From owner-cvs-all Sun Sep 30 22: 6: 1 2001 Delivered-To: cvs-all@freebsd.org Received: from elvis.mu.org (elvis.mu.org [216.33.66.196]) by hub.freebsd.org (Postfix) with ESMTP id B73C437B406; Sun, 30 Sep 2001 22:05:56 -0700 (PDT) Received: by elvis.mu.org (Postfix, from userid 1192) id A300F81D01; Mon, 1 Oct 2001 00:05:56 -0500 (CDT) Date: Mon, 1 Oct 2001 00:05:56 -0500 From: Alfred Perlstein To: Matt Dillon Cc: cvs-committers@FreeBSD.ORG, cvs-all@FreeBSD.ORG Subject: Re: cvs commit: src/sys/kern vfs_cache.c vfs_subr.c src/sys/sys vnode.h Message-ID: <20011001000556.T59854@elvis.mu.org> References: <200110010433.f914XZd11860@freefall.freebsd.org> <20010930234627.R59854@elvis.mu.org> <200110010458.f914wlZ47428@earth.backplane.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <200110010458.f914wlZ47428@earth.backplane.com>; from dillon@earth.backplane.com on Sun, Sep 30, 2001 at 09:58:47PM -0700 Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG * Matt Dillon [010930 23:58] wrote: > : > :Is _this_ the one that was messing up that guy where we kept running > :out of vnodes? :) > : > :-- > :-Alfred Perlstein [alfred@freebsd.org] > > Which guy? Take your pick, there are at least three people with the > 'vnodes ran into a wall' problem (I'm including Yahoo here, which has > some rsync machines that still run into the problem even with all the > patches). The guy at Netzero, I've lost contact and I think he was being pressured for a solution so he's probably not using FreeBSD any longer... :( > I've been trying to figure out a simple way to fix it but it looks like > I am going to have to implement another vnode list, similar to the > tobefree list, that holds vnodes with no references but which still > have VM pages. Then getnewvnode() can try to free the underlying pages > if it exceeds kern.maxvnodes ('desiredvnodes' in the source). This confuses me, shouldn't the syncer be cleaning these vnodes? Are you saying that VOP_FSYNC somehow fails when there's fragmented buffers or something? -- -Alfred Perlstein [alfred@freebsd.org] 'Instead of asking why a piece of software is using "1970s technology," start asking why software is ignoring 30 years of accumulated wisdom.' To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe cvs-all" in the body of the message