I haven't had any responses about my "weird filesystem damage" problem,
so I booted single-user & brought back the following info:

  # disklabel sd1
  # /dev/rsd1c:
  type: SCSI
  disk: SCSI disk
  label: fictitious
  flags:
  bytes/sector: 512
  sectors/track: 32
  tracks/cylinder: 64
  sectors/cylinder: 2048
  cylinders: 1021
  total sectors: 2091050
  rpm: 3600
  interleave: 1
  trackskew: 0
  cylinderskew: 0
  headswitch: 0           # milliseconds
  track-to-track seek: 0  # milliseconds
  drivedata: 0

  7 partitions:
  #        size   offset    fstype   [fsize bsize   cpg]
    a:   102400   143504      4.2BSD        0     0     0   # (Cyl.   70*- 120)
    b:    71680   245904        swap                        # (Cyl.  120*- 155)
    c:  2091050        0      unused        0     0         # (Cyl.    0 -
1021*)
    d:   143360      144         HFS                        # (Cyl.    0*- 70)
    e:   307200   317584      4.2BSD        0     0     0   # (Cyl.  155*- 305)
    f:   409600   624784      4.2BSD        0     0     0   # (Cyl.  305*- 505)
    g:  1056666  1034384      4.2BSD        0     0     0   # (Cyl.  505*-
1021*)
  #
  #
  # df
  Filesystem  1K-blocks     Used    Avail Capacity  Mounted on
  /dev/sd1a       49199    10451    33828    24%    /
  /dev/sd1f      908105   779723    37571    95%    /usr
  /dev/sd1e      709770   561735    77058    88%    /var
  /dev/sd1g      561147    59571   445461    12%    /home

The info in disklabel's "size" column matches perfectly the data I wrote
down when I set up the partitions.  Now compare the entries for /dev/sd1e
and /dev/sd1f (accounting for differences in block sizes and reserving
spare blocks).  What df shows is horribly wrong for sd1e & sd1f; I think
sd1a & sd1g are OK.


What *really* worries me is that fsck doesn't show any filesystem damage.
When I do fsck -f, the sizes it displays agrees with disklabel and it still
doesn't find any problems.  The kernel is from ...binary/sets/kern_sbc.tgz
if that makes a difference.

Now I suppose I could run newfs from BSD, or mkfs from MacOS, but then I
wouldn't know what caused the problem in the first place (thus constantly
fearing that It Could Happen Again At Any Time).  So...

Where the heck is df getting that wrong information, anyway?

If/when I do newfs on sd1e & sd1f, what should I look for or specify?

Is there any way to repair the filesystems w/o using newfs?  (I'll have to
re-install all the tarballs anyway, I've lost too many files.)

Thanks much in advance!

        Larry Kollar (kollar@stc.net)