From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from [140.186.70.92] (port=40076 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1OT9y1-0006Ft-U0
	for qemu-devel@nongnu.org; Mon, 28 Jun 2010 04:45:26 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <quentusrex@gmail.com>) id 1OT9y0-0008V8-Hf
	for qemu-devel@nongnu.org; Mon, 28 Jun 2010 04:45:25 -0400
Received: from mail-px0-f173.google.com ([209.85.212.173]:64035)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <quentusrex@gmail.com>) id 1OT9y0-0008Uv-DH
	for qemu-devel@nongnu.org; Mon, 28 Jun 2010 04:45:24 -0400
Received: by pxi2 with SMTP id 2so525377pxi.4
	for <qemu-devel@nongnu.org>; Mon, 28 Jun 2010 01:45:23 -0700 (PDT)
Message-ID: <4C286124.1020309@gmail.com>
Date: Mon, 28 Jun 2010 01:45:24 -0700
From: William King <quentusrex@gmail.com>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] Re: qemu-kvm guest network stalls
References: <4C2664B6.9070302@gmail.com>	<4C28586C.9000205@gmail.com>
	<AANLkTil-_QIXgUnrCdlJoPiaVbmyTcd2osn9_0BkCNOx@mail.gmail.com>
In-Reply-To: <AANLkTil-_QIXgUnrCdlJoPiaVbmyTcd2osn9_0BkCNOx@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: qemu-devel@nongnu.org

Thanks for the reply Stefan.

That looks to be exactly the same bug I'm dealing with here. The one
difference is that I can reproduce it easily on test hardware. So I have
two separate boxes that can go down(and run test code) without
interfering with anything in production.

My test setup has a local torrent server, and 5 other machines on the
network seeding a very large(10GB) random data file that I created. When
the guest tries to download this file from the other servers this
aggravates the issue. To the point that I could cause it to fail within
minutes, repeatedly. I created this test after someone on irc said that
the issue could have something to do with udp traffic rather than tcp
traffic.

I am running latest ubuntu 10.04 on the hosts and the guests, and I can
wipe(and reinstall) the hosts quickly and easily.

On 06/28/2010 01:28 AM, Stefan Hajnoczi wrote:
> On Mon, Jun 28, 2010 at 9:08 AM, William King <quentusrex@gmail.com> wrote:
>> Is this the wrong list to get help tracking down why the network stalls
>> under heavy load?
> 
> This list is the right one.  There is at least one known network stall bug:
> 
> http://sourceforge.net/tracker/?func=detail&atid=893831&aid=2506814&group_id=180599
> http://www.mail-archive.com/kvm@vger.kernel.org/msg06774.html
> 
> I sounds like you have the same issue?
> 
> Stefan