linux-kernel - Re: Kernel panic - not syncing: Attempted to kill init! exitcode=0x00007f00

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <f09763e4-5c1e-9293-8366-91575e0ef037@inwind.it>
Date:   Fri, 10 Mar 2023 10:18:03 +0100
From:   Valerio Vanni <valerio.vanni@...ind.it>
To:     linux-kernel@...r.kernel.org
Cc:     Slade Watkins <srw@...dewatkins.net>
Subject: Re: Kernel panic - not syncing: Attempted to kill init!
 exitcode=0x00007f00

Il 25/02/2023 13:28, Valerio Vanni ha scritto:
> Il 24/02/2023 19:06, Slade Watkins ha scritto:
> 
>>> https://bugzilla.kernel.org/show_bug.cgi?id=217083
>>>
>>> I got this error after upgrading a Debian machine (x86-64) from Stretch
>>> to Buster. Upgrade is successful, but the next boot it crashes.
>>
>> Thanks for the details. What was the last known good kernel version
>> that did not exhibit this issue on your system?
> 
> There isn't a last known good version. As I said, the issue appeared 
> after an OS upgrade, not after a kernel one.
I updated bugzilla and I update also here.

I use serial console to collect diagnostic data, but now I've found that 
it misses lines.
In kernel boot line I put "console=ttyS0 console=tty0" so I get both 
local and serial console.

But the issue is: tty0 shows more messages. Is there a way to get 
everything also on ttyS0? It would be useful, since you cant' scroll 
tty0 after a kernel panic.

When kernel loads, tty0 scrolls very fast. I could see some more line 
but wasn't able to read them.
The only way I found was adding a (slow) speed to ttyS0, 
"console=ttyS0,1200" on kernel boot line.
This slowed down ttyS0, and tty0 was forced to go at the same pace.

So I could read a line (only on tty0), between "Run /sbin/init as init 
process" and start of kernel panic.
That line complained that it could not find something in /usr/lib/.... 
At that stage, /usr was not mounted yet.

Then I restored full disk image on the machine, an before upgrading from 
stretch to buster I merged / and /usr, so that /usr/lib/... was 
available from / mount.

After that, kernel panic did not show anymore.