From owner-freebsd-questions  Sun Apr  6 01:59:00 1997
Return-Path: <owner-questions>
Received: (from root@localhost)
          by freefall.freebsd.org (8.8.5/8.8.5) id BAA12450
          for questions-outgoing; Sun, 6 Apr 1997 01:59:00 -0800 (PST)
Received: from helbig.informatik.ba-stuttgart.de (helbig.informatik.ba-stuttgart.de [141.31.166.22])
          by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id BAA12445
          for <questions@freebsd.org>; Sun, 6 Apr 1997 01:58:55 -0800 (PST)
Received: (from helbig@localhost)
	by helbig.informatik.ba-stuttgart.de (8.8.5/8.8.5) id LAA00798;
	Sun, 6 Apr 1997 11:58:04 +0200 (MET DST)
From: Wolfgang Helbig <helbig@MX.BA-Stuttgart.De>
Message-Id: <199704060958.LAA00798@helbig.informatik.ba-stuttgart.de>
Subject: Re: Corrupted swapspace
In-Reply-To: <199704060322.MAA22496@topdown.bns.com.au> from Kris Kennaway at "Apr 6, 97 12:52:57 pm"
To: kkenn@bns.com.au (Kris Kennaway)
Date: Sun, 6 Apr 1997 11:58:04 +0200 (MET DST)
Cc: questions@freebsd.org
X-Mailer: ELM [version 2.4ME+ PL30 (25)]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Sender: owner-questions@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

Hi,

> I've just been playing around with my hard disk partitions (resized the 
> primary DOS partition on my first hard drive using partition magic and 
> changed the cluster size), which is the only thing I can think of that 
> was actually changed in the meantime. I added another FreeBSD partition 
> on a second drive and moved around some directories, but I can't think 
> how that could have caused the problem below:
> 
> The FreeBSD swap space on my first hard drive seems to be corrupt: this 
> is what disklabel -r /wd0s2a makes of it:
> 
> [morden] 11:55 ~ disklabel -r /dev/wd0s2a
> # /dev/wd0s2a:
> type: ESDI
> disk: wd0s2
> label:
> flags:
> bytes/sector: 512
> sectors/track: 63
> tracks/cylinder: 64
> sectors/cylinder: 4032
> cylinders: 354
> sectors/unit: 1427328
> rpm: 3600
> interleave: 1
> trackskew: 0
> cylinderskew: 0
> headswitch: 0           # milliseconds
> track-to-track seek: 0  # milliseconds
> drivedata: 0
> 
> 8 partitions:
> #        size   offset    fstype   [fsize bsize bps/cpg]
>   a:    65536        0    4.2BSD        0     0     0   # (Cyl.    0 - 16*)
>   b:    88256    65536      swap                        # (Cyl.   16*- 38*)
>   c:  1427328        0    unused        0     0         # (Cyl.    0 - 353)
>   e:    61440   153792    4.2BSD        0     0     0   # (Cyl.   38*- 53*)
>   f:  1212096   215232    4.2BSD        0     0     0   # (Cyl.   53*- 353*)
> 
> [morden] 11:57 ~ disklabel -r /dev/wd0s2b
                                     ^ rwd0s2b
> Bad pack magic number (label is damaged, or pack is unlabeled)

That is ok, the swap partition does not have a disklabel, only the slice
has, e. g. 

disklabel -r /dev/rwd0s2

is the disklabel for the second slice.

BTW use raw devices as argument for disklabel(8). I know, disklabel(8) tries
to be smart and will automagically include the "r", but this "feature"
breaks my golden rule:

Use a raw device argument for fstab(8), disklabel(8), newfs(8).
Use the block device for mount(8).

> 
> By contrast, disklabel -r on my other swap partition on the second hard 
> drive correctly gives me the disklabel summary for the drive.
> 
That is strange! I think your second drive is not o. k.

> By rights, I should have 88M total swap with these two partitions, but 
                           ^^^
You have already 88M on your first swap partition, so the total should
be greater.

> top was reporting it as 133M total. Presumably as a consequence, my 
> processes kept on dying with signal 10 and 11 (I'm assuming this is what 
> hapened when they tried to enter the phantom swap-space). I've removed 
> the first swapon entry from fstab and it seems to be working fine so far.
> 

Use pstat -s to see what the systems thinks it should use as swap space
and compare to what you think it should use.

Good luck

Wolfgang