From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:57105) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VHGYm-0003Pg-9f for qemu-devel@nongnu.org; Wed, 04 Sep 2013 13:08:12 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1VHGYc-0008KI-4z for qemu-devel@nongnu.org; Wed, 04 Sep 2013 13:08:04 -0400 Received: from mail-yh0-x22a.google.com ([2607:f8b0:4002:c01::22a]:60661) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1VHGYc-0008JP-1B for qemu-devel@nongnu.org; Wed, 04 Sep 2013 13:07:54 -0400 Received: by mail-yh0-f42.google.com with SMTP id z12so237134yhz.1 for ; Wed, 04 Sep 2013 10:07:53 -0700 (PDT) Sender: Paolo Bonzini Message-ID: <522768E9.9080408@redhat.com> Date: Wed, 04 Sep 2013 19:07:53 +0200 From: Paolo Bonzini MIME-Version: 1.0 References: In-Reply-To: Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] When does live migration give up? List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Alex Bligh Cc: qemu-devel@nongnu.org Il 04/09/2013 17:24, Alex Bligh ha scritto: > We have seen a situation when migrating about 50 VMs at once where some > of them fail. I think this is because they are dirtying pages faster than > they can be transmitted. No, migration never "gives up". It may never converge, but it keeps trying until cancelled. Could it be that you are choosing migration server ports from a small range, and some of them are failing because two migrations pick the same random port for the destination (which is where the server socket lies)? Paolo > What algorithm controls when migration fails in this way, and is it > tunable? > > I am fully aware one answer to this question is "do not attempt to > migrate 50 busy VMs through a single 1GB/s NIC".