From owner-freebsd-questions  Tue Nov  3 12:46:54 1998
Return-Path: <owner-freebsd-questions@FreeBSD.ORG>
Received: (from majordom@localhost)
          by hub.freebsd.org (8.8.8/8.8.8) id MAA07608
          for freebsd-questions-outgoing; Tue, 3 Nov 1998 12:46:54 -0800 (PST)
          (envelope-from owner-freebsd-questions@FreeBSD.ORG)
Received: from goose.Stanford.EDU (goose.Stanford.EDU [171.64.99.76])
          by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA07603
          for <questions@freebsd.org>; Tue, 3 Nov 1998 12:46:53 -0800 (PST)
          (envelope-from yergeau@gloworm.Stanford.EDU)
Received: from localhost (yergeau@localhost)
	by goose.Stanford.EDU (8.8.8/8.8.8) with ESMTP id MAA22570;
	Tue, 3 Nov 1998 12:46:46 -0800 (PST)
Message-Id: <199811032046.MAA22570@goose.Stanford.EDU>
To: questions@FreeBSD.ORG
cc: yergeau@gloworm.Stanford.EDU
Subject: making device nodes with boot/fixit floppies (3.0-RELEASE)
Reply-To: yergeau@gloworm.Stanford.EDU
Date: Tue, 03 Nov 1998 12:46:45 -0800
From: Dan Yergeau <yergeau@gloworm.Stanford.EDU>
Sender: owner-freebsd-questions@FreeBSD.ORG
Precedence: bulk
X-Loop: FreeBSD.ORG


Am I missing something or is making device nodes (for disk slices)
under a "emergency recovery" scenario(*) with boot/fixit supposed to be
this complicated/challenging?

I didn't seen anything "obvious" in sysinstall, so I ended up 
 - booted using the boot floppy (well, it was already booted)
 - started a holographic shell
   (no /dev/MAKEDEV, so ...)
 - used sysinstall to mount the fixit floppy
 - copied fixit floppy's dev/MAKEDEV to /dev
   (discovered that MAKEDEV doesn't work due to ...)
 - added fixit floppy directories to PATH
   (and ...)
 - edited /dev/MAKEDEV to change some hardcoded paths and
   chmod/chgrp since the boot floppy's /etc/{passwd,group} ain't
   quite complete (if they even exist)
 - (finally!) "sh MAKEDEV da0s2a"

I suppose I could have faked an install, adding the
disks/partitions, then commit'ing w/o extracting anything, but that
also seems needlessly complicated and is not obvious.


Dan


(*) This instance of emergency recovery was brought on by my sloppy
attempt to wire the disk at scsi target 0 to da1 (first bios disk is
an IDE).  I did modify /etc/fstab, but I had forgotten to create the
device nodes for da1s2, so even booting from the hard disk in single
user mode wouldn't work for recovery (couldn't get it to remount /
in read/write mode).

To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-questions" in the body of the message