From tvignaud@mandrakesoft.com Fri Oct 11 01:54:24 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 11 Oct 2002 01:54:26 -0700 (PDT) Received: from vador.mandrakesoft.com (office.mandrakesoft.com [195.68.114.34]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9B8sNtG005841 for ; Fri, 11 Oct 2002 01:54:23 -0700 Received: from vador.mandrakesoft.com (localhost.localdomain [127.0.0.1]) by vador.mandrakesoft.com (Postfix) with ESMTP id 4492E1BD4C for ; Fri, 11 Oct 2002 10:55:53 +0200 (CEST) To: devfs@oss.sgi.com Subject: [BUGFIX] Re: [PATCHES] patches currently used in mandrake-linux References: X-URL: Date: Fri, 11 Oct 2002 10:55:53 +0200 In-Reply-To: (Thierry Vignaud's message of "Wed, 21 Aug 2002 17:40:18 +0200") Message-ID: User-Agent: Gnus/5.090007 (Oort Gnus v0.07) Emacs/21.2.91 (i386-mandrake-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" X-archive-position: 6 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: tvignaud@mandrakesoft.com Precedence: bulk X-list: devfs Content-Length: 1693 Lines: 14 --=-=-= Thierry Vignaud writes: > o Compacq smart array support[1] humm, some time elapsed since the last time i had access to such pets, and the patch get broken in the while. i recently had access to one of this (well, just this morning), and here come the fixed patch. --=-=-= Content-Type: text/x-patch Content-Disposition: attachment; filename=devfsd-1.3.25-ida.patch --- ./compat_name.c.ida 2002-08-17 19:12:29.000000000 +0200 +++ ./compat_name.c 2002-08-17 19:14:56.000000000 +0200 @@ -220,6 +220,13 @@ compat_name = buffer; } return (compat_name); + } + else if (strncmp (devname, "ida/", 4) == 0) { + if (strcmp (devname + namelen - 4, "disc") == 0) + sprintf ( buffer, "ida/c%dd%d", minor / 16 / 16, minor / 16 % 16); + else if (strncmp (ptr = (strrchr (devname, '/') +1), "part", 4) == 0) + sprintf (buffer, "ida/c%dd%dp%s", minor / 16 / 16, minor / 16 % 16, ptr + 4); + compat_name = buffer; } else if (strncmp (devname, "vcc/", 4) == 0) { --- ./devfsd.c.ida 2002-03-25 07:59:40.000000000 +0100 +++ ./devfsd.c 2002-08-17 19:12:29.000000000 +0200 @@ -1462,6 +1462,11 @@ case AC_RMOLDCOMPAT: compat_name = get_old_name (info->devname, info->namelen, compat_buf, info->major, info->minor); + if ( strncmp (info->devname, "ida/", 4) == 0) { + ptr = (strrchr (info->devname, '/') +1); + if ( (strncmp (ptr, "part", 4) == 0) || (strncmp (ptr, "disc", 5) == 0) ) + dest_name = info->devname+4; + } break; case AC_MKNEWCOMPAT: case AC_RMNEWCOMPAT: --=-=-= i may get access to a cciss box too; if yes, i'll add support to this specy too and 'll post the patch here. --=-=-=-- From tvignaud@mandrakesoft.com Fri Oct 11 01:56:08 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 11 Oct 2002 01:56:10 -0700 (PDT) Received: from vador.mandrakesoft.com (office.mandrakesoft.com [195.68.114.34]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9B8u7tG005852 for ; Fri, 11 Oct 2002 01:56:08 -0700 Received: from vador.mandrakesoft.com (localhost.localdomain [127.0.0.1]) by vador.mandrakesoft.com (Postfix) with ESMTP id 88F5C1BD4C for ; Fri, 11 Oct 2002 10:57:38 +0200 (CEST) To: devfs@oss.sgi.com Subject: Re: [BUGFIX] Re: [PATCHES] patches currently used in mandrake-linux References: X-URL: Date: Fri, 11 Oct 2002 10:57:38 +0200 In-Reply-To: (Thierry Vignaud's message of "Fri, 11 Oct 2002 10:55:53 +0200") Message-ID: User-Agent: Gnus/5.090007 (Oort Gnus v0.07) Emacs/21.2.91 (i386-mandrake-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" X-archive-position: 7 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: tvignaud@mandrakesoft.com Precedence: bulk X-list: devfs Content-Length: 2006 Lines: 20 --=-=-= Thierry Vignaud writes: > > o Compacq smart array support[1] > > humm, some time elapsed since the last time i had access to such pets, > and the patch get broken in the while. > i recently had access to one of this (well, just this morning), and > here come the fixed patch. > > > > i may get access to a cciss box too; if yes, i'll add support to this > specy too and 'll post the patch here. > > o Mylex support[1] the above ida patch prevent previous mylex patch to apply. here's the new one: --=-=-= Content-Type: text/x-patch Content-Disposition: attachment; filename=devfsd-1.3.25-rd.patch --- ./compat_name.c.cdrom 2002-01-25 08:06:35.000000000 +0100 +++ ./compat_name.c 2002-08-17 18:12:54.000000000 +0200 @@ -144,6 +152,15 @@ char *ptr; struct translate_struct *trans; + /* rd/disc needs to be searched before translate_table (rd/0 -> ram0) */ + if (strncmp (devname, "rd/disc", sizeof("rd/disc") - 1) == 0) { + if(strcmp (devname + namelen - 4, "disc") == 0) + sprintf ( buffer, "rd/c%dd%d", minor / 8 / 32, minor / 8 % 32); + else if (strncmp (ptr = (strrchr (devname, '/') +1), "part", 4) == 0) + sprintf (buffer, "rd/c%dd%dp%s", minor / 8 / 32, minor / 8 % 32, ptr + 4); + return buffer; + } + for (trans = translate_table; trans->match != NULL; ++trans) { size_t len = strlen (trans->match); --- ./devfsd.c.cdrom 2002-03-25 07:59:40.000000000 +0100 +++ ./devfsd.c 2002-08-17 18:06:17.000000000 +0200 @@ -1462,5 +1462,8 @@ case AC_RMOLDCOMPAT: compat_name = get_old_name (info->devname, info->namelen, compat_buf, info->major, info->minor); + if ( (strncmp (info->devname, "rd/", 3) == 0) && + (strncmp (ptr = (strrchr (info->devname, '/') +1), "part", 4) == 0) ) + dest_name = info->devname+3; if ( strncmp (info->devname, "ida/", 4) == 0) { ptr = (strrchr (info->devname, '/') +1); if ( (strncmp (ptr, "part", 4) == 0) || (strncmp (ptr, "disc", 4) == 0) ) { --=-=-=-- From rgooch@vindaloo.ras.ucalgary.ca Tue Oct 15 22:13:26 2002 Received: with ECARTIS (v1.0.0; list devfs); Tue, 15 Oct 2002 22:13:28 -0700 (PDT) Received: from vindaloo.ras.ucalgary.ca (vindaloo.ras.ucalgary.ca [136.159.55.21]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9G5DPtG027118 for ; Tue, 15 Oct 2002 22:13:26 -0700 Received: (from rgooch@localhost) by vindaloo.ras.ucalgary.ca (8.10.0/8.10.0) id g9G5D0M15828; Tue, 15 Oct 2002 23:13:00 -0600 Date: Tue, 15 Oct 2002 23:13:00 -0600 Message-Id: <200210160513.g9G5D0M15828@vindaloo.ras.ucalgary.ca> From: Richard Gooch To: linux-kernel@vger.kernel.org, devfs-announce-list@vindaloo.ras.ucalgary.ca Subject: [PATCH] devfs v199.17 available X-archive-position: 8 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: rgooch@ras.ucalgary.ca Precedence: bulk X-list: devfs Content-Length: 756 Lines: 27 Hi, all. Version 199.17 of my devfs patch is now available from: http://www.atnf.csiro.au/~rgooch/linux/kernel-patches.html The devfs FAQ is also available here. Patch directly available from: ftp://ftp.??.kernel.org/pub/linux/kernel/people/rgooch/v2.4/devfs-patch-current.gz AND: ftp://ftp.atnf.csiro.au/pub/people/rgooch/linux/kernel-patches/v2.4/devfs-patch-current.gz This is against 2.4.20-pre11. Highlights of this release: - Updated README from master HTML file - Switched lingering structure field initialiser to ISO C - Added locking when setting/clearing flags - Documentation fix in fs/devfs/util.c - Created Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From rgooch@vindaloo.ras.ucalgary.ca Tue Oct 15 22:46:24 2002 Received: with ECARTIS (v1.0.0; list devfs); Tue, 15 Oct 2002 22:46:28 -0700 (PDT) Received: from vindaloo.ras.ucalgary.ca (vindaloo.ras.ucalgary.ca [136.159.55.21]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9G5kNtG028898 for ; Tue, 15 Oct 2002 22:46:24 -0700 Received: (from rgooch@localhost) by vindaloo.ras.ucalgary.ca (8.10.0/8.10.0) id g9G5kJm18075; Tue, 15 Oct 2002 23:46:19 -0600 Date: Tue, 15 Oct 2002 23:46:19 -0600 Message-Id: <200210160546.g9G5kJm18075@vindaloo.ras.ucalgary.ca> From: Richard Gooch To: linux-kernel@vger.kernel.org, devfs-announce-list@vindaloo.ras.ucalgary.ca Subject: [PATCH] devfs v218 available X-archive-position: 9 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: rgooch@ras.ucalgary.ca Precedence: bulk X-list: devfs Content-Length: 767 Lines: 27 Hi, all. Version 218 of my devfs patch is now available from: http://www.atnf.csiro.au/~rgooch/linux/kernel-patches.html The devfs FAQ is also available here. Patch directly available from: ftp://ftp.??.kernel.org/pub/linux/kernel/people/rgooch/v2.5/devfs-patch-current.gz AND: ftp://ftp.atnf.csiro.au/pub/people/rgooch/linux/kernel-patches/v2.5/devfs-patch-current.gz NOTE: kernel 2.5.1 and later require devfsd-v1.3.19 or later. This is against 2.5.43. Highlights of this release: - Removed DEVFS_FL_AUTO_OWNER flag - Switched lingering structure field initialiser to ISO C - Added locking when setting/clearing flags - Documentation fix in fs/devfs/util.c Regards, Richard.... Permanent: rgooch@atnf.csiro.au Current: rgooch@ras.ucalgary.ca From mn3@os.inf.tu-dresden.de Wed Oct 16 05:40:34 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 05:40:36 -0700 (PDT) Received: from os.inf.tu-dresden.de (os.inf.tu-dresden.de [141.76.48.99]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GCeXtG003145 for ; Wed, 16 Oct 2002 05:40:34 -0700 Received: from mn3 (helo=localhost) by os.inf.tu-dresden.de with local-esmtp (Exim 3.36) id 181nTU-000BlE-00 for devfs@oss.sgi.com; Wed, 16 Oct 2002 14:40:32 +0200 Date: Wed, 16 Oct 2002 14:40:32 +0200 (DFT) From: Mathias Noack To: devfs@oss.sgi.com Subject: partition table Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 10 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: mn3@os.inf.tu-dresden.de Precedence: bulk X-list: devfs Content-Length: 524 Lines: 19 Hi there, I got the following problem: I created a block special inode with (from HOWTO SSI Cluster project): mknod /dev/ide/host0/bus0/target0/lun0/part3 b 7 1 After that I loop back mount a root filesystem: losetup /dev/loop1 root_fs So know I was wondering why this block device (/dev/ide/host0/bus0/target0/lun0/part3) doesn't appear in the partition table (/proc/partition)? What I have to do? I need this because the ptool from the OpenGFS project reads the partition table to creating pools. Regards, Mathias From john.pellett@adic.com Wed Oct 16 07:02:22 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 07:02:24 -0700 (PDT) Received: from penguin.adic.com ([63.81.117.28]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GE2MtG011457 for ; Wed, 16 Oct 2002 07:02:22 -0700 Received: by penguin.adic.com with Internet Mail Service (5.5.2653.19) id <4W3N1X0A>; Wed, 16 Oct 2002 07:01:54 -0700 Message-ID: <995FF289C9D69747A09E429926445954B78899@penguin.adic.com> From: john.pellett@adic.com To: devfs@oss.sgi.com Subject: devfs and RedHat Linux 8.0 Date: Wed, 16 Oct 2002 07:01:52 -0700 MIME-Version: 1.0 X-Mailer: Internet Mail Service (5.5.2653.19) Content-Type: text/plain; charset="ISO-8859-1" X-archive-position: 11 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: john.pellett@adic.com Precedence: bulk X-list: devfs Content-Length: 171 Lines: 7 Hello, I have been unable to get devfs running on RedHat linux 8.0. Has anyone been able to get this to work? BTW...I got it to work on RedHat linux 7.3. Thanks, John From tvignaud@mandrakesoft.com Wed Oct 16 09:15:12 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 09:15:15 -0700 (PDT) Received: from vador.mandrakesoft.com (office.mandrakesoft.com [195.68.114.34]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GGFBtG026120 for ; Wed, 16 Oct 2002 09:15:11 -0700 Received: from vador.mandrakesoft.com (localhost.localdomain [127.0.0.1]) by vador.mandrakesoft.com (Postfix) with ESMTP id F2C801BD01 for ; Wed, 16 Oct 2002 18:16:23 +0200 (CEST) To: devfs@oss.sgi.com Subject: Re: [BUGFIX] Re: [PATCHES] patches currently used in mandrake-linux References: X-URL: Date: Wed, 16 Oct 2002 18:16:23 +0200 In-Reply-To: (Thierry Vignaud's message of "Fri, 11 Oct 2002 10:55:53 +0200") Message-ID: User-Agent: Gnus/5.090007 (Oort Gnus v0.07) Emacs/21.2.91 (i386-mandrake-linux-gnu) MIME-Version: 1.0 Content-Type: multipart/mixed; boundary="=-=-=" X-archive-position: 12 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: tvignaud@mandrakesoft.com Precedence: bulk X-list: devfs Content-Length: 1549 Lines: 9 --=-=-= Thierry Vignaud writes: > i may get access to a cciss box too; if yes, i'll add support to > this specy too and 'll post the patch here. here it is: --=-=-= Content-Type: text/x-patch Content-Disposition: attachment; filename=devfsd-1.3.25-cciss.patch --- ./compat_name.c.cciss 2002-10-16 17:17:16.000000000 +0200 +++ ./compat_name.c 2002-10-16 17:17:37.000000000 +0200 @@ -224,6 +224,13 @@ sprintf (buffer, "ida/c%dd%dp%s", minor / 16 / 16, minor / 16 % 16, ptr + 4); compat_name = buffer; } + else if (strncmp (devname, "cciss/", 6) == 0) { + if (strcmp (devname + namelen - 4, "disc") == 0) + sprintf ( buffer, "cciss/c%dd%d", minor / 16 / 16, minor / 16 % 16); + else if (strncmp (ptr = (strrchr (devname, '/') +1), "part", 4) == 0) + sprintf (buffer, "cciss/c%dd%dp%s", minor / 16 / 16, minor / 16 % 16, ptr + 4); + compat_name = buffer; + } else if (strncmp (devname, "vcc/", 4) == 0) { sprintf (buffer, "vcs%s", devname + 4); --- ./devfsd.c.cciss 2002-10-16 17:17:16.000000000 +0200 +++ ./devfsd.c 2002-10-16 17:18:05.000000000 +0200 @@ -1467,6 +1467,11 @@ if ( (strncmp (ptr, "part", 4) == 0) || (strncmp (ptr, "disc", 5) == 0) ) dest_name = info->devname+4; } + if ( strncmp (info->devname, "cciss/", 6) == 0) { + ptr = (strrchr (info->devname, '/') +1); + if ( (strncmp (ptr, "part", 4) == 0) || (strncmp (ptr, "disc", 5) == 0) ) + dest_name = info->devname+6; + } break; case AC_MKNEWCOMPAT: case AC_RMNEWCOMPAT: --=-=-=-- From samuel@pc-62-31-74-96-ed.blueyonder.co.uk Wed Oct 16 12:27:43 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 12:27:45 -0700 (PDT) Received: from pc-62-31-74-96-ed.blueyonder.co.uk (pc-62-31-74-96-ed.blueyonder.co.uk [62.31.74.96]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GJRgtG013341 for ; Wed, 16 Oct 2002 12:27:43 -0700 Received: from pc-62-31-74-96-ed.blueyonder.co.uk (localhost.localdomain [127.0.0.1]) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6) with ESMTP id g9GJWg9b000617 for ; Wed, 16 Oct 2002 20:32:42 +0100 Received: (from samuel@localhost) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6/Submit) id g9GJWggn000616; Wed, 16 Oct 2002 20:32:42 +0100 Date: Wed, 16 Oct 2002 20:32:42 +0100 From: Sam Halliday To: devfs@oss.sgi.com Subject: too much mounted Message-Id: <20021016203242.4017569e.samuel@ma.hw.ac.uk> X-Mailer: Sylpheed version 0.8.5claws (GTK+ 1.2.10; ) X-Operating-System: Linux from Scratch 4.0 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-archive-position: 13 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: samuel@ma.hw.ac.uk Precedence: bulk X-list: devfs Content-Length: 1451 Lines: 34 hi there, i recently rebuilt my system with devfs support... i like it a lot! good idea! but unfortunately i have one small gripe... i have my /etc/mtab as a symbolic link to /proc/mounts which has always worked nicely for me and i can therefore be sure that even on a system crash or read-only root mount, my mtab will always be correct... however, using devfs (on linux-2.4.19 and i also tested 2.4.20-pre11) i get two mounts for the rootfs... here is my /proc/mounts rootfs / rootfs rw 0 0 /dev/root / ext3 rw 0 0 none /dev devfs rw 0 0 proc /proc proc rw 0 0 /dev/hda7 /home ext3 rw 0 0 none /dev/shm tmpfs rw 0 0 automount(pid77) /mnt/.autofs autofs rw 0 0 is this a bug, or is my symlink a bad idea. this is the way Linux from Scratch reccomends doing to /etc/mtab and i agree with their arguements. has anyone else experienced this? is it known already, and if so, is there a fix in the running? (hopefully before 2.4.20!) also, just to let the list know that for my kernel setup i needed to add the module 'ide-scsi' to the line for cdroms in /etc/modules.devfs This is because i use as much modularisation as possible and devfsd kept using the normal ide cdrom drivers for my cdrom when in actual fact, i required the scsi emulation for cdburning purposes. i dont see how this could cause problems on a system which doesnt have an ide cdwriter. Sam -- The chief danger in life is that you may take too many precautions. -- Alfred Adler From per@fukt.bth.se Wed Oct 16 12:42:05 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 12:42:09 -0700 (PDT) Received: from ce00521-p19-hertas5.cenara.com (root@ce00521-p19-hertas5.cenara.com [195.38.1.140]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GJg3tG014656 for ; Wed, 16 Oct 2002 12:42:05 -0700 Received: from localhost (per@localhost [127.0.0.1]) by ce00521-p19-hertas5.cenara.com (8.12.6/8.12.6) with ESMTP id g9GJg0b8009489; Wed, 16 Oct 2002 21:42:00 +0200 Date: Wed, 16 Oct 2002 21:42:00 +0200 (CEST) From: =?iso-8859-1?Q?Per_Lid=E9n?= X-X-Sender: per@ce00521-p19-hertas5.cenara.com To: Richard Gooch cc: devfs@oss.sgi.com, Subject: option "root=" doesn't work with devfs device names Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 14 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: per@fukt.bth.se Precedence: bulk X-list: devfs Content-Length: 1412 Lines: 49 Hi, The "root=" kernel option doesn't work when used together with devfs device names, e.g. root=/dev/discs/disc0/part1. in do_mounts.c: static int __init root_dev_setup(char *line) { int i; char ch; ROOT_DEV = name_to_kdev_t(line); ^^^^^^^^^^^^^^^^^^^^ The call to name_to_kdev_t() should return 0 if the device name starts with "/dev/" but was not found in root_dev_names[] (which is the case if e.g. root=/dev/discs/disc0/part1 is used as a kernel option). But instead of returning 0, name_to_kdev_t() will treat "discs/disc0/part1" as a hex-number, which results in "discs/disc0/part1" being passed to simple_strtoul(). The returned value from name_to_kdev_t() will thus be garbage, and later calls to create_dev() will fail to create the appropriate devfs entries because ROOT_DEV is non-zero. The end result is a kernel panic "VFS: Unable to mount root fs on ...". This has been broken since 2.4.19. The patch below fixes this problem. It would be nice if this problem was solved before 2.4.20 was released. /Per --- linux-2.4.20-pre10/init/do_mounts.c.old 2002-10-09 23:04:59.000000000 +0200 +++ linux-2.4.20-pre10/init/do_mounts.c 2002-10-09 23:50:46.000000000 +0200 @@ -258,6 +258,8 @@ } dev++; } while (dev->name); + if (!(dev->name)) + return to_kdev_t(0); } return to_kdev_t(base + simple_strtoul(line,NULL,base?10:16)); } From jabiru_croc@yahoo.com Wed Oct 16 13:59:02 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 13:59:03 -0700 (PDT) Received: from web40017.mail.yahoo.com (web40017.mail.yahoo.com [66.218.78.57]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GKx2tG023477 for ; Wed, 16 Oct 2002 13:59:02 -0700 Message-ID: <20021016205856.5910.qmail@web40017.mail.yahoo.com> Received: from [24.50.169.14] by web40017.mail.yahoo.com via HTTP; Wed, 16 Oct 2002 13:58:56 PDT Date: Wed, 16 Oct 2002 13:58:56 -0700 (PDT) From: Brad Chapman Subject: Re: option "root=" doesn't work with devfs device names To: "Per_Lidén" Cc: devfs@oss.sgi.com In-Reply-To: MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii X-archive-position: 15 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: jabiru_croc@yahoo.com Precedence: bulk X-list: devfs Content-Length: 1570 Lines: 52 Sir, --- Per_Lidén wrote: > Hi, > > The "root=" kernel option doesn't work when used together with devfs > device names, e.g. root=/dev/discs/disc0/part1. > > in do_mounts.c: > > static int __init root_dev_setup(char *line) > { > int i; > char ch; > > ROOT_DEV = name_to_kdev_t(line); > ^^^^^^^^^^^^^^^^^^^^ > > The call to name_to_kdev_t() should return 0 if the device name starts > with "/dev/" but was not found in root_dev_names[] (which is the case if > e.g. root=/dev/discs/disc0/part1 is used as a kernel option). But instead > of returning 0, name_to_kdev_t() will treat "discs/disc0/part1" as a > hex-number, which results in "discs/disc0/part1" being passed to > simple_strtoul(). The returned value from name_to_kdev_t() will thus be > garbage, and later calls to create_dev() will fail to create the > appropriate devfs entries because ROOT_DEV is non-zero. Ah, so THIS is why it never seemed to work right. I always thought it was a kernel problem but couldn't figure out what to look for. > > The end result is a kernel panic "VFS: Unable to mount root fs on ...". > > This has been broken since 2.4.19. > > The patch below fixes this problem. It would be nice if this problem was > solved before 2.4.20 was released. So with this patch a root= entry of /dev/discs/disc0/part1 returns the correct device entry in devfs? Brad > > /Per > __________________________________________________ Do you Yahoo!? Faith Hill - Exclusive Performances, Videos & More http://faith.yahoo.com From karmak@karmak.org Wed Oct 16 14:22:35 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 14:22:38 -0700 (PDT) Received: from ariel.karmak.org (pcp490536pcs.nash01.tn.comcast.net [68.53.1.252]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GLMYtG026264 for ; Wed, 16 Oct 2002 14:22:35 -0700 Received: from localhost (localhost [127.0.0.1]) (uid 1) by ariel.karmak.org with local; Wed, 16 Oct 2002 21:22:34 +0000 Date: Wed, 16 Oct 2002 21:22:34 +0000 From: Michael Carmack To: Sam Halliday Cc: devfs@oss.sgi.com Subject: Re: too much mounted Message-ID: <20021016212234.GH31008@ariel.karmak.org> References: <20021016203242.4017569e.samuel@ma.hw.ac.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Content-Disposition: inline In-Reply-To: <20021016203242.4017569e.samuel@ma.hw.ac.uk> User-Agent: Mutt/1.3.24i X-archive-position: 16 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: karmak@karmak.org Precedence: bulk X-list: devfs Content-Length: 1372 Lines: 34 On Wed, Oct 16, 2002 at 08:32:42PM +0100, Sam Halliday wrote: > hi there, > > i recently rebuilt my system with devfs support... i like it a lot! good > idea! but unfortunately i have one small gripe... i have my /etc/mtab as > a symbolic link to /proc/mounts which has always worked nicely for me > and i can therefore be sure that even on a system crash or read-only > root mount, my mtab will always be correct... however, using devfs (on > linux-2.4.19 and i also tested 2.4.20-pre11) i get two mounts for the > rootfs... here is my /proc/mounts > > rootfs / rootfs rw 0 0 > /dev/root / ext3 rw 0 0 > none /dev devfs rw 0 0 > proc /proc proc rw 0 0 > /dev/hda7 /home ext3 rw 0 0 > none /dev/shm tmpfs rw 0 0 > automount(pid77) /mnt/.autofs autofs rw 0 0 > > is this a bug, or is my symlink a bad idea. this is the way Linux from > Scratch reccomends doing to /etc/mtab and i agree with their arguements. > has anyone else experienced this? is it known already, and if so, is > there a fix in the running? (hopefully before 2.4.20!) I have noticed it too, and I never saw this before 2.4.19. With 2.4.17 (or was it 2.4.9?) the root fs showed up as /dev/ide/host0/... I wasn't even aware that devfs was the culprit here. I've been running several machines like this for months and have never noticed a problem, but I'm also curious why this changed recently. m. From samuel@pc-62-31-74-96-ed.blueyonder.co.uk Wed Oct 16 15:19:46 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 15:19:47 -0700 (PDT) Received: from pc-62-31-74-96-ed.blueyonder.co.uk (pc-62-31-74-96-ed.blueyonder.co.uk [62.31.74.96]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9GMJitG005737 for ; Wed, 16 Oct 2002 15:19:45 -0700 Received: from pc-62-31-74-96-ed.blueyonder.co.uk (localhost.localdomain [127.0.0.1]) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6) with ESMTP id g9GMOjFG000484 for ; Wed, 16 Oct 2002 23:24:45 +0100 Received: (from samuel@localhost) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6/Submit) id g9GMOjqb000482; Wed, 16 Oct 2002 23:24:45 +0100 Date: Wed, 16 Oct 2002 23:24:45 +0100 From: Sam Halliday To: devfs@oss.sgi.com Subject: Re: too much mounted Message-Id: <20021016232445.4aac01e8.samuel@ma.hw.ac.uk> In-Reply-To: <20021016212234.GH31008@ariel.karmak.org> References: <20021016203242.4017569e.samuel@ma.hw.ac.uk> <20021016212234.GH31008@ariel.karmak.org> X-Mailer: Sylpheed version 0.8.5claws (GTK+ 1.2.10; ) X-Operating-System: Linux from Scratch 4.0 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-archive-position: 17 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: samuel@ma.hw.ac.uk Precedence: bulk X-list: devfs Content-Length: 2149 Lines: 40 On Wed, 16 Oct 2002 21:22:34 +0000 Michael Carmack wrote: > On Wed, Oct 16, 2002 at 08:32:42PM +0100, Sam Halliday wrote: > > i recently rebuilt my system with devfs support... i like it a lot! > > good idea! but unfortunately i have one small gripe... i have my > > /etc/mtab as a symbolic link to /proc/mounts which has always worked > > nicely for me and i can therefore be sure that even on a system > > crash or read-only root mount, my mtab will always be correct... > > however, using devfs (on linux-2.4.19 and i also tested > > 2.4.20-pre11) i get two mounts for the rootfs... here is my > > /proc/mounts rootfs / rootfs rw 0 0 > > /dev/root / ext3 rw 0 0 > > none /dev devfs rw 0 0 > > proc /proc proc rw 0 0 > > /dev/hda7 /home ext3 rw 0 0 > > none /dev/shm tmpfs rw 0 0 > > automount(pid77) /mnt/.autofs autofs rw 0 0 > > is this a bug, or is my symlink a bad idea. this is the way Linux > > from Scratch reccomends doing to /etc/mtab and i agree with their > > arguements. has anyone else experienced this? is it known already, > > and if so, is there a fix in the running? (hopefully before 2.4.20!) > I have noticed it too, and I never saw this before 2.4.19. With 2.4.17 > (or was it 2.4.9?) the root fs showed up as /dev/ide/host0/... > I wasn't even aware that devfs was the culprit here. > I've been running several machines like this for months and have never > noticed a problem, but I'm also curious why this changed recently. well, i got a little too annoyed with the bugs involved in running a symlink form /etc/mtab to /proc/mounts, and i _just_ changed my system to use a real file /etc/mtab. but i can assure you that when i switched to devfs, this problem occured, so to me id say it probably is devfs to blame, sorry! Sam -- Q: How many journalists does it take to screw in a lightbulb? A: Three. One to report it as an inspired government program to bring light to the people, one to report it as a diabolical government plot to deprive the poor of darkness, and one to win a Pulitzer prize for reporting that Electric Company hired a lightbulb-assassin to break the bulb in the first place. From Andrej.Borsenkow@mow.siemens.ru Wed Oct 16 22:26:30 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 16 Oct 2002 22:26:32 -0700 (PDT) Received: from goliath.siemens.de (goliath.siemens.de [192.35.17.28]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9H5QStG029513 for ; Wed, 16 Oct 2002 22:26:29 -0700 Received: from mail3.siemens.de (mail3.siemens.de [139.25.208.14]) by goliath.siemens.de (8.11.6/8.11.6) with ESMTP id g9H5QB726205; Thu, 17 Oct 2002 07:26:11 +0200 (MEST) Received: from MOWD019A.mow.siemens.ru ([139.24.18.3]) by mail3.siemens.de (8.11.6/8.11.6) with ESMTP id g9H5QA416790; Thu, 17 Oct 2002 07:26:11 +0200 (MEST) Received: by mowd019a.mow.siemens.ru with Internet Mail Service (5.5.2653.19) id ; Thu, 17 Oct 2002 09:37:17 +0400 Received: from mw2b210c (163.242.193.12 [163.242.193.12]) by MOWD019A.mow.siemens.ru with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2653.13) id VBKKAJPA; Thu, 17 Oct 2002 09:37:12 +0400 From: Borzenkov Andrey To: "'Sam Halliday'" , devfs@oss.sgi.com Subject: RE: too much mounted Date: Thu, 17 Oct 2002 09:26:00 +0400 Message-ID: <6134254DE87BD411908B00A0C99B044F03A0B491@mowd019a.mow.siemens.ru> MIME-Version: 1.0 Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Priority: 3 (Normal) X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook, Build 10.0.4024 In-Reply-To: <20021016232445.4aac01e8.samuel@ma.hw.ac.uk> X-MIMEOLE: Produced By Microsoft MimeOLE V6.00.2800.1106 Importance: Normal X-archive-position: 18 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: Andrej.Borsenkow@mow.siemens.ru Precedence: bulk X-list: devfs Content-Length: 2363 Lines: 54 > On Wed, 16 Oct 2002 21:22:34 +0000 Michael Carmack > wrote: > > On Wed, Oct 16, 2002 at 08:32:42PM +0100, Sam Halliday wrote: > > > i recently rebuilt my system with devfs support... i like it a lot! > > > good idea! but unfortunately i have one small gripe... i have my > > > /etc/mtab as a symbolic link to /proc/mounts which has always worked > > > nicely for me and i can therefore be sure that even on a system > > > crash or read-only root mount, my mtab will always be correct... > > > however, using devfs (on linux-2.4.19 and i also tested > > > 2.4.20-pre11) i get two mounts for the rootfs... here is my > > > /proc/mounts rootfs / rootfs rw 0 0 > > > /dev/root / ext3 rw 0 0 > > > none /dev devfs rw 0 0 > > > proc /proc proc rw 0 0 > > > /dev/hda7 /home ext3 rw 0 0 > > > none /dev/shm tmpfs rw 0 0 > > > automount(pid77) /mnt/.autofs autofs rw 0 0 > > > is this a bug, or is my symlink a bad idea. this is the way Linux > > > from Scratch reccomends doing to /etc/mtab and i agree with their > > > arguements. has anyone else experienced this? is it known already, > > > and if so, is there a fix in the running? (hopefully before 2.4.20!) > > I have noticed it too, and I never saw this before 2.4.19. With 2.4.17 > > (or was it 2.4.9?) the root fs showed up as /dev/ide/host0/... > > I wasn't even aware that devfs was the culprit here. > > I've been running several machines like this for months and have never > > noticed a problem, but I'm also curious why this changed recently. > well, i got a little too annoyed with the bugs involved in running a > symlink form /etc/mtab to /proc/mounts, and i _just_ changed my system > to use a real file /etc/mtab. but i can assure you that when i switched > to devfs, this problem occured, so to me id say it probably is devfs to > blame, sorry! > It has always been so. The first mount comes from kernel mounting "generic" root without knowing real root fs type. The second one comes from actual mounting. It may be dependent on using initrd, I do not remember currently. About ide-scsi - just add hdX=ide-scsi to kernel boot string to prevent ide-cd from claiming CD-ROM. Modules.devfs already includes scsi-hostadapter alias, so the easiest way is to add probeall scsi-hostadapter drv1 ... ide-scsi ... you can of course add ide-scsi explicitly to modules.devfs. -andrey From samuel@pc-62-31-74-96-ed.blueyonder.co.uk Thu Oct 17 02:28:10 2002 Received: with ECARTIS (v1.0.0; list devfs); Thu, 17 Oct 2002 02:28:11 -0700 (PDT) Received: from pc-62-31-74-96-ed.blueyonder.co.uk (pc-62-31-74-96-ed.blueyonder.co.uk [62.31.74.96]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9H9S8tG022358 for ; Thu, 17 Oct 2002 02:28:09 -0700 Received: from pc-62-31-74-96-ed.blueyonder.co.uk (localhost.localdomain [127.0.0.1]) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6) with ESMTP id g9H9X9OF000548 for ; Thu, 17 Oct 2002 10:33:09 +0100 Received: (from samuel@localhost) by pc-62-31-74-96-ed.blueyonder.co.uk (8.12.6/8.12.6/Submit) id g9H9X9uQ000547; Thu, 17 Oct 2002 10:33:09 +0100 Date: Thu, 17 Oct 2002 10:33:09 +0100 From: Sam Halliday To: devfs@oss.sgi.com Subject: Re: too much mounted Message-Id: <20021017103309.22f8e1bc.samuel@ma.hw.ac.uk> In-Reply-To: <6134254DE87BD411908B00A0C99B044F03A0B491@mowd019a.mow.siemens.ru> References: <20021016232445.4aac01e8.samuel@ma.hw.ac.uk> <6134254DE87BD411908B00A0C99B044F03A0B491@mowd019a.mow.siemens.ru> X-Mailer: Sylpheed version 0.8.5claws (GTK+ 1.2.10; ) X-Operating-System: Linux from Scratch 4.0 Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-archive-position: 19 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: samuel@ma.hw.ac.uk Precedence: bulk X-list: devfs Content-Length: 501 Lines: 16 > About ide-scsi - just add hdX=ide-scsi to kernel boot string to > prevent ide-cd from claiming CD-ROM. i already have... > you can of course add ide-scsi explicitly to modules.devfs. thats what i had to end up doing :/ aw well, shame about the kernel mounting root twice... but it seems the kernel just isnt ready for allowing a symbolic link from /etc/mtab to the /proc/mounts. cheers everyone! Sam -- The doll's trying to kill me, and the toaster's been laughing at me. -- Homer Simpson From per@fukt.bth.se Thu Oct 17 08:29:25 2002 Received: with ECARTIS (v1.0.0; list devfs); Thu, 17 Oct 2002 08:29:29 -0700 (PDT) Received: from ce00521-p19-hertas5.cenara.com (root@ce00521-p19-hertas5.cenara.com [195.38.1.140]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9HFTNtG029644 for ; Thu, 17 Oct 2002 08:29:24 -0700 Received: from localhost (per@localhost [127.0.0.1]) by ce00521-p19-hertas5.cenara.com (8.12.6/8.12.6) with ESMTP id g9HFTLb8011823; Thu, 17 Oct 2002 17:29:21 +0200 Date: Thu, 17 Oct 2002 17:29:21 +0200 (CEST) From: =?iso-8859-1?Q?Per_Lid=E9n?= X-X-Sender: per@ce00521-p19-hertas5.cenara.com To: Brad Chapman cc: devfs@oss.sgi.com Subject: Re: option "root=" doesn't work with devfs device names In-Reply-To: <20021016205856.5910.qmail@web40017.mail.yahoo.com> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-archive-position: 20 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: per@fukt.bth.se Precedence: bulk X-list: devfs Content-Length: 285 Lines: 13 On Wed, 16 Oct 2002, Brad Chapman wrote: [...] > > The patch below fixes this problem. It would be nice if this problem was > > solved before 2.4.20 was released. > > So with this patch a root= entry of /dev/discs/disc0/part1 returns the > correct device entry in devfs? Yes. /Per From racsw@frontiernet.net Mon Oct 21 17:37:41 2002 Received: with ECARTIS (v1.0.0; list devfs); Mon, 21 Oct 2002 17:37:42 -0700 (PDT) Received: from relay03.roc.frontiernet.net (relay03.roc.frontiernet.net [66.133.131.36]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9M0beuR006687 for ; Mon, 21 Oct 2002 17:37:40 -0700 Received: (qmail 15267 invoked from network); 22 Oct 2002 00:37:40 -0000 Received: from unknown (HELO localhost.localdomain) ([65.37.13.251]) (envelope-sender ) by relay03.roc.frontiernet.net (qmail-ldap-1.03) with SMTP for ; 22 Oct 2002 00:37:40 -0000 Content-Type: text/plain; charset="us-ascii" From: Robert To: Devfs Subject: symlinks destroyed on boot for a zip drive Date: Mon, 21 Oct 2002 08:37:48 -0400 User-Agent: KMail/1.4.3 MIME-Version: 1.0 Message-Id: <200210210837.48896.racsw@frontiernet.net> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id g9M0beuR006687 X-archive-position: 21 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: racsw@frontiernet.net Precedence: bulk X-list: devfs Content-Length: 1700 Lines: 31 Hello, I linux guru friend of mine in England sent me to you folks because of the battle I've been waging with my /dev symlinks. He attributes the problem to the quirks of devfs. Here's the problem. I have Mandrake 8.2 installed, and I have, among other devices, a 256Mb internal IDE zip drive installed. When I had 8.2 installed without the Upgrade CD, things were working fine. Since upgrading, I can no longer use my zip drive with supermount. The reason is that something is wiping out my symlink called "hdd4" in the /dev directory. Originally, I had a symlink called hdd4@, a block device, that pointed to "/ide/host0/bus1/target1/lun0/part4". If I go to create the link, I have to first eliminate the /hdd4 directory that is being created each time I boot. I use this syntax: "ln -s /dev/ide/host0/bus1/target1/lun0/part4 hdd4". This successfully creates the link, and while I'm logged on, my zip functions fine as it originally did, all the files are available. When I reboot however, this symlink is wiped out, and replaced by that /hdd4 directory again, and again my zip fails to function. This happens every time. For now, seeing as I can't get around this issue unless you folks can help me out, I created the hdd4@ symlink inside of the /hdd4 directory, and I had to change my fstab file to suit this new arrangement. I can use the zip, but only if I manually mount and umount the drive. I would like to fix this problem so that the supermount feature once again works the way it's supposed to, and I won't have to mount the zip manually again anymore. Can someone help me out with this? P.S. - My system uses devfsd. Thanks, Robert From russell@coker.com.au Tue Oct 22 02:19:15 2002 Received: with ECARTIS (v1.0.0; list devfs); Tue, 22 Oct 2002 02:19:18 -0700 (PDT) Received: from tsv.sws.net.au (tsv.sws.net.au [203.36.46.2]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9M9JEuR009286 for ; Tue, 22 Oct 2002 02:19:15 -0700 Received: from lyta.coker.com.au (localhost [127.0.0.1]) by tsv.sws.net.au (Postfix) with ESMTP id EC29892724; Tue, 22 Oct 2002 19:19:14 +1000 (EST) Received: from lyta (localhost [127.0.0.1]) by lyta.coker.com.au (Postfix) with ESMTP id E92E492B3; Tue, 22 Oct 2002 11:19:06 +0200 (CEST) Content-Type: text/plain; charset="iso-8859-1" From: Russell Coker Reply-To: Russell Coker To: Robert , Devfs Subject: Re: symlinks destroyed on boot for a zip drive Date: Tue, 22 Oct 2002 11:19:06 +0200 User-Agent: KMail/1.4.3 References: <200210210837.48896.racsw@frontiernet.net> In-Reply-To: <200210210837.48896.racsw@frontiernet.net> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: <200210221119.06516.russell@coker.com.au> X-archive-position: 22 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: russell@coker.com.au Precedence: bulk X-list: devfs Content-Length: 1224 Lines: 29 On Mon, 21 Oct 2002 14:37, Robert wrote: > my zip drive with supermount. The reason is that something is wiping out > my symlink called "hdd4" in the /dev directory. Originally, I had a /dev is effectively a ram-disk, every reboot clears it out... > "/ide/host0/bus1/target1/lun0/part4". If I go to create the link, I have > to first eliminate the /hdd4 directory that is being created each time I It sounds like something in your devfsd configuration or your boot scripts is creating it. But as you didn't attach copies of any configuration files to your message it's difficult to even guess what the error might be. What is 8.2 anyway? > boot. I use this syntax: "ln -s /dev/ide/host0/bus1/target1/lun0/part4 That only creates it as a transient entry in a ram disk. When you reboot it disappears. You need to put something like the following in your devfsd config: REGISTER ^ide/.* MKOLDCOMPAT UNREGISTER ^ide/.* RMOLDCOMPAT -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From racsw@frontiernet.net Tue Oct 22 03:40:05 2002 Received: with ECARTIS (v1.0.0; list devfs); Tue, 22 Oct 2002 03:40:06 -0700 (PDT) Received: from relay04.roc.frontiernet.net (relay04.roc.frontiernet.net [66.133.131.37]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9MAe4uR017506 for ; Tue, 22 Oct 2002 03:40:04 -0700 Received: (qmail 2413 invoked from network); 22 Oct 2002 10:40:06 -0000 Received: from unknown (HELO localhost.localdomain) ([65.37.13.251]) (envelope-sender ) by relay04.roc.frontiernet.net (qmail-ldap-1.03) with SMTP for ; 22 Oct 2002 10:40:06 -0000 Content-Type: text/plain; charset="iso-8859-1" From: Robert To: Russell Coker , Devfs Subject: Re: symlinks destroyed on boot for a zip drive Date: Mon, 21 Oct 2002 14:40:13 -0400 User-Agent: KMail/1.4.3 References: <200210210837.48896.racsw@frontiernet.net> <200210221119.06516.russell@coker.com.au> In-Reply-To: <200210221119.06516.russell@coker.com.au> MIME-Version: 1.0 Message-Id: <200210211440.13958.racsw@frontiernet.net> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id g9MAe4uR017506 X-archive-position: 23 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: racsw@frontiernet.net Precedence: bulk X-list: devfs Content-Length: 1582 Lines: 40 Hi Russell, I printed out the FAQ last night and went over it, but I have to read it again more closely. I apologize for not giving you config scripts, but as I have no prior knowledge about this subject, I didn't know what you would want. When I referred to 8.2, I meant Linux Mandrake version 8.2. If you could tell me what config scripts to post, I will be happy to do that. As I said, I need to reread the FAQ again, study it, actually, so I understand a little better what's going on here. Thanks, Robert On Tuesday 22 October 2002 5:19 am, you wrote: > On Mon, 21 Oct 2002 14:37, Robert wrote: > > my zip drive with supermount. The reason is that something is wiping > > out my symlink called "hdd4" in the /dev directory. Originally, I had a > > /dev is effectively a ram-disk, every reboot clears it out... > > > "/ide/host0/bus1/target1/lun0/part4". If I go to create the link, I have > > to first eliminate the /hdd4 directory that is being created each time I > > It sounds like something in your devfsd configuration or your boot scripts > is creating it. But as you didn't attach copies of any configuration files > to your message it's difficult to even guess what the error might be. > > What is 8.2 anyway? > > > boot. I use this syntax: "ln -s /dev/ide/host0/bus1/target1/lun0/part4 > > That only creates it as a transient entry in a ram disk. When you reboot > it disappears. You need to put something like the following in your devfsd > config: > REGISTER ^ide/.* MKOLDCOMPAT > UNREGISTER ^ide/.* RMOLDCOMPAT From russell@coker.com.au Wed Oct 23 03:09:43 2002 Received: with ECARTIS (v1.0.0; list devfs); Wed, 23 Oct 2002 03:09:44 -0700 (PDT) Received: from tsv.sws.net.au (tsv.sws.net.au [203.36.46.2]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9NA9fuR007829 for ; Wed, 23 Oct 2002 03:09:42 -0700 Received: from lyta.coker.com.au (localhost [127.0.0.1]) by tsv.sws.net.au (Postfix) with ESMTP id 66A9A92713; Wed, 23 Oct 2002 20:09:46 +1000 (EST) Received: from lyta (localhost [127.0.0.1]) by lyta.coker.com.au (Postfix) with ESMTP id 6C403889A; Wed, 23 Oct 2002 12:09:38 +0200 (CEST) Content-Type: text/plain; charset="iso-8859-1" From: Russell Coker Reply-To: Russell Coker To: Robert , Devfs Subject: Re: my devfsd.conf file Date: Wed, 23 Oct 2002 12:09:38 +0200 User-Agent: KMail/1.4.3 References: <200210211959.03135.racsw@frontiernet.net> In-Reply-To: <200210211959.03135.racsw@frontiernet.net> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: <200210231209.38038.russell@coker.com.au> X-archive-position: 24 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: russell@coker.com.au Precedence: bulk X-list: devfs Content-Length: 816 Lines: 20 On Tue, 22 Oct 2002 01:59, Robert wrote: > # Enable full compatibility mode for old device names. You may comment > these # out if you don't use the old device names. Make sure you know what > you're # doing! > REGISTER .* MKOLDCOMPAT > UNREGISTER .* RMOLDCOMPAT The above will do what you need. The fact that it doesn't suggests that perhaps the conf.d files included below might be stuffing it up somehow. Please send us copies of them as attachments. > # Include package-generated files from /etc/devfs/conf.d > OPTIONAL_INCLUDE /etc/devfs/conf.d/ -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page From racsw@frontiernet.net Thu Oct 24 16:37:37 2002 Received: with ECARTIS (v1.0.0; list devfs); Thu, 24 Oct 2002 16:37:39 -0700 (PDT) Received: from relay03.roc.frontiernet.net (relay03.roc.frontiernet.net [66.133.131.36]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9ONbauR003385 for ; Thu, 24 Oct 2002 16:37:36 -0700 Received: (qmail 9926 invoked from network); 24 Oct 2002 23:37:48 -0000 Received: from unknown (HELO localhost.localdomain) ([65.37.13.251]) (envelope-sender ) by relay03.roc.frontiernet.net (qmail-ldap-1.03) with SMTP for ; 24 Oct 2002 23:37:48 -0000 Content-Type: text/plain; charset="us-ascii" From: Robert To: Devfs Subject: Symlinks being destroyed on boot in /dev Date: Wed, 23 Oct 2002 07:37:53 -0400 User-Agent: KMail/1.4.3 MIME-Version: 1.0 Message-Id: <200210230737.53169.racsw@frontiernet.net> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id g9ONbauR003385 X-archive-position: 25 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: racsw@frontiernet.net Precedence: bulk X-list: devfs Content-Length: 10468 Lines: 250 Hello, I'm reposting this problem, as another maillist person (Russell Coker) made a valiant effort to solve this problem for me, but this has him stumped as well. Well, this all started when I installed the Linux Mandrake 8.2 Upgrade CDROM a month ago. I found out quickly that I could not access my internal IDE 256Mb Zip drive. So I started to read as much as I could find about devices and such. I also had the advantage of my wife's computer, which is identical to mine, and is also running Mandrake 8.1, and her internal zip is working just fine. So I went into the /dev directory, and located a symlink called "hdd4@" which pointed to /ide/host0/bus1/target1/lun0/part4. My system, OTOH, had no "hdd4@" symlink. So I used ln -s to create this missing link. I had trouble doing this, until I noticed I had a directory called /dev/hdd4 with nothing in it that my wife's computer did not have. So I had to delete that directory before creation of the symlink would work, apparently because of a conflict. Well, the symlink was created, but was flashing red, which indicated the target didn't exist. So I checked the target dir, and the block device file "part4" was not there. So I read the entire Linux Device Documentation, and not seeing anything related to a zip for major/minor numbers, I used the numbers 22,68 to create "part4" which I got from the wife's computer. Once I did this, the symlink was the normal color, and had all the correct attributes as my wife's system. So, after many hours of fstab experimentation earlier in an effort to narrow down this problem, I returned the fstab line to it's original form, and Presto, I had a fully functioning ZIP drive and supermount once again. I was elated to say the least. The next day I powered up again to check email, and I couldn't access the zip again. Something had erased the symlink "hdd4@" and replaced it with the /dev/hdd4 directory. So essentially, I had to do everything all over again, with the exception of creating the "part4" block device file. As a result, it would only work while running, but on reboot, the system would repeat the same process again. So what I ended up doing, after many nights of frustrating experimentation, was creating the "hdd4@" symlink inside the /dev/hdd4 directory. This allowed me access to the zip only if I mounted the zip manually, supermount and the original fstab line would not function with this configuration. So I was telling all this to my Linux guru friend in England, and he said that this seemed to be a devfs problem, and to do some research on this, which I have and why I ended up here. I've stated this whole thing in an earlier post, and Russell Coker came to my rescue, examined my /etc/devfsd.conf file, plus the others that were in the /etc/devfs/con.d/ directory. I've read the Linux Devfs FAQ by Richard Gooch three times now, and while I have a much better understanding of how devfs works, I have not been able to fix my problem. If someone could help me out here, I would greatly appreciate it, because otherwise, my only other option is to completely reload my whole OS, something I desperately don't want to do after finally getting everything functioning and the tools loaded that I needed for my system. I've put in almost a whole month of nights and weekends trying to solve this on my own, but I've come up short. I just don't know enough at the system level to fix this. The FAQ mentions two files that are usually questioned when a problem arises, so I have included the the entire /etc/devfs.conf file below, as well as a small section of my /etc/rc.d/rc.sysinit that appeared to me to be related to devfs. If this is not proper to do on this list, I would appreciate a comment so I handle this correctly in the future. Hopefully, someone can help me troubleshoot this. Fingers are crossed ;-) Thanks, Robert # Sample /etc/devfsd.conf configuration file. # Richard Gooch 17-FEB-2002 # # Enable full compatibility mode for old device names. You may comment these # out if you don't use the old device names. Make sure you know what you're # doing! REGISTER .* MKOLDCOMPAT UNREGISTER .* RMOLDCOMPAT # You may comment out the above and uncomment the following if you've # configured your system to use the original "new" devfs names or the really # new names #REGISTER ^vc/ MKOLDCOMPAT #UNREGISTER ^vc/ RMOLDCOMPAT #REGISTER ^pty/ MKOLDCOMPAT #UNREGISTER ^pty/ RMOLDCOMPAT #REGISTER ^misc/ MKOLDCOMPAT #UNREGISTER ^misc/ RMOLDCOMPAT # You may comment these out if you don't use the original "new" names REGISTER .* MKNEWCOMPAT UNREGISTER .* RMNEWCOMPAT # Enable module autoloading. You may comment this out if you don't use # autoloading LOOKUP .* MODLOAD # Uncomment the following if you want to set the group to "tty" for the # pseudo-tty devices. This is necessary so that mesg(1) can later be used to # enable/disable talk requests and wall(1) messages. REGISTER ^pty/s.* PERMISSIONS -1.tty 0600 REGISTER ^pts/.* PERMISSIONS -1.tty 0600 # # Uncomment this if you want permissions to be saved and restored # Do not do this for pseudo-terminal devices REGISTER ^pt[sy] IGNORE CREATE ^pt[sy] IGNORE CHANGE ^pt[sy] IGNORE DELETE ^pt[sy] IGNORE REGISTER .* COPY /lib/dev-state/$devname $devpath CREATE .* COPY $devpath /lib/dev-state/$devname CHANGE .* COPY $devpath /lib/dev-state/$devname DELETE .* CFUNCTION GLOBAL unlink /lib/dev-state/$devname RESTORE /lib/dev-state # # Uncomment this if you want the old /dev/cdrom symlink REGISTER ^cdroms/cdrom0$ CFUNCTION GLOBAL mksymlink $devname cdrom UNREGISTER ^cdroms/cdrom0$ CFUNCTION GLOBAL unlink cdrom REGISTER ^v4l/video0$ CFUNCTION GLOBAL symlink v4l/video0 video UNREGISTER ^v4l/video0$ CFUNCTION GLOBAL unlink video # ALSA stuff #LOOKUP snd MODLOAD ACTION snd # Manage USB mouse REGISTER ^input/mouse0$ CFUNCTION GLOBAL mksymlink input/mouse0 usbmouse UNREGISTER ^input/mouse0$ CFUNCTION GLOBAL unlink usbmouse # dynamic desktop and co #REGISTER .*/part.* EXECUTE /etc/dynamic/scripts/part.script add $devpath #UNREGISTER .*/part.* EXECUTE /etc/dynamic/scripts/part.script del $devpath REGISTER v4l/video.* EXECUTE /etc/dynamic/scripts/webcam.script add $devpath UNREGISTER v4l/video.* EXECUTE /etc/dynamic/scripts/webcam.script del $devpath REGISTER usb/scanner.* EXECUTE /etc/dynamic/scripts/scanner.script add $devpath UNREGISTER usb/scanner.* EXECUTE /etc/dynamic/scripts/scanner.script del $devpath REGISTER usb/rio500 EXECUTE /etc/dynamic/scripts/rio500.script add $devpath UNREGISTER usb/rio500 EXECUTE /etc/dynamic/scripts/rio500.script del $devpath REGISTER usb/tts/[13579] EXECUTE /etc/dynamic/scripts/visor.script add $devpath UNREGISTER usb/tts/[13579] EXECUTE /etc/dynamic/scripts/visor.script del $devpath REGISTER (usb/lp.*|printers/.*) EXECUTE /etc/dynamic/scripts/lp.script add $devpath UNREGISTER (usb/lp.*|printers/.*) EXECUTE /etc/dynamic/scripts/lp.script del $devpath # Uncomment this to let PAM manage devfs REGISTER .* CFUNCTION /lib/security/pam_console_apply_devfsd.so pam_console_apply_single $devpath # Uncomment this to manage USB mouse REGISTER ^input/mouse0$ CFUNCTION GLOBAL mksymlink $devname usbmouse UNREGISTER ^input/mouse0$ CFUNCTION GLOBAL unlink usbmouse REGISTER ^input/mice$ CFUNCTION GLOBAL mksymlink $devname usbmouse UNREGISTER ^input/mice$ CFUNCTION GLOBAL unlink usbmouse # If you have removable media and want to force media revalidation when looking # up new or old compatibility names, uncomment the following lines # SCSI NEWCOMPAT /dev/sd/* names LOOKUP ^(sd/c[0-9]+b[0-9]+t[0-9]+u[0-9]+)p[0-9]+$ EXECUTE /bin/dd if=$mntpnt/\1 of=/dev/null count=1 # SCSI OLDCOMPAT /dev/sd?? names LOOKUP ^(sd[a-z]+)[0-9]+$ EXECUTE /bin/dd if=$mntpnt/\1 of=/dev/null count=1 # IDE NEWCOMPAT /dev/ide/hd/* names LOOKUP ^(ide/hd/c[0-9]+b[0-9]+t[0-9]+u[0-9]+)p[0-9]+$ EXECUTE /bin/dd if=$mntpnt/\1 of=/dev/null count=1 # IDE OLDCOMPAT /dev/hd?? names LOOKUP ^(hd[a-z])[0-9]+$ EXECUTE /bin/dd if=$mntpnt/\1 of=/dev/null count=1 # Include package-generated files from /etc/devfs/conf.d OPTIONAL_INCLUDE /etc/devfs/conf.d/ /////////////////////////////////////////////////////////////////////////////////////////////// This next section is two snippets from the /etc/rc.d/rc.sysinit file that I thought pertained to devfsd. ////////////////////////////////////////////////////////////////////////////////////////////// # Try to get devfsd if [ -c /dev/.devfsd ]; then if [ -x /sbin/devfsd ]; then action "Running DevFs daemon" devfsd /dev if [ -f /etc/sysconfig/mouse ];then . /etc/sysconfig/mouse device=`echo $device | sed -e 's/psaux/misc\/psaux/' -e 's/ttyS\(.\)/tts\/\1/g'` if [ -n "$device" ]; then rm -f /dev/mouse ln -s /dev/$device /dev/mouse fi fi else if ! grep -q "nodevfsd" /proc/cmdline;then # devfs is used but devfsd isn't installed gprintf " #### your system is currently using devfs but devfsd isn't availlable" gprintf " #### you should disable devfs (add 'devfs=nomount' to the paramaters" gprintf " #### that your bootloader gives to the kernel" # Let devfs being tested, but disable it on /dev action " Disabling devfs (was mounted on /dev)" umount /dev fi fi fi get_devfs_name() { declare -i bus target let "bus = (36#${1#hd} - 36#a)/2" let "target = (36#${1#hd} - 36#a)%2" echo "ide/host0/bus$bus/target$target/lun0" } # directly set IDE parameter via /proc interface # 1 = drive name # 2 = setting name # 3 = name setting value set_ide_setting() { [ -e /proc/ide/$1/settings ] || return 1 declare device=$1 setting=$2 value=$3 # map hdparm parameters from harddisk* case $setting in MULTIPLE_IO ) setting=multcount ;; USE_DMA ) setting=using_dma ;; EIDE_32BIT ) setting=io_32bit ;; LOOKAHEAD | EXTRA_PARAMS ) return 1 ;; esac grep -q -- "^$setting[[:space:]]*" "/proc/ide/$1/settings" || return 1 set -- $(grep -- "^$setting[[:space:]]*" "/proc/ide/$device/settings") [ "$5" = rw -o "$5" = w ] || return 1 echo -n "$setting:$value" > "/proc/ide/$device/settings" [ "$5" = w ] && return 0 set -- $(grep -- "^$setting[[:space:]]*" "/proc/ide/$device/settings") [ "$2" = "$value" ] && return 0 return 1 } From kpfleming@cox.net Fri Oct 25 10:16:32 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 25 Oct 2002 10:16:33 -0700 (PDT) Received: from fed1mtao03.cox.net (fed1mtao03.cox.net [68.6.19.242]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9PHGVuR010245 for ; Fri, 25 Oct 2002 10:16:31 -0700 Received: from jeeves.kpf.internal ([24.56.45.215]) by fed1mtao03.cox.net (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with ESMTP id <20021025171644.GLAN21152.fed1mtao03.cox.net@jeeves.kpf.internal> for ; Fri, 25 Oct 2002 13:16:44 -0400 Received: from [192.168.130.55] (helo=cox.net) by jeeves.kpf.internal with esmtp (Exim 4.05) id 18584h-0002Cu-00 for devfs@oss.sgi.com; Fri, 25 Oct 2002 10:16:43 -0700 Message-ID: <3DB97C7B.3090305@cox.net> Date: Fri, 25 Oct 2002 10:16:43 -0700 From: "Kevin P. Fleming" User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.2b) Gecko/20021016 X-Accept-Language: en-us, en MIME-Version: 1.0 To: devfs@oss.sgi.com Subject: Re: Symlinks being destroyed on boot in /dev References: <200210230737.53169.racsw@frontiernet.net> In-Reply-To: <200210230737.53169.racsw@frontiernet.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 26 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: kpfleming@cox.net Precedence: bulk X-list: devfs Content-Length: 1894 Lines: 36 Robert wrote: > Hello, > I'm reposting this problem, as another maillist person (Russell Coker) made > a valiant effort to solve this problem for me, but this has him stumped as > well. > Well, this all started when I installed the Linux Mandrake 8.2 Upgrade > CDROM a month ago. I found out quickly that I could not access my internal > IDE 256Mb Zip drive. So I started to read as much as I could find about Stop! You've gone way, way beyond what was necessary here! The root of your problem is that in current 2.4.x kernels, if you don't have a piece of media in your removable media ATAPI drive (your ZIP drive) when you boot the kernel, and you are using devfs, the drive will not show up in /dev/discs. Also, if you are using devfsd, no compatibility symlinks will be created, since the main entry was not created. Even if you do have media in the drive when you boot the kernel, if you later change media the kernel will not do a very good job of updating the /dev entries to match the new media. This has been discussed before on this list, I'm surprised you didn't find it when you checked the list archives . The simple solution is to wait for the 2.4.20 kernel to be released, it should have a patch to make the drive appear even without media in it at boot time. There are also patches on my web site (URL below) for making the media change handling better; they have been sent to the 2.4 kernel maintainer, but I don't know when they'll get incorporated into the mainline kernel. In the interim, you can apply them yourself and get these problems pretty well solved. If you're not up to compiling and installing your own kernel, then you're going to have a problem. devfs and ZIP drives don't get along well without the patches (which is not devfs' fault, by the way). http://members.cox.net/kpfleming/ide-floppy has the patches mentioned above. From racsw@frontiernet.net Fri Oct 25 14:37:47 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 25 Oct 2002 14:37:50 -0700 (PDT) Received: from relay04.roc.frontiernet.net (relay04.roc.frontiernet.net [66.133.131.37]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9PLbkuR024076 for ; Fri, 25 Oct 2002 14:37:47 -0700 Received: (qmail 31773 invoked from network); 25 Oct 2002 21:38:01 -0000 Received: from unknown (HELO localhost.localdomain) ([65.37.13.251]) (envelope-sender ) by relay04.roc.frontiernet.net (qmail-ldap-1.03) with SMTP for ; 25 Oct 2002 21:38:01 -0000 Content-Type: text/plain; charset="iso-8859-1" From: Robert To: "Kevin P. Fleming" , Devfs Subject: Re: Symlinks being destroyed on boot in /dev Date: Thu, 24 Oct 2002 01:38:05 -0400 User-Agent: KMail/1.4.3 References: <200210230737.53169.racsw@frontiernet.net> <3DB97C7B.3090305@cox.net> In-Reply-To: <3DB97C7B.3090305@cox.net> MIME-Version: 1.0 Message-Id: <200210240138.05869.racsw@frontiernet.net> Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by oss.sgi.com id g9PLbkuR024076 X-archive-position: 27 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: racsw@frontiernet.net Precedence: bulk X-list: devfs Content-Length: 2654 Lines: 60 Hi Kevin, Thanksyou for you reply. While I did not search the archives for this problem, I feel as though I have read more material in my Linux books and other documents trying to solve this than I did in the first two years of college. ;-) In my experience with technical newsgroups, the criticism most often levied to those posting is not supplying either enough information or information not detailed enough to enable others to adequately determine the problem without a series of more questions. It was my intent to describe the problem as best I could. Sorry it was so lengthy. Thankyou for the patches. Thanks again, Robert On Friday 25 October 2002 1:16 pm, you wrote: > Robert wrote: > > Hello, > > I'm reposting this problem, as another maillist person (Russell Coker) > > made a valiant effort to solve this problem for me, but this has him > > stumped as well. > > Well, this all started when I installed the Linux Mandrake 8.2 Upgrade > > CDROM a month ago. I found out quickly that I could not access my > > internal IDE 256Mb Zip drive. So I started to read as much as I could > > find about > > Stop! You've gone way, way beyond what was necessary here! > > The root of your problem is that in current 2.4.x kernels, if you don't > have a piece of media in your removable media ATAPI drive (your ZIP drive) > when you boot the kernel, and you are using devfs, the drive will not show > up in /dev/discs. Also, if you are using devfsd, no compatibility symlinks > will be created, since the main entry was not created. Even if you do have > media in the drive when you boot the kernel, if you later change media the > kernel will not do a very good job of updating the /dev entries to match > the new media. > > This has been discussed before on this list, I'm surprised you didn't find > it when you checked the list archives . The simple solution is to wait > for the 2.4.20 kernel to be released, it should have a patch to make the > drive appear even without media in it at boot time. > > There are also patches on my web site (URL below) for making the media > change handling better; they have been sent to the 2.4 kernel maintainer, > but I don't know when they'll get incorporated into the mainline kernel. In > the interim, you can apply them yourself and get these problems pretty well > solved. > > If you're not up to compiling and installing your own kernel, then you're > going to have a problem. devfs and ZIP drives don't get along well without > the patches (which is not devfs' fault, by the way). > > http://members.cox.net/kpfleming/ide-floppy has the patches mentioned > above. From kpfleming@cox.net Fri Oct 25 14:48:44 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 25 Oct 2002 14:48:46 -0700 (PDT) Received: from fed1mtao01.cox.net (fed1mtao01.cox.net [68.6.19.244]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9PLmiuR024161 for ; Fri, 25 Oct 2002 14:48:44 -0700 Received: from jeeves.kpf.internal ([24.56.45.215]) by fed1mtao01.cox.net (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with ESMTP id <20021025214857.GHUU14888.fed1mtao01.cox.net@jeeves.kpf.internal>; Fri, 25 Oct 2002 17:48:57 -0400 Received: from [192.168.130.55] (helo=cox.net) by jeeves.kpf.internal with esmtp (Exim 4.05) id 185CK9-0002bZ-00; Fri, 25 Oct 2002 14:48:57 -0700 Message-ID: <3DB9BC48.8090100@cox.net> Date: Fri, 25 Oct 2002 14:48:56 -0700 From: "Kevin P. Fleming" User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.2b) Gecko/20021016 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Robert CC: Devfs Subject: Re: Symlinks being destroyed on boot in /dev References: <200210230737.53169.racsw@frontiernet.net> <3DB97C7B.3090305@cox.net> <200210240138.05869.racsw@frontiernet.net> In-Reply-To: <200210230737.53169.racsw@frontiernet.net> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-archive-position: 28 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: kpfleming@cox.net Precedence: bulk X-list: devfs Content-Length: 1227 Lines: 33 Robert wrote: > Hi Kevin, > Thanksyou for you reply. While I did not search the archives for this > problem, I feel as though I have read more material in my Linux books and > other documents trying to solve this than I did in the first two years of > college. ;-) No problem, we all have to learn somehow. > > In my experience with technical newsgroups, the criticism most often levied > to those posting is not supplying either enough information or information > not detailed enough to enable others to adequately determine the problem > without a series of more questions. It was my intent to describe the > problem as best I could. Sorry it was so lengthy. Absolutely true... Unfortunately when I read your message I knew the solution to the problem in the first few sentences. I'm actually surprised I didn't see your previous discussion of this with Russell; did that discussion happen on the devfs list? I've been watching for these things to come up ever since I created the patches to fix the problem and I usually jump right in and suggest the fix. > > > Thankyou for the patches. You're welcome. Let us know if anything still doesn't work the way you expect after you give them a try. From russell@coker.com.au Fri Oct 25 16:31:38 2002 Received: with ECARTIS (v1.0.0; list devfs); Fri, 25 Oct 2002 16:31:40 -0700 (PDT) Received: from tsv.sws.net.au (tsv.sws.net.au [203.36.46.2]) by oss.sgi.com (8.12.5/8.12.5) with SMTP id g9PNVbuR025337 for ; Fri, 25 Oct 2002 16:31:38 -0700 Received: from lyta.coker.com.au (localhost [127.0.0.1]) by tsv.sws.net.au (Postfix) with ESMTP id 635D192724; Sat, 26 Oct 2002 09:31:53 +1000 (EST) Received: from lyta (localhost [127.0.0.1]) by lyta.coker.com.au (Postfix) with ESMTP id BDF7E947B; Sat, 26 Oct 2002 01:31:44 +0200 (CEST) Content-Type: text/plain; charset="iso-8859-1" From: Russell Coker Reply-To: Russell Coker To: "Kevin P. Fleming" , Robert Subject: Re: Symlinks being destroyed on boot in /dev Date: Sat, 26 Oct 2002 01:31:44 +0200 User-Agent: KMail/1.4.3 Cc: Devfs References: <200210230737.53169.racsw@frontiernet.net> <200210240138.05869.racsw@frontiernet.net> <3DB9BC48.8090100@cox.net> In-Reply-To: <3DB9BC48.8090100@cox.net> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Message-Id: <200210260131.44404.russell@coker.com.au> X-archive-position: 29 X-ecartis-version: Ecartis v1.0.0 Sender: devfs-bounce@oss.sgi.com Errors-to: devfs-bounce@oss.sgi.com X-original-sender: russell@coker.com.au Precedence: bulk X-list: devfs Content-Length: 951 Lines: 21 On Fri, 25 Oct 2002 23:48, Kevin P. Fleming wrote: > Absolutely true... Unfortunately when I read your message I knew the > solution to the problem in the first few sentences. I'm actually > surprised I didn't see your previous discussion of this with Russell; did > that discussion happen on the devfs list? I've been watching for these > things to come up ever since I created the patches to fix the problem and I > usually jump right in and suggest the fix. There was only a couple of messages on the list, then we discussed it by private mail until I realised I couldn't solve it. After seeing your answer I recalled the previous discussion... I should have checked the list archives! -- http://www.coker.com.au/selinux/ My NSA Security Enhanced Linux packages http://www.coker.com.au/bonnie++/ Bonnie++ hard drive benchmark http://www.coker.com.au/postal/ Postal SMTP/POP benchmark http://www.coker.com.au/~russell/ My home page