From owner-freebsd-current Mon Mar 10 14:41:53 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id OAA04196 for current-outgoing; Mon, 10 Mar 1997 14:41:53 -0800 (PST) Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.50]) by freefall.freebsd.org (8.8.5/8.8.5) with SMTP id OAA04191 for ; Mon, 10 Mar 1997 14:41:49 -0800 (PST) Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id PAA23787; Mon, 10 Mar 1997 15:33:53 -0700 From: Terry Lambert Message-Id: <199703102233.PAA23787@phaeton.artisoft.com> Subject: Re: Yikes! Current / Lite2 kernel has MAJOR NFS client problems To: karl@Mcs.Net (Karl Denninger) Date: Mon, 10 Mar 1997 15:33:53 -0700 (MST) Cc: current@freebsd.org In-Reply-To: <19970309145605.07984@Jupiter.Mcs.Net> from "Karl Denninger" at Mar 9, 97 02:56:05 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Be EXTREMELY careful if you try to run a LITE2 kernel in an NFS environment. > > As it sits right now, you'll get tons of panics out of "page not present" > faults in the event any executable is grabbed over NFS. > > I'll start looking at this if someone has an idea where it would be coming > from. Is this a "dirty page" problem in that the page isn't being fetched > when it should be? Look at the NFS differences for the pre-merge code vs. the Lite code that should be the first rev in the tree. Basically, the cache unification process has changes the FS/buffer interface, and you need to propagate some stuff down to the FS that NFS is a client of for it to be truly happy. The diff should pinpoint it for you. I thought John Dyson said he was already looking at this... avoid his toes if possible, I think... Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.