From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:45940) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WBlLk-0007Xp-5l for qemu-devel@nongnu.org; Fri, 07 Feb 2014 08:20:16 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WBlLa-0001kr-UU for qemu-devel@nongnu.org; Fri, 07 Feb 2014 08:20:08 -0500 Received: from mail-ee0-x22a.google.com ([2a00:1450:4013:c00::22a]:65214) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WBlLa-0001j8-O4 for qemu-devel@nongnu.org; Fri, 07 Feb 2014 08:19:58 -0500 Received: by mail-ee0-f42.google.com with SMTP id b15so1541109eek.29 for ; Fri, 07 Feb 2014 05:19:57 -0800 (PST) Sender: Paolo Bonzini Message-ID: <52F4DD77.7000209@redhat.com> Date: Fri, 07 Feb 2014 14:19:51 +0100 From: Paolo Bonzini MIME-Version: 1.0 References: <50562927-0498-460b-8f29-98722e332149@mailpro> <52F496B4.5030505@profihost.ag> <20140207091513.GC2374@work-vm> <52F4A554.1080400@profihost.ag> <20140207093159.GD2374@work-vm> <52F4A952.6080007@profihost.ag> <52F4CB3E.3030203@profihost.ag> <20140207122157.GF2374@work-vm> <52F4D1CA.6030501@profihost.ag> <52F4D523.9040809@redhat.com> <52F4D9E2.4010600@profihost.ag> In-Reply-To: <52F4D9E2.4010600@profihost.ag> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [pve-devel] QEMU LIve Migration - swap_free: Bad swap file entry List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Stefan Priebe - Profihost AG , "Dr. David Alan Gilbert" Cc: m.gibula@beyond.pl, qemu-devel , Alexandre DERUMIER Il 07/02/2014 14:04, Stefan Priebe - Profihost AG ha scritto: > first of all i've now a memory image of a VM where i can reproduce it. You mean you start that VM with -incoming 'exec:cat /path/to/vm.img'? But google stress test doesn't report any error until you start migration _and_ it finishes? That sounds good enough. Can you upload the image somewhere (doesn't have to be a public place, you can contact David or others offlist)? > reproducing does NOT work if i boot the VM freshly i need to let it run > for some hours. > > Then just when the migration finishes there is a short time frame where > the google stress app reports memory errors than when the migration > finishes it runs fine again. > > It seems to me it is related to pause and unpause/resume? > >> > - with Unix migration between two QEMUs on the same host > now tested => same issue > >> > - with different hosts > already tested => same issue > >> > - with a different network (e.g. just a cross cable between two machines) > already tested => same issue Another test: - start the VM with -S, migrate, do errors appear on the destination? Thanks, Paolo