linux-kernel - Dying hardware?: Eeek! page_mapcount(page) went negative! (-1)

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>] [day] [month] [year] [list]

Message-ID: <495A4EE4.5020701@ribosome.natur.cuni.cz>
Date:	Tue, 30 Dec 2008 17:40:04 +0100
From:	Martin MOKREJŠ <mokrejsm@...osome.natur.cuni.cz>
To:	LKML <linux-kernel@...r.kernel.org>
Subject: Dying hardware?: Eeek! page_mapcount(page) went negative! (-1)

Hi,
  I have a machine with Intel Desktop board D865GBF which
ran fine last 3-4 years. My last kernel was 2.6.24.7 since
May 12 2008. Last few days I get daily crashes of the kernel
with "Eeek! page_mapcount(page) went negative! (-1)" while
the stacktraces differ. I tried even 2.6.28 but again, after
one hour uptime I hit this again. I am attaching the dmesg
output. One of the other crashes I had retyped on a piece of
paper and it seemed reiserfs was involved. I could retype
it next emails if somebody is interested. It caused some
extra kernel crash but it was 2.6.24.7 kernel so maybe
nobody is interested in those traces anymore. I tried
memtest for a while and it did not find any problems, but
did not let it run more than few percent.

  I have attached serial console now and maybe would manage
setup the console over firewire and somebody tells me what
to do if the crash happens again. ;-)

Thanks for any clues,
Martin
P.S.: Please Cc: me in replies.

View attachment "dmesg-2.6.28.txt" of type "text/plain" (25965 bytes)

Download attachment "config.gz" of type "application/gzip" (12047 bytes)