From mboxrd@z Thu Jan 1 00:00:00 1970 From: Rick Jones Subject: Re: [iproute PATCH 0/2] Netns performance improvements Date: Thu, 7 Jul 2016 09:16:20 -0700 Message-ID: <577E8054.6040603@hpe.com> References: <1467729773-16751-1-git-send-email-phil@nwl.cc> <87twg4ywjz.fsf@x220.int.ebiederm.org> <20160705205103.GE620@orbyte.nwl.cc> <87h9c259ip.fsf@x220.int.ebiederm.org> <20160707111718.GL620@orbyte.nwl.cc> <577E5244.8030601@6wind.com> <20160707154809.GN620@orbyte.nwl.cc> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE To: Phil Sutter , Nicolas Dichtel , "Eric W. Biederman" , Stephen Hemminger , netdev@vger.kernel.org Return-path: Received: from g2t2354.austin.hpe.com ([15.233.44.27]:13004 "EHLO g2t2354.austin.hpe.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751220AbcGGQQ2 (ORCPT ); Thu, 7 Jul 2016 12:16:28 -0400 In-Reply-To: <20160707154809.GN620@orbyte.nwl.cc> Sender: netdev-owner@vger.kernel.org List-ID: On 07/07/2016 08:48 AM, Phil Sutter wrote: > On Thu, Jul 07, 2016 at 02:59:48PM +0200, Nicolas Dichtel wrote: >> Le 07/07/2016 13:17, Phil Sutter a =C3=A9crit : >> [snip] >>> The issue came up during OpenStack Neutron testing, see this ticket= for >>> reference: >>> >>> https://bugzilla.redhat.com/show_bug.cgi?id=3D1310795 >> Access to this ticket is not public :( > > *Sigh* OK, here are a few quotes: > > "OpenStack Neutron controller nodes, when undergoing testing, are > locking up specifically during creation and mounting of namespaces. > They appear to be blocking behind vfsmount_lock, and contention for t= he > namespace_sem" > > "During the scale testing, we have 300 routers, 600 dhcp namespaces > spread across four neutron network nodes. When then start as one set = of > standard Openstack Rally benchmark test cycle against neutron. An > example scenario is creating 10x networks, list them, delete them and > repeat 10x times. The second set performs an L3 benchmark test betwee= n > two instances." > Those 300 routers will each have at least one namespace along with the=20 dhcp namespaces. Depending on the nature of the routers (Distributed=20 versus Centralized Virtual Routers - DVR vs CVR) and whether the router= s=20 are supposed to be "HA" there can be more than one namespace for a give= n=20 router. 300 routers is far from the upper limit/goal. Back in HP Public Cloud,= =20 we were running as many as 700 routers per network node (*), and more=20 than four network nodes. (back then it was just the one namespace per=20 router and network). Mileage will of course vary based on the "oomph" o= f=20 one's network node(s). happy benchmarking, rick jones * Didn't want to go much higher than that because each router had a por= t=20 on a common linux bridge and getting to > 1024 would be an unpleasant d= ay.