Subject: HDD-woes on Sparcstation 5
To: None <port-sparc@netbsd.org>
From: Victor Lagerkvist <plumsaren@gmail.com>
List: port-sparc
Date: 11/17/2006 19:26:44
Hello,
I've encountered a weird behavior with NetBSD 3.1 on my Sparcstation 5. First
of all, the hard drive is divided into three partitions: one for /, one for
swap and one for miscellaneous data. The disklabel that the slice uses looks
like this:
bock# disklabel -e /dev/sd0
# /dev/sd0c:
type: SCSI
disk: mydisk
label: SUN0535 cyl 1866
flags:
bytes/sector: 512
sectors/track: 80
tracks/cylinder: 7
sectors/cylinder: 560
cylinders: 1866
total sectors: 1044960
rpm: 5400
interleave: 1
trackskew: 0
cylinderskew: 0
headswitch: 0 # microseconds
track-to-track seek: 0 # microseconds
drivedata: 0
8 partitions:
# size offset fstype [fsize bsize cpg/sgs]
a: 500000 0 4.2BSD 1024 8192 41672
# (Cyl. 0 - 892*)
b: 392000 522480 4.2BSD 1024 8192 39200
# (Cyl. 933 - 1632)
c: 112000 914480 swap
# (Cyl. 1633 - 1832)
d: 1044960 0 unused 0 0
# (Cyl. 0 - 1865)
If it looks a little funny, it's because I manually created the disklabel
since the normal installer that I tried first failed, and thus I just ran the
INSTALL-script which worked. Now the problem; just to see whether the second
partition worked as expected I tried the following:
bock# df -ha
Filesystem Size Used Avail Capacity Mounted on
/dev/sd0a 237M 175M 50M 77% /
/dev/sd0b 186M 1.0K 176M 0% /mnt
bock# du -m /usr/share/man #Just to see how much diskspace it occupied on sd0a
18 /usr/share/man
bock# cp -r /usr/share/man /mnt/
The first weird thing is that this simple procedure takes almost 10 minutes to
complete, the second is the diskspace it occupies on /dev/sd0b:
bock# df -ha
Filesystem Size Used Avail Capacity Mounted on
/dev/sd0a 237M 175M 50M 77% /
/dev/sd0b 186M 87M 90M 49% /mnt
bock# du -m /mnt/man
87 /mnt/man
What!? How's that possible? The "man" folder is the only folder on /mnt, and I
don't understand how it managed to grow over 4 times the original size. It
might also be worth to mention that I used "newfs" to create the filesystem
on /dev/sd0b. Might it be a sign that the disk is going bananas or is it just
a misconfiguration on my part?