] | [prev] | [next] | [standalone]

#85296

[Multipart message — attachments visible in raw view] — view raw

Woah! I think I can reproduce this!

I was trying to manually install Nvidia drivers to see how that would go!
But didn't get that far. Only got as far as creating a file in
/etc/modprobe.d to disable nouveau, with contents:

```
blacklist nouveau
options nouveau modeset=0
```

Once that is in place, I get the regular freezes on startup (as above).
I've not even installed any nvidia packages.

I'm going to try just removing the nouveau packages entirely, see if that
does anything.

mark

>

[toc] | [prev] | [next] | [standalone]

#85299

[Multipart message — attachments visible in raw view] — view raw

Installed the manual proprietary Nvidia drivers and the crash is
back 🤷🏻‍♂️

I think this is the end of me debugging this.

>

[toc] | [prev] | [next] | [standalone]

#85302

HI,

On Sun, Jan 26, 2025 at 11:19:45AM -0800, Mark Mandel wrote:
> Installed the manual proprietary Nvidia drivers and the crash is
> back 🤷🏻‍♂️
> 
> I think this is the end of me debugging this.

So I interpret the results as having identified the problem might just
be in src:nvidia-graphics-drivers.

If this is correct we would reassign the bug, but will wait for your
reply.

Regards,
Salvatore

[toc] | [prev] | [next] | [standalone]

#85303

[Multipart message — attachments visible in raw view] — view raw

I'll be honest, I'm not entirely sure. I am now trying again on a different
Alienware laptop to see if the issue resurfaces.

On Mon, 27 Jan 2025 at 09:52, Salvatore Bonaccorso <carnil@debian.org>
wrote:

> HI,
>
> On Sun, Jan 26, 2025 at 11:19:45AM -0800, Mark Mandel wrote:
> > Installed the manual proprietary Nvidia drivers and the crash is
> > back 🤷🏻‍♂️
> >
> > I think this is the end of me debugging this.
>
> So I interpret the results as having identified the problem might just
> be in src:nvidia-graphics-drivers.
>
> If this is correct we would reassign the bug, but will wait for your
> reply.
>
> Regards,
> Salvatore
>

[toc] | [prev] | [standalone]

Back to top | Article view | linux.debian.kernel

csiph-web