I have a SCSI disk drive with an MSDOS (FAT) file system on it, and
I'm trying to read it under NetBSD 1.1 on an i486 with an Adaptec 1542
host adapter.

The drive in question is my second SCSI drive (sd1); I boot NetBSD
from sd0.

I'm trying to mount sd1 with the following command:
	mount_msdos /dev/sd1a /mnt

When I try the command for the first time after
booting, the kernel prints the message:
	sd1: no disk label

and then I get two copies of the message:
	mount_msdos: mount: device not configured

Subsequent attempts to mount the drive result in no kernel message,
and only a single copy of the non-kernel message (i.e., mount_msdos:
mount: device not configured).

So my question is: What could "device not configured" possibly mean? I
see the same message if I try to disklabel the drive, so I assume it's
some "standard" error message (from perror?). But I can't find
anything that looks like it in 'man errno'.

One reason I find the message confusing is that the device *is*
configured in my kernel (at least it appears in my config file).
The following lines are from my kernel config file:

aha0	at isa? port 0x330 irq ? drq ?	# Adaptec 154[02] SCSI controllers
scsibus* at aha?
sd*	at scsibus? target ? lun ?	# SCSI disk drives

The device gets probed and recognized correctly by the kernel at boot
time. Here are some messages the kernel prints while booting:

aha0: 1542C/CF detected, unlocking mailbox
aha0 at isa0 port 0x330-0x333 irq 11 drq 5
scsibus0 at aha0
aha0 targ 0 lun 0: <MICROP, 1528-15MD1066702, DD24> SCSI1 0/direct fixed
sd0 at scsibus0: 1280MB, 2100 cyl, 15 head, 83 sec, 512 bytes/sec
aha0 targ 6 lun 0: <CDC, 94191-15, 0136> SCSI1 0/direct fixed
sd1 at scsibus0: 632MB, 1632 cyl, 15 head, 52 sec, 512 bytes/sec

All of those messages look good to me.

Here is an excerpt from my /dev directory:
[101] robert@grover> ls -l /dev/sd1*
brw-r-----  1 root  operator    4,   8 Feb 22 15:30 /dev/sd1a
brw-r-----  1 root  operator    4,   9 Feb 22 15:30 /dev/sd1b
brw-r-----  1 root  operator    4,  10 Feb 22 15:30 /dev/sd1c
brw-r-----  1 root  operator    4,  11 Feb 22 15:30 /dev/sd1d
brw-r-----  1 root  operator    4,  12 Feb 22 15:30 /dev/sd1e
brw-r-----  1 root  operator    4,  13 Feb 22 15:30 /dev/sd1f
brw-r-----  1 root  operator    4,  14 Feb 22 15:30 /dev/sd1g
brw-r-----  1 root  operator    4,  15 Feb 22 15:30 /dev/sd1h

Can anyone tell me what I'm doing wrong? Why can't I mount_msdos this
disk? I have successfully mount_msdos'ed many floppies before --
what's the catch with SCSI disks? Bear in mind while answering that my
sd0 (partitioned in the traditional Un*x way, and containing two ffs
filesystems) works just fine. Thanks for any info!

	-- Robert Kennedy