From owner-freebsd-current Sat Apr 19 07:41:38 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id HAA19463 for current-outgoing; Sat, 19 Apr 1997 07:41:38 -0700 (PDT) Received: from nlsystems.com (nlsys.demon.co.uk [158.152.125.33]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id HAA19455 for ; Sat, 19 Apr 1997 07:41:30 -0700 (PDT) Received: from herring.nlsystems.com (herring.nlsystems.com [10.0.0.2]) by nlsystems.com (8.8.5/8.8.5) with SMTP id PAA01107; Sat, 19 Apr 1997 15:40:58 +0100 (BST) Date: Sat, 19 Apr 1997 15:40:58 +0100 (BST) From: Doug Rabson To: Karl Denninger cc: current@freebsd.org Subject: Re: Danger Will Robinson In-Reply-To: <19970419093254.63764@Jupiter.Mcs.Net> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk On Sat, 19 Apr 1997, Karl Denninger wrote: > On Sat, Apr 19, 1997 at 03:31:09PM +0100, Doug Rabson wrote: > > On Sat, 19 Apr 1997, Karl Denninger wrote: > > > > > > > > It is not NFS specific. > > > > > > Try running a web server on -CURRENT using httpd 1.5.2. > > > > > > The kernel will blow chunks within minutes with a "page not present" fault. > > > > > > This is true whether the executable is on a LOCAL disk or over NFS. > > > > > > The -CURRENT kernel is NOT stable in this sort of environment. > > > > In that case, none of my NFS fixes will even come close. I assume you are > > working closely with John and David to diagnose this? > > > > -- > > Doug Rabson Mail: dfr@nlsystems.com > > Nonlinear Systems Ltd. Phone: +44 181 951 1891 > > John has a core dump and kernel file showing the problem. > > David has not yet contacted me. > > Can you delineate the NFS fixes and what they are expected to correct? Most recently, I fixed some race conditions when vnodes were recycled from NFS to UFS. These showed up as panics when nfs_inactive ended up calling ufs_unlock. I also fixed a couple of problems with faults and odd behaviour for programs writing lots of small blocks over NFS. Your venerable log file problem where many processes are appending to the same log file is (probably) not fixed. I may work on that one soon. -- Doug Rabson Mail: dfr@nlsystems.com Nonlinear Systems Ltd. Phone: +44 181 951 1891