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

Groups > linux.debian.kernel > #79111 > unrolled thread

Bug#1036644: linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast

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#1036644: linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast Diederik de Haas <didi.debian@cknow.org> - 2023-05-23 22:40 +0200
    Processed: Re: Bug#1036644: linux-image-6.1.0-9-amd64: System  crashes. Netconsole reports CPUs not responding to MCE broadcast "Debian Bug Tracking System" <owner@bugs.debian.org> - 2023-05-23 22:40 +0200
    Bug#1036644: linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast Olivier Berger <olivier.berger@telecom-sudparis.eu> - 2023-05-24 08:30 +0200

#79111 — Bug#1036644: linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast

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

Control: found -1 6.1.25-1
Control: retitle -1 Kernel panic - not syncing: Timeout: Not all CPUs entered broadcast exception handler

On Tuesday, 23 May 2023 18:49:00 CEST Olivier Berger wrote:
> It used to work fine with 6.1.0-7 but has had problems with the 2 later
> updates of the testing kernel.

The stack traces should be useful for someone who understands those (which
isn't me), but I did notice several other items:

- [  465.284645] GPT: Use GNU Parted to correct GPT errors
That happened after you plugged in an USB drive?
I would follow that advice, but it would be useful to get that USB drive
'out of the equation'.
Does the issue also occur when that USB drive isn't used?
The kernel seems to assign both sda and sdb before settling on sda(1)?
Not sure what to make of that, but it doesn't look good

- [  535.857315] EXT4-fs (dm-0): recovery complete
I can understand a FS recovery when you're dealing with a freeze/crash,
but I find the timing a 'bit' unusual. After 9.5 minutes, I doubt it's the
primary/boot drive (and we had the USB drive before that), so where
is that coming from?

- [  543.576681] systemd-journald[428]: Sent WATCHDOG=1 notification
I'm not really sure what that means, but afaik a watchdog is used to
(automatically) reboot the machine if the system hangs.
So seeing that message numerous times, is worrisome. And it looks like it
doesn't do its actual job?

- BIOS T70 Ver. 01.13.01 03/30/2023
Can you check whether there is a newer BIOS version available?
I believe 'NMI' is BIOS related, so it may have an effect.

[toc] | [next] | [standalone]

#79112 — Processed: Re: Bug#1036644: linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast

Processing control commands:

> found -1 6.1.25-1
Bug #1036644 [src:linux] linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast
Marked as found in versions linux/6.1.25-1.
> retitle -1 Kernel panic - not syncing: Timeout: Not all CPUs entered broadcast exception handler
Bug #1036644 [src:linux] linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast
Changed Bug title to 'Kernel panic - not syncing: Timeout: Not all CPUs entered broadcast exception handler' from 'linux-image-6.1.0-9-amd64: System crashes. Netconsole reports CPUs not responding to MCE broadcast'.

-- 
1036644: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1036644
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems

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

#79113

Hi.

Diederik de Haas <didi.debian@cknow.org> writes:

>
> The stack traces should be useful for someone who understands those (which
> isn't me), but I did notice several other items:
>
> - [  465.284645] GPT: Use GNU Parted to correct GPT errors
> That happened after you plugged in an USB drive?
> I would follow that advice, but it would be useful to get that USB drive
> 'out of the equation'.
> Does the issue also occur when that USB drive isn't used?
> The kernel seems to assign both sda and sdb before settling on sda(1)?
> Not sure what to make of that, but it doesn't look good
>

I guess the USB drive has nothing to do with the issue, AFAIU. Actually,
I just wanted to be sure that netconsole was indeed capturing kernel
events, as suggested by a howto on remote debugging of kernel panics
with netconsole. And FYI, this is a USB key that embeds a SD card
reader, hence the 2 drives that popup... as for GPT, dunno, maybe a
formatting mistake.
In any case, the laptop crashed in the past whenever no such USB key was
being plugged.

> - [  535.857315] EXT4-fs (dm-0): recovery complete
> I can understand a FS recovery when you're dealing with a freeze/crash,
> but I find the timing a 'bit' unusual. After 9.5 minutes, I doubt it's the
> primary/boot drive (and we had the USB drive before that), so where
> is that coming from?
>

Thats a LUKS partition being mounted after a while by me, for secrets
stored on the hard drive in a dedicated partition. As the laptop crashed
in the previous execution with the partition mounted, it explains the
FS recovery at mount time.

Nothing strange here either.

> - [  543.576681] systemd-journald[428]: Sent WATCHDOG=1 notification
> I'm not really sure what that means, but afaik a watchdog is used to
> (automatically) reboot the machine if the system hangs.
> So seeing that message numerous times, is worrisome. And it looks like it
> doesn't do its actual job?
>

I booted with 'debug ignore_loglevel' as kernel arguments... maybe that
explains the occurence of such logs... dunno exactly if this is
worrysome.

> - BIOS T70 Ver. 01.13.01 03/30/2023
> Can you check whether there is a newer BIOS version available?
> I believe 'NMI' is BIOS related, so it may have an effect.

I just updated the HP BIOS to the latest available the last day, but
crashes were occuring before too... maybe related, but nothing can be
updated more for the moment, at least from what the Windows HP Support
Assistant can show.

Thanks for your help.

Best regards,

-- 
Olivier BERGER
https://www-public.imtbs-tsp.eu/~berger_o/ - OpenPGP 2048R/0xF9EAE3A65819D7E8
Ingenieur Recherche - Dept INF
Institut Mines-Telecom, Telecom SudParis, Evry (France)

[toc] | [prev] | [standalone]

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

csiph-web