From owner-devfs@oss.sgi.com Sun Apr 1 09:12:15 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f31GCFe13017 for devfs-outgoing; Sun, 1 Apr 2001 09:12:15 -0700 Received: from cmailg3.svr.pol.co.uk (cmailg3.svr.pol.co.uk [195.92.195.173]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f31GCDM13014 for ; Sun, 1 Apr 2001 09:12:14 -0700 Received: from modem-219.fire-goby.dialup.pol.co.uk ([62.137.11.219]) by cmailg3.svr.pol.co.uk with esmtp (Exim 3.13 #0) id 14jkSZ-000094-00 for devfs@oss.sgi.com; Sun, 01 Apr 2001 17:12:12 +0100 Date: Sun, 1 Apr 2001 17:14:54 +0100 (BST) From: Will Newton X-X-Sender: To: Subject: Temp files and symlinks Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-devfs@oss.sgi.com Precedence: bulk In my persistance storage directory (/dev-state) I get some temporary files. I would be grateful if someone could tell me if they should be there or not. They are of the form: dri-temp-1002 and fd0u1040 relating to DRI and floppy drivers respectively. Also symlinks are stored, which seems pointless as they are created by devfsd, and when I change the config I want them to disappear. From owner-devfs@oss.sgi.com Mon Apr 2 07:55:05 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f32Et5j16235 for devfs-outgoing; Mon, 2 Apr 2001 07:55:05 -0700 Received: from cmailg7.svr.pol.co.uk (cmailg7.svr.pol.co.uk [195.92.195.177]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f32Et4M16232 for ; Mon, 2 Apr 2001 07:55:04 -0700 Received: from modem-141.blue-star-damsel.dialup.pol.co.uk ([62.136.240.141]) by cmailg7.svr.pol.co.uk with esmtp (Exim 3.13 #0) id 14k5jS-0005TP-00 for devfs@oss.sgi.com; Mon, 02 Apr 2001 15:55:02 +0100 Date: Mon, 2 Apr 2001 15:57:45 +0100 (BST) From: Will Newton X-X-Sender: To: Subject: devfs warning messages Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-devfs@oss.sgi.com Precedence: bulk Can anyone help me understand what's going on here: devfs: devfs_register(): device already registered: "2" devfs: devfs_register(): device already registered: "a2" devfs: devfs_register(): device already registered: "3" devfs: devfs_register(): device already registered: "a3" Power management code maybe? I get these messages on every boot. From owner-devfs@oss.sgi.com Thu Apr 5 01:48:39 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f358md811051 for devfs-outgoing; Thu, 5 Apr 2001 01:48:39 -0700 Received: from web8003.mail.in.yahoo.com (web8003.in.yahoo.com [203.199.70.21]) by oss.sgi.com (8.11.3/8.11.3) with SMTP id f358mbM11046 for ; Thu, 5 Apr 2001 01:48:38 -0700 Message-ID: <20010405085127.11966.qmail@web8003.mail.in.yahoo.com> Received: from [203.197.143.2] by web8003.mail.in.yahoo.com; Thu, 05 Apr 2001 09:51:27 BST Date: Thu, 5 Apr 2001 09:51:27 +0100 (BST) From: =?iso-8859-1?q?Karthikeyan=20Nagaraj?= Subject: Re: devfs warning messages To: devfs@oss.sgi.com In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Sender: owner-devfs@oss.sgi.com Precedence: bulk HI, I dont know exactly what is going on but I beleive that the kernel has support for /dev file system... When u r compiling a kernel there is a option of selecting it in the section of filesystem... Try disabling it adn then try the new kernel... Might work out though I am not sure of it.... If I am sounding too stupid dont fire at me just press delete .... thanks for bearing with me..... Thanks & rgds, Karthik --- Will Newton wrote: > > Can anyone help me understand what's going on here: > > devfs: devfs_register(): device already registered: > "2" > devfs: devfs_register(): device already registered: > "a2" > devfs: devfs_register(): device already registered: > "3" > devfs: devfs_register(): device already registered: > "a3" > > Power management code maybe? > > I get these messages on every boot. > > ____________________________________________________________ Do You Yahoo!? For regular News updates go to http://in.news.yahoo.com From owner-devfs@oss.sgi.com Thu Apr 5 22:32:33 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f365WXQ16615 for devfs-outgoing; Thu, 5 Apr 2001 22:32:33 -0700 Received: from pervalidus.dyndns.org (exim@200-191-140-143-as.acessonet.com.br [200.191.140.143]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f365WFM16612 for ; Thu, 5 Apr 2001 22:32:31 -0700 Received: from pervalidus by pervalidus.dyndns.org with local (Exim 3.22 #1) id 14lOrH-00028D-00 for devfs@oss.sgi.com; Fri, 06 Apr 2001 02:32:31 -0300 Date: Fri, 6 Apr 2001 02:32:31 -0300 From: =?iso-8859-1?B?RnLpZOlyaWMgTC4gVy4=?= Meunier <0@pervalidus.net> To: devfs@oss.sgi.com Subject: error calling: "symlink" in "GLOBAL" Message-ID: <20010406023231.W111@pervalidus> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.3.16i X-Mailer: Mutt/1.3.16i - Linux 2.4.3 X-URL: http://www.pervalidus.net/ Sender: owner-devfs@oss.sgi.com Precedence: bulk Hi. I'm using #REGISTER vc/.* MKOLDCOMPAT #UNREGISTER vc/.* RMOLDCOMPAT and want to create some symlinks instead. Using REGISTER vc/1 CFUNCTION GLOBAL symlink vc/1 tty1 UNREGISTER vc/1 CFUNCTION GLOBAL unlink tty1 works but I get the following messages: Started device management daemon for /dev error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" error calling: "symlink" in "GLOBAL" What's wrong ? I'm a bit confused about the devfsd man page syntax for symlinks: REGISTER mydir/mydev CFUNCTION GLOBAL symlink $devname mydev UNREGISTER mydir/mydev CFUNCTION GLOBAL unlink mydev -- 0@pervalidus.{net, {dyndns.}org} Tel: 55-21-717-2399 (Niterói-RJ BR) From owner-devfs@oss.sgi.com Fri Apr 6 06:27:34 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f36DRYv27875 for devfs-outgoing; Fri, 6 Apr 2001 06:27:34 -0700 Received: from notes.informed.net (charcot.informed.net [205.167.80.72]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f36DRWM27872 for ; Fri, 6 Apr 2001 06:27:33 -0700 Subject: FYI: Oracle Install To: devfs@oss.sgi.com X-Mailer: Lotus Notes Release 5.0.4a July 24, 2000 Message-ID: From: mclinden@informed.net Date: Fri, 6 Apr 2001 08:35:37 -0400 X-MIMETrack: Serialize by Router on notes.informed.net/informed dot net(Release 5.0.5 |September 22, 2000) at 04/06/2001 08:35:35 AM MIME-Version: 1.0 Content-type: text/plain; charset=us-ascii Sender: owner-devfs@oss.sgi.com Precedence: bulk This is slightly off-topic but may be important to some users. I have confirmed that devfs enabled Linux systems may exhibit problems related to the installation of Oracle 8i for Linux. Apparently the problem is related to the Oracle Installer's inability to properly parse the output of the df(1) command in order to calculate available disk space (I don't have the source code for the Installer). The identical install with devfs disabled has no problems. There does NOT appear to be a problem enabling devfs after installing Oracle, however. Sean McLinden Outcome Technology Associates, Inc. From owner-devfs@oss.sgi.com Sat Apr 7 22:26:38 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f385Qcl04929 for devfs-outgoing; Sat, 7 Apr 2001 22:26:38 -0700 Received: from mailin8.bigpond.com (juicer39.bigpond.com [139.134.6.96]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f385QaM04926 for ; Sat, 7 Apr 2001 22:26:37 -0700 Received: from mobilix.atnf.CSIRO.AU ([139.134.4.56]) by mailin8.bigpond.com (Netscape Messaging Server 4.15) with SMTP id GBGKOP00.FM9; Sun, 8 Apr 2001 15:31:37 +1000 Received: from CPE-61-9-168-62.vic.bigpond.net.au ([61.9.168.62]) by mail3.bigpond.com (Claudes-Interim-MailRouter V2.9c 5/3401762); 08 Apr 2001 15:26:17 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.10.0/8.10.0) id f385QU906141; Sun, 8 Apr 2001 15:26:30 +1000 Date: Sun, 8 Apr 2001 15:26:30 +1000 Message-Id: <200104080526.f385QU906141@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: =?iso-8859-1?B?RnLpZOlyaWMgTC4gVy4=?= Meunier <0@pervalidus.net> Cc: devfs@oss.sgi.com Subject: Re: error calling: "symlink" in "GLOBAL" In-Reply-To: <20010406023231.W111@pervalidus> References: <20010406023231.W111@pervalidus> Sender: owner-devfs@oss.sgi.com Precedence: bulk 0@pervalidus.net writes: > Hi. I'm using > > #REGISTER vc/.* MKOLDCOMPAT > #UNREGISTER vc/.* RMOLDCOMPAT > > and want to create some symlinks instead. Using > > REGISTER vc/1 CFUNCTION GLOBAL symlink vc/1 tty1 > UNREGISTER vc/1 CFUNCTION GLOBAL unlink tty1 > > works but I get the following messages: > > Started device management daemon for /dev > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > error calling: "symlink" in "GLOBAL" > > What's wrong ? I'm a bit confused about the devfsd man page > syntax for symlinks: > > REGISTER mydir/mydev CFUNCTION GLOBAL symlink $devname mydev > UNREGISTER mydir/mydev CFUNCTION GLOBAL unlink mydev What version of devfsd are you running? Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Sun Apr 8 07:30:56 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f38EUu313780 for devfs-outgoing; Sun, 8 Apr 2001 07:30:56 -0700 Received: from pervalidus.dyndns.org (200-191-140-22-as.acessonet.com.br [200.191.140.22]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f38EUrM13777 for ; Sun, 8 Apr 2001 07:30:54 -0700 Received: from pervalidus by pervalidus.dyndns.org with local (Exim 3.22 #1) id 14mGDE-0002SF-00; Sun, 08 Apr 2001 11:30:44 -0300 Date: Sun, 8 Apr 2001 11:30:44 -0300 From: =?iso-8859-1?B?RnLpZOlyaWMgTC4gVy4=?= Meunier <0@pervalidus.net> To: Richard Gooch Cc: devfs@oss.sgi.com Subject: Re: error calling: "symlink" in "GLOBAL" Message-ID: <20010408113044.I111@pervalidus.net> References: <20010406023231.W111@pervalidus> <200104080526.f385QU906141@mobilix.atnf.CSIRO.AU> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.3.16i In-Reply-To: <200104080526.f385QU906141@mobilix.atnf.CSIRO.AU>; from rgooch@atnf.csiro.au on Sun, Apr 08, 2001 at 03:26:30PM +1000 X-Mailer: Mutt/1.3.16i - Linux 2.4.3 X-URL: http://www.pervalidus.net/ Sender: owner-devfs@oss.sgi.com Precedence: bulk Hi Richard. I'm using devfsd 1.3.11 on 2.4.3 Kernel. On Sun, Apr 08, 2001 at 03:26:30PM +1000, Richard Gooch wrote: > 0@pervalidus.net writes: > > Hi. I'm using > > > > #REGISTER vc/.* MKOLDCOMPAT > > #UNREGISTER vc/.* RMOLDCOMPAT > > > > and want to create some symlinks instead. Using > > > > REGISTER vc/1 CFUNCTION GLOBAL symlink vc/1 tty1 > > UNREGISTER vc/1 CFUNCTION GLOBAL unlink tty1 > > > > works but I get the following messages: > > > > Started device management daemon for /dev > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > error calling: "symlink" in "GLOBAL" > > > > What's wrong ? I'm a bit confused about the devfsd man page > > syntax for symlinks: > > > > REGISTER mydir/mydev CFUNCTION GLOBAL symlink $devname mydev > > UNREGISTER mydir/mydev CFUNCTION GLOBAL unlink mydev > > What version of devfsd are you running? -- 0@pervalidus.{net, {dyndns.}org} Tel: 55-21-717-2399 (Niterói-RJ BR) From owner-devfs@oss.sgi.com Thu Apr 12 15:30:24 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3CMUOb22448 for devfs-outgoing; Thu, 12 Apr 2001 15:30:24 -0700 Received: from geneva.vmware.com (geneva.vmware.com [63.93.12.3]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3CMUOM22445 for ; Thu, 12 Apr 2001 15:30:24 -0700 Received: from lester.vmware.com (lester.vmware.com [172.16.15.251]) by geneva.vmware.com (8.9.3/8.9.3) with ESMTP id PAA27316 for ; Thu, 12 Apr 2001 15:30:23 -0700 Received: by lester.vmware.com (Postfix, from userid 865) id 9BB6E7288; Thu, 12 Apr 2001 15:30:23 -0700 (PDT) To: devfs@oss.sgi.com Subject: using modules +/- CONFIG_DEVFS_FS Message-Id: <20010412223023.9BB6E7288@lester.vmware.com> Date: Thu, 12 Apr 2001 15:30:23 -0700 (PDT) From: rmanches@vmware.com (Robert Manchester) Sender: owner-devfs@oss.sgi.com Precedence: bulk if I compile a module which uses devfs I would like it to work on kernels with and without devfs enabled. Currently, (kernel 2.4.3) the devfs_fs_kernel.h has two different implementaions of the devfs_* functions. This seems wrong to me, as the implemetaion differences should not be in the header file. I would like to compile one module for kernel version xxx have it work on both devfs and non devfs systems. Is there something inherently wrong in doing this that I am missing? -rob From owner-devfs@oss.sgi.com Thu Apr 12 15:57:22 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3CMvMj23357 for devfs-outgoing; Thu, 12 Apr 2001 15:57:22 -0700 Received: from mailgate.rz.uni-karlsruhe.de (mailgate.rz.uni-karlsruhe.de [129.13.64.97]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3CMvLM23354 for ; Thu, 12 Apr 2001 15:57:22 -0700 Received: from nce2.hadiko.de (hadince2.hadiko.uni-karlsruhe.de [172.20.32.2]) by mailgate.rz.uni-karlsruhe.de with esmtp (Exim 3.16 #1) id 14nq1b-0003Fk-00; Fri, 13 Apr 2001 00:57:15 +0200 Received: from panorama.hadiko.de (hadii309.hadiko.uni-karlsruhe.de [172.20.44.69]) by nce2.hadiko.de (8.9.3/8.9.3) with SMTP id AAA22655 for ; Fri, 13 Apr 2001 00:57:09 +0200 (MET DST) Received: (qmail 24049 invoked from network); 12 Apr 2001 22:57:02 -0000 Received: from localhost (127.0.0.1) by localhost with SMTP; 12 Apr 2001 22:57:02 -0000 To: rmanches@vmware.com Cc: devfs@oss.sgi.com Subject: Re: using modules +/- CONFIG_DEVFS_FS From: Robert Siemer In-Reply-To: <20010412223023.9BB6E7288@lester.vmware.com> References: <20010412223023.9BB6E7288@lester.vmware.com> X-Mailer: Mew version 1.94b25 on Emacs 20.5 / Mule 4.0 (HANANOEN) Reply-To: Robert Siemer Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: <20010413005658I.siemer@panorama.hadiko.de> Date: Fri, 13 Apr 2001 00:56:58 +0200 X-Dispatcher: imput version 990425(IM115) Lines: 12 Sender: owner-devfs@oss.sgi.com Precedence: bulk From: rmanches@vmware.com (Robert Manchester) > I would like to compile one module for kernel version xxx have it work > on both devfs and non devfs systems. Is there something inherently > wrong in doing this that I am missing? Kernel modules are not intended to work for different kernel configurations (even in the same kernel version). Bye, Robert From owner-devfs@oss.sgi.com Thu Apr 12 16:29:16 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3CNTGp24057 for devfs-outgoing; Thu, 12 Apr 2001 16:29:16 -0700 Received: from geneva.vmware.com (geneva.vmware.com [63.93.12.3]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3CNTFM24054 for ; Thu, 12 Apr 2001 16:29:15 -0700 Received: from lester.vmware.com (lester.vmware.com [172.16.15.251]) by geneva.vmware.com (8.9.3/8.9.3) with ESMTP id QAA02299 for ; Thu, 12 Apr 2001 16:29:15 -0700 Received: by lester.vmware.com (Postfix, from userid 865) id 9811F7288; Thu, 12 Apr 2001 16:29:15 -0700 (PDT) Date: Thu, 12 Apr 2001 16:29:15 -0700 From: Robert Manchester To: devfs@oss.sgi.com Subject: Re: using modules +/- CONFIG_DEVFS_FS Message-ID: <20010412162915.A7239@lester.vmware.com> References: <20010412223023.9BB6E7288@lester.vmware.com> <20010413005658I.siemer@panorama.hadiko.de> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010413005658I.siemer@panorama.hadiko.de>; from Robert.Siemer@gmx.de on Fri, Apr 13, 2001 at 12:56:58AM +0200 Sender: owner-devfs@oss.sgi.com Precedence: bulk * Robert Siemer (Robert.Siemer@gmx.de) [010412 15:59]: > From: rmanches@vmware.com (Robert Manchester) > > > I would like to compile one module for kernel version xxx have it work > > on both devfs and non devfs systems. Is there something inherently > > wrong in doing this that I am missing? > > Kernel modules are not intended to work for different kernel > configurations (even in the same kernel version). How am I supposed to release binary modules if they only work on the machine that I compiled them on? -rob From owner-devfs@oss.sgi.com Thu Apr 12 17:13:35 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3D0DZQ25368 for devfs-outgoing; Thu, 12 Apr 2001 17:13:35 -0700 Received: from pneumatic-tube.sgi.com (pneumatic-tube.sgi.com [204.94.214.22]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3D0DYM25365 for ; Thu, 12 Apr 2001 17:13:34 -0700 Received: from miine.engr.sgi.com (miine.engr.sgi.com [163.154.6.142]) by pneumatic-tube.sgi.com (980327.SGI.8.8.8-aspam/980310.SGI-aspam) via ESMTP id RAA01678 for ; Thu, 12 Apr 2001 17:23:47 -0700 (PDT) mail_from (mee@miine.engr.sgi.com) Received: (from mee@localhost) by miine.engr.sgi.com (SGI-8.9.3/8.9.3) id RAA12976; Thu, 12 Apr 2001 17:09:47 -0700 (PDT) Date: Thu, 12 Apr 2001 17:09:47 -0700 From: Jeremy Brown To: Robert Manchester Cc: devfs@oss.sgi.com Subject: Re: using modules +/- CONFIG_DEVFS_FS Message-ID: <20010412170947.T74774@miine.engr.sgi.com> References: <20010412223023.9BB6E7288@lester.vmware.com> <20010413005658I.siemer@panorama.hadiko.de> <20010412162915.A7239@lester.vmware.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010412162915.A7239@lester.vmware.com>; from rmanches@vmware.com on Thu, Apr 12, 2001 at 04:29:15PM -0700 Sender: owner-devfs@oss.sgi.com Precedence: bulk To solve this problem, I would suggest looking at NVIDIA's Linux driver on their site (www.nvidia.com). They distribute binary-only modules for common distributions, as well as a source rpm that contains open-source wrapper code and a binary-only library that is compiled into a kernel module. Of course, this whole problem goes away if your driver is open-source. :) Jeremy Brown SGI On Thu, Apr 12, 2001 at 04:29:15PM -0700, Robert Manchester wrote: > * Robert Siemer (Robert.Siemer@gmx.de) [010412 15:59]: > > From: rmanches@vmware.com (Robert Manchester) > > > > > I would like to compile one module for kernel version xxx have it work > > > on both devfs and non devfs systems. Is there something inherently > > > wrong in doing this that I am missing? > > > > Kernel modules are not intended to work for different kernel > > configurations (even in the same kernel version). > > How am I supposed to release binary modules if they only work on the > machine that I compiled them on? > > -rob From owner-devfs@oss.sgi.com Thu Apr 12 17:30:07 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3D0U7s25699 for devfs-outgoing; Thu, 12 Apr 2001 17:30:07 -0700 Received: from geneva.vmware.com (geneva.vmware.com [63.93.12.3]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3D0U6M25696 for ; Thu, 12 Apr 2001 17:30:06 -0700 Received: from lester.vmware.com (lester.vmware.com [172.16.15.251]) by geneva.vmware.com (8.9.3/8.9.3) with ESMTP id RAA09300 for ; Thu, 12 Apr 2001 17:30:05 -0700 Received: by lester.vmware.com (Postfix, from userid 865) id D1BDC7288; Thu, 12 Apr 2001 17:30:05 -0700 (PDT) Date: Thu, 12 Apr 2001 17:30:05 -0700 From: Robert Manchester To: devfs@oss.sgi.com Subject: Re: using modules +/- CONFIG_DEVFS_FS Message-ID: <20010412173005.A12675@lester.vmware.com> References: <20010412223023.9BB6E7288@lester.vmware.com> <20010413005658I.siemer@panorama.hadiko.de> <20010412162915.A7239@lester.vmware.com> <20010412170947.T74774@miine.engr.sgi.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: <20010412170947.T74774@miine.engr.sgi.com>; from mee@miine.engr.sgi.com on Thu, Apr 12, 2001 at 05:09:47PM -0700 Sender: owner-devfs@oss.sgi.com Precedence: bulk * Jeremy Brown (mee@miine.engr.sgi.com) [010412 17:12]: > Of course, this whole problem goes away if your driver is open-source. :) Not really. The reason I would like to have a binary module is so my users don't have to know how to compile it. So many people have problems getting the right header files, compiler, etc, that it is easier for them if everything just works. Distributing the module as a binary doesn't mean that I don't want to distribute the source code with it, I just want to minimize the user from screwing up and ultimately being turned off from running my code because it is too hard to install. another reason would be I want to run two 2.4.3 kernels one with devfs and one without devfs using the same modules for both kernels. -rob From owner-devfs@oss.sgi.com Thu Apr 12 18:31:41 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3D1VfD27728 for devfs-outgoing; Thu, 12 Apr 2001 18:31:41 -0700 Received: from mailin1.bigpond.com (juicer13.bigpond.com [139.134.6.21]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3D1VdM27721 for ; Thu, 12 Apr 2001 18:31:39 -0700 Received: from mobilix.atnf.CSIRO.AU ([139.134.4.58]) by mailin1.bigpond.com (Netscape Messaging Server 4.15) with SMTP id GBPJ4R00.J10; Fri, 13 Apr 2001 11:36:27 +1000 Received: from CPE-61-9-170-29.vic.bigpond.net.au ([61.9.170.29]) by mail1.bigpond.com (Claudes-Wiggly-MailRouter V2.9c 1/1416224); 13 Apr 2001 11:30:56 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.10.0/8.10.0) id f3D1VIE03330; Fri, 13 Apr 2001 11:31:18 +1000 Date: Fri, 13 Apr 2001 11:31:18 +1000 Message-Id: <200104130131.f3D1VIE03330@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Robert Manchester Cc: devfs@oss.sgi.com Subject: Re: using modules +/- CONFIG_DEVFS_FS In-Reply-To: <20010412173005.A12675@lester.vmware.com> References: <20010412223023.9BB6E7288@lester.vmware.com> <20010413005658I.siemer@panorama.hadiko.de> <20010412162915.A7239@lester.vmware.com> <20010412170947.T74774@miine.engr.sgi.com> <20010412173005.A12675@lester.vmware.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Robert Manchester writes: > * Jeremy Brown (mee@miine.engr.sgi.com) [010412 17:12]: > > Of course, this whole problem goes away if your driver is open-source. :) > > Not really. The reason I would like to have a binary module is so my > users don't have to know how to compile it. So many people have > problems getting the right header files, compiler, etc, that it is > easier for them if everything just works. Distributing the module as a > binary doesn't mean that I don't want to distribute the source code with > it, I just want to minimize the user from screwing up and ultimately > being turned off from running my code because it is too hard to install. > > another reason would be I want to run two 2.4.3 kernels one with devfs > and one without devfs using the same modules for both kernels. It's not just devfs where you'll have this problem. It's a range of other CONFIG_ options. The obvious one is CONFIG_SMP. That's just doubled the number of binaries you need to ship. Every config option that changes the driver ABI will multiply the number of binaries you need to ship. You're better off giving up on trying to ship binaries for all/most configurations, and instead ship source code. Or at least a source wrapper to a binary library (if your management is still under the illusion that you have IP worth hiding:-). Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Sat Apr 14 10:49:41 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3EHnf326269 for devfs-outgoing; Sat, 14 Apr 2001 10:49:41 -0700 Received: from mail15.jump.net (mail15.jump.net [206.196.91.15]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3EHneM26266 for ; Sat, 14 Apr 2001 10:49:40 -0700 Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26]) by mail15.jump.net (8.10.2/) with ESMTP id f3EHnch01369 for ; Sat, 14 Apr 2001 12:49:38 -0500 (CDT) Message-ID: <3AD88E44.BBE8A438@sgi.com> Date: Sat, 14 Apr 2001 12:52:04 -0500 From: Eric Sandeen X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: making /dev/cdrom work transparently Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Ok, before you flame me for hitting this topic _again_, I understand that there are many problems with links like /dev/cdrom and /dev/mouse. But right now our biggest problem is users mailing the linux-xfs list after they use our installer, complaining "Hey! There's no /dev/cdrom or /dev/mouse!" :) So I thought I'd embark on an experiment of making /dev/cdrom show up automagically if that's what the user really wants. The other constraint I've imposed upon myself is to do it from within the devfsd package, rather than relying on initscripts, so that our installer doesn't have to contain another modified version of a standard Red Hat package (initscripts) that will break if they upgrade from Red Hat. So here's what I've tried, maybe somebody can tell me where I went wrong. It works most of the time, but after some unmounts or module unloads, it breaks. The plan is that if someone tries to access /dev/cdrom, it will: 1) load the necessary modules 2) create a link from /dev/cdrom to /dev/cdroms/cdromX where X is the lowest available number. First, I added this line to devfsd.conf, so that when a user tries to access /dev/cdrom, appropriate action will be taken: LOOKUP cdrom EXECUTE /etc/rc.d/rc.devfs-cdrom and then clean up after it on unload: UNREGISTER cdroms/.* CFUNCTION GLOBAL unlink cdrom Then, I added this to /etc/modules.conf, so that devfsd module autoloading will work for /dev/cdrom: alias /dev/cdrom* /dev/cdroms And finally, here's the script that gets EXECUTEd from devfsd.conf: #!/bin/bash # # /etc/rc.devfs-cdrom - run by devfsd when something looks for /dev/cdrom rm -f /dev/cdrom # in case of old links if [ -c /dev/.devfsd -a -x /sbin/devfsd ]; then for i in /dev/cdroms/*; do ln -sf $i /dev/ echo $i if [ ! -x /dev/cdrom ]; then ln -sf $i /dev/cdrom # link the first cd fi done fi # End of script But, this doesn't always work. Any suggestions? One more quick question - why does X in /dev/cdroms/cdromX get incremented each time the ide-cd module is reloaded? Is that the expected behaviour? Thanks, -Eric -- Eric Sandeen XFS for Linux http://oss.sgi.com/projects/xfs sandeen@sgi.com SGI, Inc. From owner-devfs@oss.sgi.com Sat Apr 14 14:41:36 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3ELfag31902 for devfs-outgoing; Sat, 14 Apr 2001 14:41:36 -0700 Received: from mailin9.bigpond.com (juicer34.bigpond.com [139.134.6.86]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3ELfYM31899 for ; Sat, 14 Apr 2001 14:41:35 -0700 Received: from mobilix.atnf.CSIRO.AU ([139.134.4.54]) by mailin9.bigpond.com (Netscape Messaging Server 4.15) with SMTP id GBSXTI00.7RA; Sun, 15 Apr 2001 07:46:30 +1000 Received: from CPE-61-9-170-29.vic.bigpond.net.au ([61.9.170.29]) by mail6.bigpond.com (Claudes-Emblazoned-MailRouter V2.9c 11/2109046); 15 Apr 2001 07:40:33 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.10.0/8.10.0) id f3ELfMq00437; Sun, 15 Apr 2001 07:41:22 +1000 Date: Sun, 15 Apr 2001 07:41:22 +1000 Message-Id: <200104142141.f3ELfMq00437@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Eric Sandeen Cc: devfs@oss.sgi.com Subject: Re: making /dev/cdrom work transparently In-Reply-To: <3AD88E44.BBE8A438@sgi.com> References: <3AD88E44.BBE8A438@sgi.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Eric Sandeen writes: > Ok, before you flame me for hitting this topic _again_, I understand > that there are many problems with links like /dev/cdrom and /dev/mouse. > But right now our biggest problem is users mailing the linux-xfs list > after they use our installer, complaining "Hey! There's no /dev/cdrom > or /dev/mouse!" :) > > So I thought I'd embark on an experiment of making /dev/cdrom show up > automagically if that's what the user really wants. The other > constraint I've imposed upon myself is to do it from within the devfsd > package, rather than relying on initscripts, so that our installer > doesn't have to contain another modified version of a standard Red Hat > package (initscripts) that will break if they upgrade from Red Hat. Talk to Tom Duffy and Jeremy Brown over in building #30. I talked to them about this just under two weeks ago when I visited. > One more quick question - why does X in /dev/cdroms/cdromX get > incremented each time the ide-cd module is reloaded? Is that the > expected behaviour? Known problem. It's due to an incementing counter. I'll be submitting a fix. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Sat Apr 14 14:53:14 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3ELrEq32121 for devfs-outgoing; Sat, 14 Apr 2001 14:53:14 -0700 Received: from sgi.com (sgi.SGI.COM [192.48.153.1]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3ELrEM32118 for ; Sat, 14 Apr 2001 14:53:14 -0700 Received: from cthulhu.engr.sgi.com (cthulhu.engr.sgi.com [192.26.80.2]) by sgi.com (980327.SGI.8.8.8-aspam/980304.SGI-aspam: SGI does not authorize the use of its proprietary systems or networks for unsolicited or bulk email from the Internet.) via ESMTP id OAA04023 for ; Sat, 14 Apr 2001 14:53:10 -0700 (PDT) mail_from (tduffy@engr.sgi.com) Received: from dbear.engr.sgi.com (dbear.engr.sgi.com [163.154.18.85]) by cthulhu.engr.sgi.com (SGI-8.9.3/8.9.3) with ESMTP id OAA05037; Sat, 14 Apr 2001 14:51:52 -0700 (PDT) Date: Sat, 14 Apr 2001 14:47:02 -0700 (PDT) From: Tom Duffy To: Richard Gooch cc: Eric Sandeen , Subject: Re: making /dev/cdrom work transparently In-Reply-To: <200104142141.f3ELfMq00437@mobilix.atnf.CSIRO.AU> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-devfs@oss.sgi.com Precedence: bulk > Talk to Tom Duffy and Jeremy Brown over in building #30. I talked to > them about this just under two weeks ago when I visited. right, eric, jeremy is currently working on fixing this (the correct way) for LBS stuff. he will have the fixes checked into that tree real soon now... but the basic idea is that you add a special entry into your devfsd.conf file (no need for the script). jeremy knows the exact syntax to put in. he is also on this list, so he can answer what to do for xfs 1.0 -tduffy From owner-devfs@oss.sgi.com Wed Apr 18 20:28:46 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3J3SkF05127 for devfs-outgoing; Wed, 18 Apr 2001 20:28:46 -0700 Received: from mail15.jump.net (mail15.jump.net [206.196.91.15]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3J3SiM05124 for ; Wed, 18 Apr 2001 20:28:44 -0700 Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26]) by mail15.jump.net (8.10.2/) with ESMTP id f3J3SiA28603 for ; Wed, 18 Apr 2001 22:28:44 -0500 (CDT) Message-ID: <3ADE5BBF.CEE82281@sgi.com> Date: Wed, 18 Apr 2001 22:30:07 -0500 From: Eric Sandeen X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: why mount --bind for permissions? Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk I was experimenting with saving permissions in /dev via devfsd, and I was wondering - why is mount --bind /dev-state /dev necessary? It seems that REGISTER .* COPY /dev-state/$devname $devpath CHANGE .* COPY $devpath /dev-state/$devname CREATE .* COPY $devpath /dev-state/$devname in devfsd.conf would copy everything back and forth whenever appropriate - I'm not that familiar with "mount --bind" - but what does it add to the equation? Sorry if I'm being dense. :) Thanks, -Eric -- Eric Sandeen XFS for Linux http://oss.sgi.com/projects/xfs sandeen@sgi.com SGI, Inc. From owner-devfs@oss.sgi.com Wed Apr 18 20:44:17 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3J3iHW05715 for devfs-outgoing; Wed, 18 Apr 2001 20:44:17 -0700 Received: from mobilix.atnf.CSIRO.AU (tinylinux.tip.CSIRO.AU [130.155.192.102]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3J3iFM05709 for ; Wed, 18 Apr 2001 20:44:15 -0700 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.10.0/8.10.0) id f3J3i3G00877; Thu, 19 Apr 2001 13:44:03 +1000 Date: Thu, 19 Apr 2001 13:44:03 +1000 Message-Id: <200104190344.f3J3i3G00877@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Eric Sandeen Cc: devfs@oss.sgi.com Subject: Re: why mount --bind for permissions? In-Reply-To: <3ADE5BBF.CEE82281@sgi.com> References: <3ADE5BBF.CEE82281@sgi.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Eric Sandeen writes: > I was experimenting with saving permissions in /dev via devfsd, and I > was wondering - why is > > mount --bind /dev-state /dev > > necessary? It seems that > > REGISTER .* COPY /dev-state/$devname $devpath > CHANGE .* COPY $devpath /dev-state/$devname > CREATE .* COPY $devpath /dev-state/$devname > > in devfsd.conf would copy everything back and forth whenever appropriate > - I'm not that familiar with "mount --bind" - but what does it add to > the equation? Sorry if I'm being dense. :) The FAQ tells what what the difference is. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Wed Apr 18 21:15:27 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3J4FRe06727 for devfs-outgoing; Wed, 18 Apr 2001 21:15:27 -0700 Received: from mail15.jump.net (mail15.jump.net [206.196.91.15]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3J4FQM06724 for ; Wed, 18 Apr 2001 21:15:27 -0700 Received: from sgi.com (aus-dsl-dhcp1-26.customer.jump.net [63.163.168.26]) by mail15.jump.net (8.10.2/) with ESMTP id f3J4FRA10073 for ; Wed, 18 Apr 2001 23:15:27 -0500 (CDT) Message-ID: <3ADE66B2.9181DC8D@sgi.com> Date: Wed, 18 Apr 2001 23:16:50 -0500 From: Eric Sandeen X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.4.2-XFS i686) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: Re: why mount --bind for permissions? References: <3ADE5BBF.CEE82281@sgi.com> <200104190344.f3J3i3G00877@mobilix.atnf.CSIRO.AU> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Richard Gooch wrote: > The FAQ tells what what the difference is. I'm sorry. Next time I'll read the faq more carefully, at least twice, before I ask a question of the list. :) Incidentally, there seems to be a typo in the faq, under the "add the following lines near the bvery beginning..." in the permissions section: mount --bind /dev /dev-state mount -t devfs none /dev devfsd /dev /dev. <--- what's this? Thanks, -Eric -- Eric Sandeen XFS for Linux http://oss.sgi.com/projects/xfs sandeen@sgi.com SGI, Inc. From owner-devfs@oss.sgi.com Wed Apr 18 22:16:43 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3J5GhG09145 for devfs-outgoing; Wed, 18 Apr 2001 22:16:43 -0700 Received: from mobilix.atnf.CSIRO.AU (tinylinux.tip.CSIRO.AU [130.155.192.102]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3J5GfM09138 for ; Wed, 18 Apr 2001 22:16:41 -0700 Received: (from rgooch@localhost) by mobilix.atnf.CSIRO.AU (8.10.0/8.10.0) id f3J5GYg01359; Thu, 19 Apr 2001 15:16:34 +1000 Date: Thu, 19 Apr 2001 15:16:34 +1000 Message-Id: <200104190516.f3J5GYg01359@mobilix.atnf.CSIRO.AU> From: Richard Gooch To: Eric Sandeen Cc: devfs@oss.sgi.com Subject: Re: why mount --bind for permissions? In-Reply-To: <3ADE66B2.9181DC8D@sgi.com> References: <3ADE5BBF.CEE82281@sgi.com> <200104190344.f3J3i3G00877@mobilix.atnf.CSIRO.AU> <3ADE66B2.9181DC8D@sgi.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Eric Sandeen writes: > Incidentally, there seems to be a typo in the faq, under the "add the > following lines near the bvery beginning..." in the permissions section: > > mount --bind /dev /dev-state > mount -t devfs none /dev > devfsd /dev > > /dev. <--- what's this? Erk! Cut-and-paste disease. Fixed. Thanks. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Mon Apr 23 10:50:13 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3NHoDP03732 for devfs-outgoing; Mon, 23 Apr 2001 10:50:13 -0700 Received: from mg03.austin.ibm.com (mg03.austin.ibm.com [192.35.232.20]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3NHoCM03729 for ; Mon, 23 Apr 2001 10:50:12 -0700 Received: from austin.ibm.com (netmail1.austin.ibm.com [9.53.250.96]) by mg03.austin.ibm.com (AIX4.3/8.9.3/8.9.3) with ESMTP id MAA21516; Mon, 23 Apr 2001 12:52:01 -0500 Received: from sully3.austin.ibm.com (sully3.austin.ibm.com [9.53.126.17]) by austin.ibm.com (AIX4.3/8.9.3/8.9.3) with ESMTP id MAA32446; Mon, 23 Apr 2001 12:50:09 -0500 Received: by sully3.austin.ibm.com (Postfix, from userid 500) id 9299E24640; Mon, 23 Apr 2001 23:49:01 -0500 (CDT) Date: Mon, 23 Apr 2001 23:49:01 -0500 From: sullivan To: devfs@oss.sgi.com Subject: devfs superblock error Message-ID: <20010423234901.E1209@sully3.austin.ibm.com> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5i In-Reply-To: ; from sullivam@us.ibm.com on Mon, Apr 23, 2001 at 12:47:50PM -0500 Sender: owner-devfs@oss.sgi.com Precedence: bulk On a 2.4.3 kernel I built with devfs configured but not auto mounted at boot time. I intended to manually mount it onto /devfs to play around with it. On reboot I received a message saying that my ext2 superblock was corrupted. After looking closer I noticed that my /etc/fstab was configured with the fs_spec field set to LABEL=/ Unfortunately when the fsck utility was run it used the entries provided in /proc/partitions to open and read partition label information to find a match. Since devfs was compiled in, the /proc/partitions entries contain the devfs names (ie. scsi/host0/bus0/target0/lun0/part1). But since devfs wasn't mounted at boot, these entries didn't exist (only the old /dev nodes are present), causing the error to be reported :( In /etc/fstab, replacing LABEL=/ with /dev/sda1 allowed fsck to skip the label matching code and run successfully :) Richard, could you add a note to your doc (maybe by the devfs=nomount details) to warn users to avoid using the LABEL directive in their fstab? From owner-devfs@oss.sgi.com Wed Apr 25 01:20:34 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3P8KYr23623 for devfs-outgoing; Wed, 25 Apr 2001 01:20:34 -0700 Received: from home-330.kienet (klndi4-145-253-152-063.arcor-ip.net [145.253.152.63]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3P8KVM23620 for ; Wed, 25 Apr 2001 01:20:32 -0700 Received: from kienemunet.de (localhost [127.0.0.1]) by home-330.kienet (8.11.2/8.11.2/SuSE Linux 8.11.1-0.5) with ESMTP id f3P8If817930; Wed, 25 Apr 2001 10:18:51 +0200 Message-ID: <3AE68861.B1CE5272@kienemunet.de> Date: Wed, 25 Apr 2001 10:18:41 +0200 From: Markus Kienemund X-Mailer: Mozilla 4.76 [en] (X11; U; Linux 2.2.18-SMP i686) X-Accept-Language: de-DE, en MIME-Version: 1.0 To: mclinden@informed.net CC: devfs@oss.sgi.com Subject: Re: FYI: Oracle Install References: Content-Type: multipart/mixed; boundary="------------B08F2C3F45F317928BBB553E" Sender: owner-devfs@oss.sgi.com Precedence: bulk This is a multi-part message in MIME format. --------------B08F2C3F45F317928BBB553E Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sean, I learned the same story from other sources. Its related to dbassist before creating a default database, it checks the available disk-space. Because pathnames /dev/scsci/lun.... using devfs are much longer than before, df uses 2 lines to display the size of a filesystem, and dbassist cannot cope with this. I personally think this is a problem that goes to Oracle, not devfs. I put a workaround into the SGI knowledgebase (search for "Oracle 8.1.6 installation hangs"), which is a "mv" of the long device path to a shorter one. Then dbassists survives and creates the database. Regards, Markus Kienemund SGI Germany mclinden@informed.net wrote: > > This is slightly off-topic but may be important to some users. > > I have confirmed that devfs enabled Linux systems may exhibit problems > related to the installation of Oracle 8i for Linux. Apparently the problem > is related to the Oracle Installer's inability to properly parse the output > of the df(1) command in order to calculate available disk space (I don't > have the source code for the Installer). The identical install with devfs > disabled has no problems. > > There does NOT appear to be a problem enabling devfs after installing > Oracle, however. > > Sean McLinden > Outcome Technology Associates, Inc. --------------B08F2C3F45F317928BBB553E Content-Type: text/x-vcard; charset=us-ascii; name="markus.vcf" Content-Transfer-Encoding: 7bit Content-Description: Card for Markus Kienemund Content-Disposition: attachment; filename="markus.vcf" begin:vcard n:Kienemund;Markus tel;pager:markusk@t-d1.sms.de tel;cell:0171-3040142 tel;fax:02225-948586 (please call to prepare) tel;work:Home Office: 02225-948585 x-mozilla-html:FALSE url:www.kienemunet.de org:SGI GmbH;Customer Service adr:;;Am Beckmannplatz 5;53340 Meckenheim;NRW;Germany; version:2.1 email;internet:mki@sgi.com title:System Engineer x-mozilla-cpt:;-544 fn:Markus Kienemund end:vcard --------------B08F2C3F45F317928BBB553E-- From owner-devfs@oss.sgi.com Wed Apr 25 06:45:16 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3PDjG401416 for devfs-outgoing; Wed, 25 Apr 2001 06:45:16 -0700 Received: from notes.informed.net (charcot.informed.net [205.167.80.72]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3PDjEM01408 for ; Wed, 25 Apr 2001 06:45:14 -0700 Subject: Re: FYI: Oracle Install To: Markus Kienemund Cc: devfs@oss.sgi.com X-Mailer: Lotus Notes Release 5.0.4a July 24, 2000 Message-ID: From: mclinden@informed.net Date: Wed, 25 Apr 2001 08:52:00 -0400 X-MIMETrack: Serialize by Router on notes.informed.net/informed dot net(Release 5.0.5 |September 22, 2000) at 04/25/2001 08:52:02 AM MIME-Version: 1.0 Content-type: multipart/mixed; Boundary="0__=85256A3900461B418f9e8a93df938690918c85256A3900461B41" Content-Disposition: inline Sender: owner-devfs@oss.sgi.com Precedence: bulk --0__=85256A3900461B418f9e8a93df938690918c85256A3900461B41 Content-type: text/plain; charset=us-ascii Markus: I agree that the problem is the way that Oracle reads the output from the "df" command and not devfs, per se. In fact, the Java-based Oracle Installer is one of the worst kludges that I have seen in a long time. Thanks, Sean Markus Kienemund To: mclinden@informed.net Subject: Re: FYI: Oracle Install Sent by: mki@home-330. kienet 04/25/2001 04:18 AM Sean, I learned the same story from other sources. Its related to dbassist before creating a default database, it checks the available disk-space. Because pathnames /dev/scsci/lun.... using devfs are much longer than before, df uses 2 lines to display the size of a filesystem, and dbassist cannot cope with this. I personally think this is a problem that goes to Oracle, not devfs. I put a workaround into the SGI knowledgebase (search for "Oracle 8.1.6 installation hangs"), which is a "mv" of the long device path to a shorter one. Then dbassists survives and creates the database. Regards, Markus Kienemund SGI Germany mclinden@informed.net wrote: > > This is slightly off-topic but may be important to some users. > > I have confirmed that devfs enabled Linux systems may exhibit problems > related to the installation of Oracle 8i for Linux. Apparently the problem > is related to the Oracle Installer's inability to properly parse the output > of the df(1) command in order to calculate available disk space (I don't > have the source code for the Installer). The identical install with devfs > disabled has no problems. > > There does NOT appear to be a problem enabling devfs after installing > Oracle, however. > > Sean McLinden > Outcome Technology Associates, Inc.(See attached file: markus.vcf) --0__=85256A3900461B418f9e8a93df938690918c85256A3900461B41 Content-type: application/octet-stream; name="markus.vcf" Content-Disposition: attachment; filename="markus.vcf" Content-transfer-encoding: base64 YmVnaW46dmNhcmQgDQpuOktpZW5lbXVuZDtNYXJrdXMNCnRlbDtwYWdlcjptYXJrdXNrQHQtZDEu c21zLmRlDQp0ZWw7Y2VsbDowMTcxLTMwNDAxNDINCnRlbDtmYXg6MDIyMjUtOTQ4NTg2IChwbGVh c2UgY2FsbCB0byBwcmVwYXJlKQ0KdGVsO3dvcms6SG9tZSBPZmZpY2U6IDAyMjI1LTk0ODU4NQ0K eC1tb3ppbGxhLWh0bWw6RkFMU0UNCnVybDp3d3cua2llbmVtdW5ldC5kZQ0Kb3JnOlNHSSBHbWJI O0N1c3RvbWVyIFNlcnZpY2UNCmFkcjo7O0FtIEJlY2ttYW5ucGxhdHogNTs1MzM0MCBNZWNrZW5o ZWltO05SVztHZXJtYW55Ow0KdmVyc2lvbjoyLjENCmVtYWlsO2ludGVybmV0Om1raUBzZ2kuY29t DQp0aXRsZTpTeXN0ZW0gRW5naW5lZXINCngtbW96aWxsYS1jcHQ6Oy01NDQNCmZuOk1hcmt1cyBL aWVuZW11bmQNCmVuZDp2Y2FyZA0K --0__=85256A3900461B418f9e8a93df938690918c85256A3900461B41-- From owner-devfs@oss.sgi.com Thu Apr 26 21:05:26 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3R45Qj04000 for devfs-outgoing; Thu, 26 Apr 2001 21:05:26 -0700 Received: from mail.labsysgrp.com (phnx1-blk2-hfc-0251-d1db10f1.rdc1.az.coxatwork.com [209.219.16.241]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3R45MM03997 for ; Thu, 26 Apr 2001 21:05:22 -0700 Received: from jeeves.kpf.internal ([192.168.170.1]) by mail.labsysgrp.com with esmtp (Exim 3.22 #2) id 14szVK-0001Ta-00 for devfs@oss.sgi.com; Thu, 26 Apr 2001 21:05:15 -0700 Received: from [192.168.170.101] (helo=Kevin) by jeeves.kpf.internal with smtp (Exim 3.22 #3) id 14sroH-0004Ut-00 for devfs@oss.sgi.com; Thu, 26 Apr 2001 12:52:17 -0700 Message-ID: <001101c0ce8a$d79ad0b0$65aaa8c0@Kevin> From: "Kevin P. Fleming" To: Subject: devfsd link problem and module loading questions Date: Thu, 26 Apr 2001 12:55:28 -0700 Organization: LSG, Inc. MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.50.4133.2400 X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400 Sender: owner-devfs@oss.sgi.com Precedence: bulk I'm using RH7.1, with kernel 2.4.3 and devfs mounting at boot time, plus devfsd (the one distributed with RH7.1). I've converted pretty much everything on my system to use the new names (and will be turning off old compatibility names soon), and it's all working quite well. There are two issues I've identified, though: - I have the Universal TUN/TAP driver compiled as a module for my kernel for some IP tunneling. This module, in a non-devfs environment, needs a device inode at /dev/net/tun with major 10/minor 200. With devfs loaded, when the module loads, a device inode is created at /dev/misc/net/tun with major 10/minor 200, and a link is created at /dev/net/tun. Unfortunately that link is pointed to "misc/net/tun", not "../misc/net/tun", so it's broken. The TUN/TAP driver only seems to be asking for the misc/net/tun inode to be created, so I can't find the code that is creating the link to fix it... anyone have a clue where the link is coming from? - I have an IDE (ATAPI) ZIP 250 drive that works just fine with the ide-floppy driver. When I have this driver compiled as a module, there is no /dev entry for this drive in /dev/discs if the module has not been loaded. There is an entry at /dev/ide/host0/bus0/target1/lun0, but it has no entries below since again the module has not been loaded. Obviously, I can create an entry in modules.conf to alias "/dev/discs/disc1" to ide-floppy, or even alias "/dev/ide/host0/bus0/target1/lun0/*" to ide-floppy, but I'd rather not have to tie this drive down to being "disc1" or at that specific IDE bus location. Other than forcibly loading/unloading the ide-floppy module during the boot process, is there any better solution for this problem? Ideally, it'd be very cool if the modules that are compiled for the running kernel were all given the chance to identify their associated devices and register their entries with devfs, then unload for later (possible) usage. This would take care of the devices that are internal to the machine, getting all the /dev entries created. For external devices that may not be attached this would be no help... but still. Even for things like the TUN/TAP driver, it seems to make sense to me to not make the end-user have to put an alias in modules.conf for the driver, when it could easily do a devfs_register/devfs_unregister to get the proper /dev entries created. From owner-devfs@oss.sgi.com Thu Apr 26 22:06:56 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3R56uA05444 for devfs-outgoing; Thu, 26 Apr 2001 22:06:56 -0700 Received: from vindaloo.ras.ucalgary.ca (vindaloo.ras.ucalgary.ca [136.159.55.21]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3R56tM05441 for ; Thu, 26 Apr 2001 22:06:55 -0700 Received: (from rgooch@localhost) by vindaloo.ras.ucalgary.ca (8.10.0/8.10.0) id f3R56ZM30997; Thu, 26 Apr 2001 23:06:35 -0600 Date: Thu, 26 Apr 2001 23:06:35 -0600 Message-Id: <200104270506.f3R56ZM30997@vindaloo.ras.ucalgary.ca> From: Richard Gooch To: sullivan Cc: devfs@oss.sgi.com Subject: Re: devfs superblock error In-Reply-To: <20010423234901.E1209@sully3.austin.ibm.com> References: <20010423234901.E1209@sully3.austin.ibm.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk sullivan@austin.ibm.com writes: > On a 2.4.3 kernel I built with devfs configured but not > auto mounted at boot time. I intended to manually mount > it onto /devfs to play around with it. On reboot I received a > message saying that my ext2 superblock was corrupted. > > After looking closer I noticed that my /etc/fstab was > configured with the fs_spec field set to LABEL=/ > > Unfortunately when the fsck utility was run it > used the entries provided in /proc/partitions to open > and read partition label information to find a match. > Since devfs was compiled in, the /proc/partitions entries > contain the devfs names (ie. scsi/host0/bus0/target0/lun0/part1). > But since devfs wasn't mounted at boot, these entries didn't exist > (only the old /dev nodes are present), causing the error to > be reported :( > > In /etc/fstab, replacing LABEL=/ with /dev/sda1 allowed > fsck to skip the label matching code and run successfully :) > > Richard, could you add a note to your doc (maybe by the > devfs=nomount details) to warn users to avoid using the > LABEL directive in their fstab? Done. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Mon Apr 30 02:43:30 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3U9hUq03181 for devfs-outgoing; Mon, 30 Apr 2001 02:43:30 -0700 Received: from gec.gecpalau.com (gec.gecpalau.com [206.49.60.67]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3U9hMM03178 for ; Mon, 30 Apr 2001 02:43:25 -0700 Received: from gecpalau.com (rieacs.gecpalau.com [206.49.60.69]) by gec.gecpalau.com (8.11.2/8.11.2) with ESMTP id f3U9hB110272 for ; Mon, 30 Apr 2001 18:43:11 +0900 Message-ID: <3AED33A3.27E2F77D@gecpalau.com> Date: Mon, 30 Apr 2001 18:42:59 +0900 From: Glenn Shannon X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.4.3-GECKERN i686) X-Accept-Language: en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: DevFS and 2.4.4 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Hi all! Sorry, I just grepped the archive and couldn't find the last few days (post-2.4.4 release) so I will ask now... Is there some sort of problem with 2.4.4 and devfs or did I miss a beat? I have no ptys now, and my lvm is sorta broken (I need to symlink all the drives/partitions on boot) etc. etc. Any ideas? Thanks! Glenn From owner-devfs@oss.sgi.com Mon Apr 30 16:34:19 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3UNYJg31464 for devfs-outgoing; Mon, 30 Apr 2001 16:34:19 -0700 Received: from vindaloo.ras.ucalgary.ca (vindaloo.ras.ucalgary.ca [136.159.55.21]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3UNYIM31459 for ; Mon, 30 Apr 2001 16:34:18 -0700 Received: (from rgooch@localhost) by vindaloo.ras.ucalgary.ca (8.10.0/8.10.0) id f3UNXwZ01539; Mon, 30 Apr 2001 17:33:58 -0600 Date: Mon, 30 Apr 2001 17:33:58 -0600 Message-Id: <200104302333.f3UNXwZ01539@vindaloo.ras.ucalgary.ca> From: Richard Gooch To: Glenn Shannon Cc: devfs@oss.sgi.com Subject: Re: DevFS and 2.4.4 In-Reply-To: <3AED33A3.27E2F77D@gecpalau.com> References: <3AED33A3.27E2F77D@gecpalau.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Glenn Shannon writes: > Hi all! > > Sorry, I just grepped the archive and couldn't find the last few days > (post-2.4.4 release) so I will ask now... > > Is there some sort of problem with 2.4.4 and devfs or did I miss a beat? > > I have no ptys now, and my lvm is sorta broken (I need to symlink all > the drives/partitions on boot) etc. etc. I haven't had any problems. I don't use Unix98 ptys nor do I use LVM, though. And I don't see any changes to the pty or LVM code in patch-2.4.4. Please investigate further and let us know if you get more information. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Mon Apr 30 16:38:48 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3UNcmK31621 for devfs-outgoing; Mon, 30 Apr 2001 16:38:48 -0700 Received: from mailout01.sul.t-online.com (mailout01.sul.t-online.com [194.25.134.80]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3UNclM31618 for ; Mon, 30 Apr 2001 16:38:47 -0700 Received: from fwd00.sul.t-online.com by mailout01.sul.t-online.com with smtp id 14uNFd-00062q-01; Tue, 01 May 2001 01:38:45 +0200 Received: from [192.168.1.1] (520025986886-0001@[217.2.201.35]) by fmrl00.sul.t-online.com with esmtp id 14uNFi-0PBckCC; Tue, 1 May 2001 01:38:50 +0200 Date: Tue, 1 May 2001 01:38:19 +0200 (MEST) From: Marcel Beltz To: devfs@oss.sgi.com Subject: webcam Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Sender: 520025986886-0001@t-dialin.net Sender: owner-devfs@oss.sgi.com Precedence: bulk Hello i am new on this list and have a question: i want to use a webcam, and i use linux 2.4.1 with devfs. it works good, but the client for the webcam can't connect to it, because it expect major and minor numbers. Howto config devfs that it provied a /dev/video0 and /dev/video1 with major number 81. The rest of the /dev-filesystem should be untouched. Please help mfg Marcel Beltz From owner-devfs@oss.sgi.com Mon Apr 30 16:59:00 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f3UNx0t32155 for devfs-outgoing; Mon, 30 Apr 2001 16:59:00 -0700 Received: from gec.gecpalau.com (gec.gecpalau.com [206.49.60.67]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f3UNwmM32136 for ; Mon, 30 Apr 2001 16:58:51 -0700 Received: from gecpalau.com (rieacs.gecpalau.com [206.49.60.69]) by gec.gecpalau.com (8.11.2/8.11.2) with ESMTP id f3UNun113434; Tue, 1 May 2001 08:56:49 +0900 Message-ID: <3AEDFBB9.FB230204@gecpalau.com> Date: Tue, 01 May 2001 08:56:41 +0900 From: Glenn Shannon X-Mailer: Mozilla 4.72 [en] (X11; U; Linux 2.4.3-GECKERN i686) X-Accept-Language: en MIME-Version: 1.0 To: Richard Gooch CC: devfs@oss.sgi.com Subject: Re: DevFS and 2.4.4 References: <3AED33A3.27E2F77D@gecpalau.com> <200104302333.f3UNXwZ01539@vindaloo.ras.ucalgary.ca> Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Sender: owner-devfs@oss.sgi.com Precedence: bulk Richard Gooch wrote: > Glenn Shannon writes: > > Hi all! > > > > Sorry, I just grepped the archive and couldn't find the last few days > > (post-2.4.4 release) so I will ask now... > > > > Is there some sort of problem with 2.4.4 and devfs or did I miss a beat? > > > > I have no ptys now, and my lvm is sorta broken (I need to symlink all > > the drives/partitions on boot) etc. etc. > > I haven't had any problems. I don't use Unix98 ptys nor do I use LVM, > though. And I don't see any changes to the pty or LVM code in > patch-2.4.4. > > Please investigate further and let us know if you get more > information. > > Regards, > > Richard.... > Permanent: rgooch@atnf.csiro.au > Current: rgooch@ras.ucalgary.ca Well, I got DevFS and LVM to play nice by having a bunch of symlinks created at boot...but ptys still aren't working....I have no ptys available and so telnetd httpd etc. don't work Is there some way I can produce some debugging information for you? Glenn From owner-devfs@oss.sgi.com Mon Apr 30 17:23:18 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f410NIp00628 for devfs-outgoing; Mon, 30 Apr 2001 17:23:18 -0700 Received: from vindaloo.ras.ucalgary.ca (vindaloo.ras.ucalgary.ca [136.159.55.21]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f410NGM00625 for ; Mon, 30 Apr 2001 17:23:16 -0700 Received: (from rgooch@localhost) by vindaloo.ras.ucalgary.ca (8.10.0/8.10.0) id f410Mdj02176; Mon, 30 Apr 2001 18:22:39 -0600 Date: Mon, 30 Apr 2001 18:22:39 -0600 Message-Id: <200105010022.f410Mdj02176@vindaloo.ras.ucalgary.ca> From: Richard Gooch To: Glenn Shannon Cc: devfs@oss.sgi.com Subject: Re: DevFS and 2.4.4 In-Reply-To: <3AEDFBB9.FB230204@gecpalau.com> References: <3AED33A3.27E2F77D@gecpalau.com> <200104302333.f3UNXwZ01539@vindaloo.ras.ucalgary.ca> <3AEDFBB9.FB230204@gecpalau.com> Sender: owner-devfs@oss.sgi.com Precedence: bulk Glenn Shannon writes: > Richard Gooch wrote: > > > Glenn Shannon writes: > > > Hi all! > > > > > > Sorry, I just grepped the archive and couldn't find the last few days > > > (post-2.4.4 release) so I will ask now... > > > > > > Is there some sort of problem with 2.4.4 and devfs or did I miss a beat? > > > > > > I have no ptys now, and my lvm is sorta broken (I need to symlink all > > > the drives/partitions on boot) etc. etc. > > > > I haven't had any problems. I don't use Unix98 ptys nor do I use LVM, > > though. And I don't see any changes to the pty or LVM code in > > patch-2.4.4. > > > > Please investigate further and let us know if you get more > > information. > > > > Regards, > > > > Richard.... > > Permanent: rgooch@atnf.csiro.au > > Current: rgooch@ras.ucalgary.ca > > Well, I got DevFS and LVM to play nice by having a bunch of symlinks created > at boot...but ptys still aren't working....I have no ptys available and so > telnetd httpd etc. don't work > > Is there some way I can produce some debugging information for you? Well, send the output of "ls -lFR /dev" after the LVM module is loaded, but *before* you manually created those links. Then send a list of the links that you had to create. Specify whether you're using Unix98 ptys or BSD ptys. And whether devpts is compiled or not, and whether it's mounted or not. Also state which kernel broke. If the last kernel you tried wasn't 2.4.3, then go back in time compiling old kernels and find out which one broke. Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From owner-devfs@oss.sgi.com Mon Apr 30 17:47:08 2001 Received: (from majordomo@localhost) by oss.sgi.com (8.11.3/8.11.3) id f410l8n01376 for devfs-outgoing; Mon, 30 Apr 2001 17:47:08 -0700 Received: from mailgate.rz.uni-karlsruhe.de (mailgate.rz.uni-karlsruhe.de [129.13.64.97]) by oss.sgi.com (8.11.3/8.11.3) with ESMTP id f410l7M01373 for ; Mon, 30 Apr 2001 17:47:07 -0700 Received: from nce2.hadiko.de (hadince2.hadiko.uni-karlsruhe.de [172.20.32.2]) by mailgate.rz.uni-karlsruhe.de with esmtp (Exim 3.16 #1) id 14uOJd-0004jI-00; Tue, 01 May 2001 02:46:57 +0200 Received: from panorama.hadiko.de (hadii309.hadiko.uni-karlsruhe.de [172.20.44.69]) by nce2.hadiko.de (8.9.3/8.9.3) with SMTP id CAA00581 for ; Tue, 1 May 2001 02:46:56 +0200 (MET DST) Received: (qmail 4027 invoked from network); 1 May 2001 00:46:53 -0000 Received: from localhost (127.0.0.1) by localhost with SMTP; 1 May 2001 00:46:53 -0000 To: glenn@gecpalau.com, devfs@oss.sgi.com Subject: Re: DevFS and 2.4.4 From: Robert Siemer In-Reply-To: <3AEDFBB9.FB230204@gecpalau.com> References: <3AED33A3.27E2F77D@gecpalau.com> <200104302333.f3UNXwZ01539@vindaloo.ras.ucalgary.ca> <3AEDFBB9.FB230204@gecpalau.com> X-Mailer: Mew version 1.94b25 on Emacs 20.5 / Mule 4.0 (HANANOEN) Reply-To: Robert Siemer Mime-Version: 1.0 Content-Type: Text/Plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-Id: <20010501024650B.siemer@panorama.hadiko.de> Date: Tue, 01 May 2001 02:46:50 +0200 X-Dispatcher: imput version 990425(IM115) Lines: 15 Sender: owner-devfs@oss.sgi.com Precedence: bulk From: Glenn Shannon > Well, I got DevFS and LVM to play nice by having a bunch of symlinks > created at boot...but ptys still aren't working....I have no ptys > available and so telnetd httpd etc. don't work httpd uses ptys?? I've problems to recognize what special devices a httpd uses at all... One think I can suggest: compile unix-domain-sockets (network options) into the kernel. Don't use modules here... (But this is completely unrelated to devfs.) Bye, Robert