linux-kernel - Re: 2.6.25-rc4 OOMs itself dead on bootup

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

Date:	Thu, 6 Mar 2008 19:49:54 +0100
From:	Ingo Molnar <mingo@...e.hu>
To:	Frank Sorenson <frank@...rocks.com>
Cc:	LKML <linux-kernel@...r.kernel.org>, linux-mm@...ck.org,
	"Rafael J. Wysocki" <rjw@...k.pl>
Subject: Re: 2.6.25-rc4 OOMs itself dead on bootup


* Frank Sorenson <frank@...rocks.com> wrote:

> 2.6.25-rc4 invokes the oom-killer repeatedly when attempting to boot, 
> eventually panicing with "Out of memory and no killable processes." 
> This happens since at least 2.6.25-rc3, but 2.6.24 boots just fine,
> 
> The system is a Dell Inspiron E1705 running Fedora 8 (x86_64).
> 
> My .config is at http://tuxrocks.com/tmp/config-2.6.25-rc4, and a 
> syslog of the system up until the point where it oom-killed syslog 
> (just before the panic) is at 
> http://tuxrocks.com/tmp/oom-2.6.25-rc4.txt

i've picked up your .config and enabled a few drivers in it to make it 
boot on a testsystem of mine and it doesnt OOM:

 20:47:17 up 10 min,  1 user,  load average: 0.01, 0.07, 0.03
              total       used       free     shared    buffers     cached
 Mem:       1025768     258544     767224          0      12956     169556
 -/+ buffers/cache:      76032     949736
 Swap:      3911816          0    3911816

(config and bootlog from my box attached.)

So it's probably not a .config dependent generic kernel problem, but 
probably something specific to your hardware.

Since the first oom happens about 9 minutes into the bootup:

 [  569.755853] sh invoked oom-killer: gfp_mask=0x1201d2, order=0, oomkilladj=-17

do you have any chance to log in and capture MM statistics? The 
following script will capture a bunch of statistics:

 http://people.redhat.com/mingo/cfs-scheduler/tools/cfs-debug-info.sh

and takes less than a minute to run - should be enough time in theory. 
If it's possible to run it then send us the output file it produces.

	Ingo

View attachment "config" of type "text/plain" (59705 bytes)

View attachment "boot.log" of type "text/plain" (109099 bytes)