From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43) id 1MGdFc-0007BA-VZ for qemu-devel@nongnu.org; Tue, 16 Jun 2009 14:19:16 -0400 Received: from exim by lists.gnu.org with spam-scanned (Exim 4.43) id 1MGdFY-0007Ay-Sg for qemu-devel@nongnu.org; Tue, 16 Jun 2009 14:19:16 -0400 Received: from [199.232.76.173] (port=47735 helo=monty-python.gnu.org) by lists.gnu.org with esmtp (Exim 4.43) id 1MGdFY-0007Av-Qp for qemu-devel@nongnu.org; Tue, 16 Jun 2009 14:19:12 -0400 Received: from main.gmane.org ([80.91.229.2]:37781 helo=ciao.gmane.org) by monty-python.gnu.org with esmtps (TLS-1.0:RSA_AES_256_CBC_SHA1:32) (Exim 4.60) (envelope-from ) id 1MGdFY-0006uS-0n for qemu-devel@nongnu.org; Tue, 16 Jun 2009 14:19:12 -0400 Received: from list by ciao.gmane.org with local (Exim 4.43) id 1MGdFT-0006V6-9P for qemu-devel@nongnu.org; Tue, 16 Jun 2009 18:19:07 +0000 Received: from 143.166.197.6 ([143.166.197.6]) by main.gmane.org with esmtp (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 16 Jun 2009 18:19:07 +0000 Received: from Charles_Duffy by 143.166.197.6 with local (Gmexim 0.1 (Debian)) id 1AlnuQ-0007hv-00 for ; Tue, 16 Jun 2009 18:19:07 +0000 From: Charles Duffy Date: Tue, 16 Jun 2009 13:19:05 -0500 Message-ID: References: <4A36B025.2080602@us.ibm.com> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit In-Reply-To: <4A36B025.2080602@us.ibm.com> Sender: news Subject: [Qemu-devel] Re: Live migration broken when under heavy IO List-Id: qemu-devel.nongnu.org List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: qemu-devel@nongnu.org I'm not sure if this is related, but to comment -- I'm seeing what appears to be occasional dropped I/O on non-live migration -- that is to say, stop, migrate-to-disk, shuffle files around, migrate-from-disk, cont. I have not yet been able to reproduce this with cache=off. As I understand it, the bulk of discussion in this thread is about I/O submitted by the guest after the migration starts; as I'm stopping the guest CPU _before_ initiating migration, there would appear to be outstanding issues not limited to this scenario. [qemu-kvm-0.10.5, IDE, qcow2]