This is an archived copy of the Xen.org mailing list, which we have preserved to ensure that existing links to archives are not broken. The live archive, which contains the latest emails, can be found at http://lists.xen.org/
Home Products Support Community News


[Xen-devel] [PATCH] [HVM] fix a HVM save/restore race condition after r1

To: Tim <Tim.Deegan@xxxxxxxxxxxxx>
Subject: [Xen-devel] [PATCH] [HVM] fix a HVM save/restore race condition after r15639
From: "Zhai, Edwin" <edwin.zhai@xxxxxxxxx>
Date: Mon, 13 Aug 2007 22:55:59 +0800
Cc: xen-devel@xxxxxxxxxxxxxxxxxxx, "Zhai, Edwin" <edwin.zhai@xxxxxxxxx>
Delivery-date: Mon, 13 Aug 2007 08:03:35 -0700
Envelope-to: www-data@xxxxxxxxxxxxxxxxxx
List-help: <mailto:xen-devel-request@lists.xensource.com?subject=help>
List-id: Xen developer discussion <xen-devel.lists.xensource.com>
List-post: <mailto:xen-devel@lists.xensource.com>
List-subscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=subscribe>
List-unsubscribe: <http://lists.xensource.com/cgi-bin/mailman/listinfo/xen-devel>, <mailto:xen-devel-request@lists.xensource.com?subject=unsubscribe>
Sender: xen-devel-bounces@xxxxxxxxxxxxxxxxxxx
User-agent: Mutt/1.5.11

The qemu save command sending is moved from saveInputHandler to 
refreshShutdown(too late) in r15639, so higher possibility that control panel 
read a not ready qemu state file especially for live migration.

If reading happened before qemu create this file, a python missing file 
exception would occur. If before qemu put the record length in the file header 
as the last step, dest machine would try to restore a zero length qemu file to 
lead state lost in each device.

Attached patch can fix it, pls. have a look.


best rgds,

Attachment: hvm_live_migration_fix_r15716.patch
Description: Text document

Xen-devel mailing list
<Prev in Thread] Current Thread [Next in Thread>
  • [Xen-devel] [PATCH] [HVM] fix a HVM save/restore race condition after r15639, Zhai, Edwin <=