From mboxrd@z Thu Jan 1 00:00:00 1970 From: ebiederm@xmission.com (Eric W. Biederman) Subject: Re: unregister_netdevice: waiting for lo to become free. Usage count = 8 Date: Fri, 15 Apr 2011 00:27:20 -0700 Message-ID: References: <201104150901.47214.hans@schillstrom.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Julian Anastasov , Simon Horman , netdev@vger.kernel.org, lvs-devel@vger.kernel.org To: Hans Schillstrom Return-path: In-Reply-To: <201104150901.47214.hans@schillstrom.com> (Hans Schillstrom's message of "Fri, 15 Apr 2011 09:01:46 +0200") Sender: lvs-devel-owner@vger.kernel.org List-Id: netdev.vger.kernel.org Hans Schillstrom writes: > Hello Julian > > I'm trying to fix the cleanup process when a namespace get "killed", > which is a new feature for ipvs. However an old problem appears again > > When there has been traffic trough ipvs where the destination is unreachable > the usage count on loopback dev increases one for every packet.... > I guess thats because of this rule : > > # ip route list table all > ... > unreachable default dev lo table 0 proto kernel metric 4294967295 error -101 hoplimit 25 > ... > > I made a test just forwarding packets through the same container (ipvs loaded) > to an unreachable destination and that test had a balanced count i.e. it was possible to reboot the container. > > Do you have an idea why this happens in the ipvs case ? Hans. I do know that most outstanding references when you clean up a container get moved to the loopback device. So it may not originally be the loopback device itself where the reference counting is wrong. Eric