From owner-freebsd-questions  Tue Aug 19 16:15:39 1997
Return-Path: <owner-freebsd-questions>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.5/8.8.5) id QAA12491
          for questions-outgoing; Tue, 19 Aug 1997 16:15:39 -0700 (PDT)
Received: from whqvax.picker.com (whqvax.picker.com [144.54.1.1])
          by hub.freebsd.org (8.8.5/8.8.5) with SMTP id QAA12484
          for <questions@freebsd.org>; Tue, 19 Aug 1997 16:15:34 -0700 (PDT)
Received: from ct.picker.com by whqvax.picker.com with SMTP;
          Tue, 19 Aug 1997 19:15:03 -0400 (EDT)
Received: from elmer.ct.picker.com ([144.54.57.34]) by ct.picker.com (4.1/SMI-4.1)
	id AA16765; Tue, 19 Aug 97 19:15:01 EDT
Received: by elmer.ct.picker.com (SMI-8.6/SMI-SVR4)
	id TAA07644; Tue, 19 Aug 1997 19:12:12 -0400
Message-Id: <19970819191211.29132@ct.picker.com>
Date: Tue, 19 Aug 1997 19:12:11 -0400
From: Randall Hopper <rhh@ct.picker.com>
To: questions@freebsd.org
Subject: Question about "sysinstall" Errors
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 0.81
Sender: owner-freebsd-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

     I'm trying to do something I think should be very simple -- run
sysinstall on an installed system to allocate a sizable free slice of disk
to FreeBSD.  This will be "wd1s2", with the intent of having:

      wd1s2b: 128Meg   - SWAP 
      wd1s2e: the rest - UFS @ /share3

FDISKING in sysinstall worked flawlessly:

      Custom-Partition-wd1-Create-<CR>-<CR>-Write-Yes

Then, I got the new slice layed out in the disklabel editor:

      Label-Create-128M-SWAP-Create-<CR>-FS-/share3

So far, looks good.  I try to "Write" and I get:

      "Unable to add /dev/wd1s2b as a swap device: Device not configured"

ktracing sysinstall, I see it's trying to:

      swapon( "/dev/wd1s2b" )

I'm wonderering why--I can do that later.  Going on, it then immediately
forks off:

      newfs -b 8192 -f 1024 /dev/rwd1s2e

It appears there was no "disklabel" before this, so newfs fails.  I guess
sysinstall doesn't check the newfs command for failure since it then goes
on and tries to mount the partition:

      mount( ..., "/share3", ..., "/dev/wd1s2e" )

which returns Invalid argument yielding the:

      "Error mounting /dev/wd1s2e on /share3 : Invalid argument" dialog.  

If I try the "newfs -b 8192 -f 1024 /dev/rwd1s2e" by hand in the shell, I 
get:

      "newfs: /dev/rwd1s2e: `e' partition is unavailable"

The /dev/rwd1s2e does exist, and "disklabel wd1s2" in the shell confirms
the slice was never disklabeled (all blocks are in "c").

Am I doing something wrong here?  I'd welcome any tips, suggestions, and
even "Enlightening Flames".  Before I resort to disklabeling via cmd-line,
I'd like to make sure that sysinstall isn't an option -- it's definitely
the kinder, gentler, safer way to go for users if it works.

Thanks,

Randall