Greetings, all.

I've been following the discussion about adding a disk, and have recently
added a second NetBSD disk in my Quadra 800 (hitherto 40/500, GENERIC
kernel).  Unfortunately, I think something has gone awry, and I'd like
some help.  Essentially, my brand-new partitions are mountable, but don't
appear to be empty, appearing progressively more full in order of their
creation/indexing.  A full (i.e., very long) description follows.


I formatted and partitioned the disk (SCSI ID 1) into six, using Apple's
Drive Setup 1.7.3.  I started with HFS partitions, which seemed fine under
MacOS, then dismounted them and ran the Mkfs utility to convert the
partitions to NetBSD as follows:

 root
 swap
 usr
 usr
 usr
 usr

I didn't run the installer, since I already boot from sd0, and I'll use
dump/restore to make a backup of my root partion from sd0 on sd1.  Booting
up in NetBSD (from sd0) shows that all my disks are recognized: 

==>>

sd0 at scsibus0 targ 0 lun 0: <SEAGATE, ST3600N, 8674> SCSI1 0/direct
fixed
sd0: 500MB, 1872 cyl, 7 head, 78 sec, 512 bytes/sect x 1025920 sectors
sd1 at scsibus0 targ 1 lun 0: <IBM, CP30540 545MB !Q, ADB7> SCSI2 0/direct
fixed
sd1: 520MB, 2242 cyl, 6 head, 79 sec, 512 bytes/sect x 1065912 sectors
sd2 at scsibus0 targ 2 lun 0: <QUANTUM, LP80S  980809404, 2.9> SCSI2 
0/direct fixed
sd2: 80MB, 921 cyl, 4 head, 44 sec, 512 bytes/sect x 164139 sectors
cd0 at scsibus0 targ 3 lun 0: <MATSHITA, CD-ROM CR-8004A, 2.0a> SCSI2
5/cdrom removable


disklabel reports that my partitions are present:

==>>

# disklabel sd1 
# /dev/rsd1c:
type: SCSI
disk: CP30540 545MB !
label: fictitious
flags:
bytes/sector: 512
sectors/track: 79
tracks/cylinder: 6
sectors/cylinder: 474
cylinders: 2242
total sectors: 1065912
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   cpg]
  a:    82655      704     4.2BSD        0     0     0   # (Cyl.    1*-
175*)
  b:    82655    83359       swap                        # (Cyl.  175*-
350*)
  c:  1065912        0     unused        0     0         # (Cyl.    0 -
2248*)
  d:      512      192    unknown                        # (Cyl.    0*-
1*)
  e:   166406   326789     4.2BSD        0     0     0   # (Cyl.  689*-
1040*)
  f:   411881   493195     4.2BSD        0     0     0   # (Cyl. 1040*-
1909*)
  g:   160775   166014     4.2BSD        0     0     0   # (Cyl.  350*-
689*)
  h:   160826   905076     4.2BSD        0     0     0   # (Cyl. 1909*-
2248*)
disklabel: boot block size 0
disklabel: super block size 0


I ran fsck on each partition (redundant lines omitted):

==>>

** /dev/rsd1a
2 files, 9 used, 39861 free (21 frags, 4980 blocks, 0.1% fragmentation)

** /dev/rsd1b
2 files, 9 used, 39861 free (21 frags, 4980 blocks, 0.0% fragmentation)

** /dev/rsd1e
2 files, 9 used, 80297 free (17 frags, 10035 blocks, 0.0% fragmentation)

** /dev/rsd1f
2 files, 9 used, 199002 free (18 frags, 24873 blocks, 0.0% fragmentation)

** /dev/rsd1g
2 files, 9 used, 77625 free (17 frags, 9701 blocks, 0.0% fragmentation)

** /dev/rsd1h
2 files, 9 used, 77651 free (19 frags, 9704 blocks, 0.0% fragmentation)

I answered "n" to each question about whether or not to mark that
filesystem clean.


I mount all the partitions on some test directories to check things out:

==>>

# mount -vr /dev/sd1a /bkproot
exec: mount_ffs -o ro /dev/sd1a /bkproot
/dev/sd1a on /bkproot type ffs (local, read-only)
# mount -vr /dev/sd1f /pkg
exec: mount_ffs -o ro /dev/sd1f /pkg
/dev/sd1f on /pkg type ffs (local, read-only)
# mount -vr /dev/sd1e /tmp2
exec: mount_ffs -o ro /dev/sd1e /tmp2
/dev/sd1e on /tmp2 type ffs (local, read-only)
# mount -vr /dev/sd1h /home2
exec: mount_ffs -o ro /dev/sd1h /home2
/dev/sd1h on /home2 type ffs (local, read-only)
# mount -vr /dev/sd1g /work
exec: mount_ffs -o ro /dev/sd1g /work
/dev/sd1g on /work type ffs (local, read-only)


So far so good.  The problem is this - each of the new filesystems should
be empty, but df reports that almost all of them contain data, in a
suspiciously incremented manner:

==>

# df -k /bkproot /pkg /tmp2 /home2 /work

Filesystem  1K-blocks     Used    Avail Capacity  Mounted on
/dev/sd1a       39870        9    35874     0%    /bkproot
/dev/sd1f      436691   237689   155332    60%    /pkg
/dev/sd1e      237680   157383    56529    73%    /tmp2
/dev/sd1h      514351   436700    26215    94%    /home2
/dev/sd1g      157374    79749    61887    56%    /work


I find the behaviour a little odd, but I won't rule out that I might be
doing something wrong.  Do I need some incantation with disklabel or
newfs?  Should I have done this in an entirely different manner?  Is this
a bug?  Is there some weirdness with this model drive?


/*======================================================================
"Don't die wondering..."                http://www.cldc.howard.edu/~cpg
                                              email: cpg@scs.howard.edu
chris out-              Christopher P. Gill
  peace.        C.L.D.C. Senior System Operator (Ret.)
======================================================================*/