Reformatting external harddrive

[Jerry McAllister]

On Tue, May 12, 2009 at 01:41:37PM -0400, Daniel Underwood wrote:

> Thanks for all the advice. This evening, when I get home to work, I
> will try these suggestions.
> 
> I have no idea why there is more than 1 partition on this disk. I must
> have inadvertently created multiple partitions when I was struggling
> to reformat this disk in linux.  Every time I tried to fdisk (or
> perhaps it was mkfs.ext3) on in linux, I got errors about a "bad
> superblock" (which I understand somehow relates to the journaling
> mechanism of ext3).

According to the fdisk output you include, there is only 1 slice (called
primary partition in MS land) being used on the disk.   
It has about 305242 MegaBytes which seems to be what you are looking for.

It is possible to have up to 4 slices (primary partitions) on a disk.
The fdisk output shows one being used and the other three empty and
not being used.    So, you are all fine.

Ignore those bogus messages about BIOS and partitions not in cyl 1, etc.
They are not relevant.


You can now either use bsdlabel to create partitions within that slice
or just use the slice as is.   In either case, you have to newfs the
unit to create a filesystem in it so you can mount it and write/read it.

If you use it as is without creating partitions, then do:
  newfs /dev/da0s1

If you create one single partition within that slice - say a: for example
then the newfs would be:
  newfs /dev/da0s1a

If you create partitions, lets say a:, d: and e: then do:

  newfs /dev/da0s1a
  newfs /dev/da0s1d
  newfs /dev/da0s1e

In any case, to not use  partition  c:   and for sanity's sake
avoid  b: .    c:  is reserved to describe the whole slice and, by 
convention, b:  is used for swap though that is not required.
Some people suggest avoiding  a:  because it is most often a 
bootable partition.   It is not required either, but it can reduce
confusion to avoid it in non-bootable slices/disks.

If your backups are small enough so more than one full dump can
fit on the disk, then I suggest slicing and partitioning so that
each full dump series can have its own partition.  It is not a
requirement.  It is just easier to keep track of on those groggy
odd hours.

////jerry

> >>
> >> ******* Working on device /dev/da0 *******
> >> parameters extracted from in-core disklabel are:
> >> cylinders=38913 heads=255 sectors/track=63 (16065 blks/cyl)
> >>
> >> Figures below won't work with BIOS for partitions not in cyl 1
> >> parameters to be used for BIOS calculations are:
> >> cylinders=38913 heads=255 sectors/track=63 (16065 blks/cyl)
> >>
> >> Media sector size is 512
> >> Warning: BIOS sector numbering starts with sector 1
> >> Information from DOS bootblock is:
> >> The data for partition 1 is:
> >> sysid 165 (0xa5),(FreeBSD/NetBSD/386BSD)
> >>    start 63, size 625137282 (305242 Meg), flag 80 (active)
> >>        beg: cyl 0/ head 1/ sector 1;
> >>        end: cyl 0/ head 254/ sector 63
> >> The data for partition 2 is:
> >> <UNUSED>
> >> The data for partition 3 is:
> >> <UNUSED>
> >> The data for partition 4 is:
> >> <UNUSED>
> >