linux-kernel - Re: 2.6.36 io bring the system to its knees

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]

Message-ID: <AANLkTimc4wphG4BXv=rN8MVvFDhnkjtWn_hgHzyJVucw@mail.gmail.com>
Date:	Sun, 31 Oct 2010 18:49:39 +0100
From:	Corrado Zoccolo <czoccolo@...il.com>
To:	"Ted Ts'o" <tytso@....edu>, Aidar Kultayev <the.aidar@...il.com>,
	Ingo Molnar <mingo@...e.hu>, Pekka Enberg <penberg@...nel.org>,
	Chris Mason <chris.mason@...cle.com>,
	linux-kernel@...r.kernel.org, linux-mm@...ck.org,
	Linus Torvalds <torvalds@...ux-foundation.org>,
	Andrew Morton <akpm@...ux-foundation.org>,
	Jens Axboe <axboe@...nel.dk>,
	Peter Zijlstra <a.p.zijlstra@...llo.nl>,
	Nick Piggin <npiggin@...e.de>,
	Arjan van de Ven <arjan@...radead.org>,
	Thomas Gleixner <tglx@...utronix.de>
Subject: Re: 2.6.36 io bring the system to its knees

On Sun, Oct 31, 2010 at 3:31 AM, Ted Ts'o <tytso@....edu> wrote:
> On Sat, Oct 30, 2010 at 07:02:35PM +0600, Aidar Kultayev wrote:
>> the system is/was doing :
>> .dd if=/dev/zero of=test.10g bs=1M count=10000;rm test.10g
>> .netbeans
>> .compiling gcc-4.5.1
>> .running VBox, which wasn't doing any IO. The guest os was idle in other words
>> .vlc
>> .chromium
>> .firefox
>> and bunch of other small stuff.
>>
>> Even without having running DD, the mouse cursor would occasionally
>> lag. The alt+tab effect in KWin would take 5+seconds to workout.
>> When I run DD on top of the workload it consistently made system much
>> more laggy. The cursor would freeze much more frequent. It is like if
>> you drag your mouse physically, but the cursor on the screen would
>> jump discretely, in other words there is no continuity.
>> Music would stop.
>
> If you start shutting down tasks, Vbox, netbeans, chromium, etc., at
> what point does the cursor start tracking the system easily?  Is the
> system swapping?  Do you know how to use tools like dstat or iostat to
> see if the system is actively writing to the swap partition?  (And are
> you using a swap partition or a swap file?)
>
> The fact that cursor isn't tracking well even when the dd is running,
> and presumably the only source of I/O is the gcc and vlc, makes me
> suspect that you may be swapping pretty heavily.  Have you tried
> investigating that possibility, and made sure it has been ruled out?

Something to try is also to raise X cpu scheduling priority, since I
would be really surprised if we evict from memory the routine that
draws the cursor.
BTW, I've seen the cursor jumping problem even when not swapping, and
with minimal *real* disk activity (but with heavy usage of a fuse
filesystem providing remote resources), and high cpu activity.
Raising X priority solved the problem with the mouse pointer, but the
gui programs still didn't respond quickly...

Thanks
Corrado

>
>                                                - Ted
>
> --
> To unsubscribe, send a message with 'unsubscribe linux-mm' in
> the body to majordomo@...ck.org.  For more info on Linux MM,
> see: http://www.linux-mm.org/ .
> Fight unfair telecom policy in Canada: sign http://dissolvethecrtc.ca/
> Don't email: <a href=mailto:"dont@...ck.org"> email@...ck.org </a>
>
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/