From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Michael S. Tsirkin" Subject: Re: Query: Is it possible to lose interrupts between vhost and virtio_net during migration? Date: Tue, 5 Aug 2014 11:49:57 +0200 Message-ID: <20140805094957.GB24619@redhat.com> References: <53DA2CCC.1040606@huawei.com> <20140731143100.GA3834@redhat.com> <20140731143725.GA3875@redhat.com> <53DB705F.2000405@redhat.com> <53DB76A9.8010305@redhat.com> <53E079C8.7050700@huawei.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jason Wang , netdev@vger.kernel.org, qinchuanyu@huawei.com, liuyongan@huawei.com, davem@davemloft.net To: "Zhangjie (HZ)" Return-path: Received: from mx1.redhat.com ([209.132.183.28]:47831 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933035AbaHEJuG (ORCPT ); Tue, 5 Aug 2014 05:50:06 -0400 Content-Disposition: inline In-Reply-To: <53E079C8.7050700@huawei.com> Sender: netdev-owner@vger.kernel.org List-ID: On Tue, Aug 05, 2014 at 02:29:28PM +0800, Zhangjie (HZ) wrote: > Jason is right, the new order is not the cause of network unreachable. > Changing order seems not work. After about 40 times, the problem occurs again. > Maybe there is other hidden reasons for that. To make sure, you tested the patch that I posted to list: "vhost_net: stop guest notifiers after backend"? Please confirm. -- MST