From: Alexander Nyberg <alexn@dsv.su.se>
To: Steffen Michalke <StMichalke@web.de>
Cc: linux-kernel@vger.kernel.org
Subject: Re: Strange crashes of kernel v2.6.11
Date: Fri, 04 Mar 2005 13:23:13 +0100 [thread overview]
Message-ID: <1109938993.2285.13.camel@boxen> (raw)
In-Reply-To: <1109936036.6712.8.camel@pinky.local>
> > > I recently upgraded from linux kernel v2.6.10 to v2.6.11.
> > > Some programs like evolution 2.0 and leafnode2 crash the whole system
> > > immediatedly now.
> >
> > You mean when you run evolution the box hangs up completely? (you can't
> > kill X, switch to another console etc.)
>
> Thank you for your hints.
>
> When I looked into that problem recently, I remarked that the system
> does not actually crash but is locked totally:
>
> I use the device-mapper modules for encrypting files (loopback devices
> with aes-i586-encryption). They can be set up in the usual manner, but
> filesystem operations now lock the accessing processes, which cannot be
> killed afterwards.
> If the kernel has been compiled with preemption the system slows down
> considerably after those operations; enabling prempting The Big Kernel
> Lock locks the whole system at filesystem access (that looked like a
> system crash). That's why I could not find any messages in the logs.
>
> If I use a non-preemptive v2.6.11-kernel (vanilla, by the way) the
> system keeps on running the normal way, but every process which tries to
> work with files in device-mapped directories is unkillable locked.
>
> It seems to be a problem with the dm-*- or loop-modules.
Ok, let's try without preempt. Under the kernel config make sure you
select the following:
kernel hacking => kernel debugging => magic sysrq key
Also increase the kernel log buffer at, set it to 17 just to be safe
general setup => kernel log buffer size
Do what you do so that the processes become unkillable, then press
AltGr+SysRq+t 2-3 times and send the dmesg over here.
(sysrq normally being on the same key as 'print screen', at least on my
keyboards).
This should tell us what is going on.
Thanks!
prev parent reply other threads:[~2005-03-04 12:28 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-03-02 18:17 Strange crashes of kernel v2.6.11 Steffen Michalke
2005-03-02 21:32 ` Andrew Morton
2005-03-02 21:34 ` Alexander Nyberg
2005-03-04 11:33 ` Steffen Michalke
2005-03-04 12:23 ` Alexander Nyberg [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1109938993.2285.13.camel@boxen \
--to=alexn@dsv.su.se \
--cc=StMichalke@web.de \
--cc=linux-kernel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox