I've been getting this strange report from df all along, but I realized
only in the last hour that it's strange:

  Filesystem  1K-blocks     Used    Avail Capacity  Mounted on
  /dev/sd1a       49199     9644    34635    22%    /
  /dev/sd1f      908105   779661    37633    95%    /usr
  /dev/sd1e      709770   561730    77063    88%    /var
  /dev/sd1g      561147    59570   445462    12%    /home
  kernfs              1        1        0   100%    /kern
  procfs              4        4        0   100%    /proc

The block counts looks about right for the root & /home partitions,
but the other two are VERY WRONG (/usr should be ~204000, /var ~153000).
I rebooted into single-user mode & run fsck -f on each partition (w/o
mounting them), but fsck didn't find anything wrong.

Now I find that /usr is horked but good.  I don't know if there was
damage creeping through all along, but it hosed several programs by
turning off the x bits.  Turning them on was probably a mistake ("bad
system architecture" or something like that -- I'm in MacOS now).

Has anyone else noticed problems with the [ef] partitions on a Jaz
drive?  Am I the only one who has ever partitioned a Jaz like this?

Sys info: Mac IIsi, 17MB RAM, MacOS 7.5.3, NetBSD 1.3, generic SBC
kernel.  All NetBSD partitions are on the Jaz drive.

    Larry Kollar (kollar@stc.net)