From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:51887) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1c1PvL-0003BT-Cw for qemu-devel@nongnu.org; Mon, 31 Oct 2016 23:39:44 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1c1PvG-0008JU-MY for qemu-devel@nongnu.org; Mon, 31 Oct 2016 23:39:43 -0400 Received: from szxga01-in.huawei.com ([58.251.152.64]:9124) by eggs.gnu.org with esmtps (TLS1.0:RSA_ARCFOUR_SHA1:16) (Exim 4.71) (envelope-from ) id 1c1PvG-0008I6-Ar for qemu-devel@nongnu.org; Mon, 31 Oct 2016 23:39:38 -0400 References: <1477824430-1460-1-git-send-email-amit.shah@redhat.com> <1477824430-1460-4-git-send-email-amit.shah@redhat.com> <75e4c102-8057-90f6-28fd-9657dccc9b46@redhat.com> From: Hailiang Zhang Message-ID: <58180E5B.8040405@huawei.com> Date: Tue, 1 Nov 2016 11:39:07 +0800 MIME-Version: 1.0 In-Reply-To: <75e4c102-8057-90f6-28fd-9657dccc9b46@redhat.com> Content-Type: text/plain; charset="utf-8"; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] [PULL 03/18] migration: Enter into COLO mode after migration if COLO is enabled List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Eric Blake , Amit Shah , Peter Maydell Cc: qemu list , "Dr. David Alan Gilbert" , Juan Quintela On 2016/11/1 6:27, Eric Blake wrote: > On 10/30/2016 05:46 AM, Amit Shah wrote: >> From: zhanghailiang >> >> Add a new migration state: MIGRATION_STATUS_COLO. Migration source side >> enters this state after the first live migration successfully finished >> if COLO is enabled by command 'migrate_set_capability x-colo on'. >> >> We reuse migration thread, so the process of checkpointing will be handled >> in migration thread. >> > >> +++ b/qapi-schema.json >> @@ -459,12 +459,14 @@ >> # >> # @failed: some error occurred during migration process. >> # >> +# @colo: VM is in the process of fault tolerance. (since 2.8) >> +# > > The commit message mentions it, but the public documentation does not, > so it might be worth a followup patch that clarifies in the schema that > the 'colo' state is not possible without using the 'x-colo' migration > property (so that applications like libvirt know that they don't have to > worry about handling an unexpected state, at least not in 2.8 or until > the 'x-colo' is promoted to plain 'colo'). > Sounds reasonable, I'll send a patch to add such clarification. Thanks, Hailiang