From: Greg Banks Subject: Re: [PATCH 000 of 11] knfsd: NUMAisation Date: Thu, 27 Jul 2006 10:47:00 +1000 Message-ID: <1153961220.21040.1648.camel@hole.melbourne.sgi.com> References: <1153805274.21040.38.camel@hole.melbourne.sgi.com> <17605.50325.842862.8823@cse.unsw.edu.au> <76bd70e30607250705h3e8c9eefjc794397e0c45a21b@mail.gmail.com> <76bd70e30607250836v35110d6eo20043dccc4d284b7@mail.gmail.com> <1153895119.21040.494.camel@hole.melbourne.sgi.com> <76bd70e30607261126s28003af9neae6366f001fefe3@mail.gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Cc: Neil Brown , Paul Jimenez , Linux NFS Mailing List Return-path: Received: from sc8-sf-mx2-b.sourceforge.net ([10.3.1.92] helo=mail.sourceforge.net) by sc8-sf-list2-new.sourceforge.net with esmtp (Exim 4.43) id 1G64q7-00029v-Vz for nfs@lists.sourceforge.net; Thu, 27 Jul 2006 05:19:44 -0700 Received: from externalmx-1.sourceforge.net ([12.152.184.25]) by mail.sourceforge.net with esmtps (TLSv1:AES256-SHA:256) (Exim 4.44) id 1G64q1-0004iJ-NQ for nfs@lists.sourceforge.net; Thu, 27 Jul 2006 05:19:39 -0700 Received: from omx2-ext.sgi.com ([192.48.171.19] helo=omx2.sgi.com) by externalmx-1.sourceforge.net with esmtp (Exim 4.41) id 1G5u1x-0005W0-EW for nfs@lists.sourceforge.net; Wed, 26 Jul 2006 17:47:13 -0700 To: Chuck Lever In-Reply-To: <76bd70e30607261126s28003af9neae6366f001fefe3@mail.gmail.com> List-Id: "Discussion of NFS under Linux development, interoperability, and testing." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: nfs-bounces@lists.sourceforge.net Errors-To: nfs-bounces@lists.sourceforge.net On Thu, 2006-07-27 at 04:26, Chuck Lever wrote: > On 7/26/06, Greg Banks wrote: > > On Wed, 2006-07-26 at 01:36, Chuck Lever wrote: > > > On 7/25/06, Paul Jimenez wrote: > > > > > > > My vote is to zap the least recently used connection. That is overall > > > the safest compromise. > > > > Agreed. The following patch (on top of my earlier patch 001) does > > that. I've verified that it enforces the connection limit and > > doesn't kill the oldest socket. > > Looks like you're only dropping sockets that aren't registered with > the local portmapper. Yes, I've preserved that logic. > Should you do something similar for the > permanent sockets, No. > or am I missing something? At this time there are two permanent sockets: the UDP socket and the TCP rendezvous socket (i.e. the one in LISTEN state). Neither of these is ever connected; neither should they ever be dropped (because they're holding their respective ports, and dropping them might allow a userspace program to bind to those ports, which would be a security issue). Greg. -- Greg Banks, R&D Software Engineer, SGI Australian Software Group. I don't speak for SGI. ------------------------------------------------------------------------- Take Surveys. Earn Cash. Influence the Future of IT Join SourceForge.net's Techsay panel and you'll get the chance to share your opinions on IT & business topics through brief surveys -- and earn cash http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV _______________________________________________ NFS maillist - NFS@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/nfs