DragonFly BSD
DragonFly bugs List (threaded) for 2004-08
[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]

Re: Bootloader problems


From: Chris Pressey <cpressey@xxxxxxxxxxxxxxx>
Date: Mon, 30 Aug 2004 11:57:24 -0700

On Sun, 29 Aug 2004 21:48:47 +0300
all <all@xxxxxxxxxxxx> wrote:

> p.s. when I delete all my partition tables via dd if=/dev/zero
> of=/dev/ad0 bs=32k count=16, then fdisk, 3 partitions are unused, 4th,
> last one, is entire disk.

[and]

On Mon, 30 Aug 2004 20:02:00 +0200
"M. Schatzl" <lists@xxxxxxxxxxxxxxx> wrote:

> While checking the installation again with the live-CD, fdisk -i 
> showed me all partitons except the last one empty and without any 
> label. Slice 4 had a DragonFly disklabel and the size I gave ad0s2.

These are both occurances of the odd behaviour of fdisk.

When fdisk looks at a disk that has an invalid partition table, it says
the first 3 partitions are unused and that the 4th is a valid DragonFly
partition.  This is purely a delusion.

However...

On Mon, 30 Aug 2004 20:02:00 +0200
"M. Schatzl" <lists@xxxxxxxxxxxxxxx> wrote:

> After rebooting the machine, the only thing i saw was an BTX loader 
> error-message. The loader knew about the two systems (DOS/BSD) on 
> disk.

This implies that the partition table was not in fact invalid at this
point, as the BTX loader could decipher it.

> While checking the installation again with the live-CD, fdisk -i 
                                                          ^^^^^^^^
> showed me all partitons except the last one empty and without any 
> label.

  -i   Initialize sector 0 of the disk.  This implies -u, unless -f is
       given.

So I hope that's what you meant to do :)

Note that I've never had use for this option (it seems to be one cause
of the delusion I described above); fdisk -I is IMO the better choice. 
(Going from the man page alone I would have a hard time describing what
the intended difference between -i and -I is supposed to be...)

Sorry if this isn't much help - I don't know much about the BTX loader
itself (which is what seems to actually be failing in these instances.)

-Chris



[Date Prev][Date Next]  [Thread Prev][Thread Next]  [Date Index][Thread Index]