9.1 Gbyte SCSI Drive

9.1 Gbyte SCSI Drive

[Rick Cook]

Hello,

I have a Power Macintosh 7300/200 that currently has two 2 Gbyte SCSI Hard
Drives run Debian 2.2 with partitions mounted from both drives. Everything is
working.

I have a Seagate 9.1 Gbyte SCSI ST39173LC that I would like to install as a
replacement for one of the 2 Gbyte drives. The dreaded "the 9.1 Gbyte drive
works fine in a Pentium 120 system". However, when I install it into the
7300, the system crashes when I attempt to mke2fs any of the partitions (I
have tried several partition size combinations). I don't remember if I have
tried this since I initialized the partition map on the Pentium Machine.

Is this a known problem in the powerpc kernel (I seem to remember a
relatively recent CVS log post for the Macintosh version of m68k about a
change related to this)?

Any suggestions about where to start looking?

Thanks,

Rick Cook
--
rcook@ntlug.org


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Michel Lanners]

On  26 Apr, this message from Rick Cook echoed through cyberspace:
> I have a Seagate 9.1 Gbyte SCSI ST39173LC that I would like to install as a
> replacement for one of the 2 Gbyte drives. The dreaded "the 9.1 Gbyte drive
> works fine in a Pentium 120 system". However, when I install it into the
> 7300, the system crashes when I attempt to mke2fs any of the partitions (I
> have tried several partition size combinations). I don't remember if I have
> tried this since I initialized the partition map on the Pentium Machine.

Did you repartition the drive on the PowerMac before making filesystems?

I vaguely remember an issue with, IIRC, drives that had PC-style
partition maps. Not sure about the details anymore, sorry.

Anyway, the point was make sure you erase the PC partition map
_completely_, with something like this:

dd if=/dev/null of=/dev/sd<drive> bs=512 count=10

This should scratch the first few blocks on disk, some of which are not
touched by Mac-style partition maps, and make it possible to do a
_clean_ Mac partition map.

Obviously, the above commands destroys anything that was in those blocks
before, effectively erasing all partition information. So use with
caution, make sure you target the right disk, and only if you don't
weant to keep any data on the target drive.

Cheers

Michel

-------------------------------------------------------------------------
Michel Lanners                 |  " Read Philosophy.  Study Art.
23, Rue Paul Henkes            |    Ask Questions.  Make Mistakes.
L-1710 Luxembourg              |
email   mlan@cpu.lu            |
http://www.cpu.lu/~mlan        |                     Learn Always. "


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Hollis R Blanchard]

On Thu, 26 Apr 2001, Michel Lanners wrote:
>
> Anyway, the point was make sure you erase the PC partition map
> _completely_, with something like this:
>
> dd if=/dev/null of=/dev/sd<drive> bs=512 count=10
>
> This should scratch the first few blocks on disk, some of which are not
> touched by Mac-style partition maps, and make it possible to do a
> _clean_ Mac partition map.

As long as you don't try to make it with pdisk, which claims all disks
without an existing valid Apple partition map are 512.0 GB.

parted (or Drive Setup) should work fine.

-Hollis


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Ethan Benson]

[-- Attachment #1: Type: text/plain, Size: 559 bytes --]

On Thu, Apr 26, 2001 at 03:00:44PM -0400, Hollis R Blanchard wrote:
> 
> As long as you don't try to make it with pdisk, which claims all disks
> without an existing valid Apple partition map are 512.0 GB.

there is a workaround for that, just tell it how many blocks the disk
has, this is discussed in my mac-fdisk-basics doc at
http://www.alaska.net/~erbenson/doc and
http://penguinppc.org/usr/ybin/doc/

> parted (or Drive Setup) should work fine.

parted is fine. Drive setup is evil.  

-- 
Ethan Benson
http://www.alaska.net/~erbenson/

[-- Attachment #2: Type: application/pgp-signature, Size: 0 bytes --]

Re: 9.1 Gbyte SCSI Drive

[Giuliano Pochini]

Hollis R Blanchard wrote:
>
> On Thu, 26 Apr 2001, Michel Lanners wrote:
> >
> > Anyway, the point was make sure you erase the PC partition map
> > _completely_, with something like this:
> >
> > dd if=/dev/null of=/dev/sd<drive> bs=512 count=10
> >
> > This should scratch the first few blocks on disk, some of which are not
> > touched by Mac-style partition maps, and make it possible to do a
> > _clean_ Mac partition map.
>
> As long as you don't try to make it with pdisk, which claims all disks
> without an existing valid Apple partition map are 512.0 GB.


Uh?

version 0.7a3 (18 February 1998)

[root@Jay linux]# /sbin/pdisk -i
Top level command (? for help): e /dev/sdb
Edit /dev/sdb -
Command (? for help): i
map already exists
do you want to reinit? [n/y]: y
A physical block is 512 bytes: 2048
A logical block is 512 bytes: 2048
size of 'device' is 555555 blocks (2048 byte blocks): 123456
new size of 'device' is 123456 blocks (2048 byte blocks)
Command (? for help):


Bye.


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Giuliano Pochini]

>> version 0.7a3 (18 February 1998)
>>
>> [root@Jay linux]# /sbin/pdisk -i
>> Top level command (? for help): e /dev/sdb
>> Edit /dev/sdb -
>> Command (? for help): i
>> map already exists
>   ^^^ seems to suggest that you did not use dd to zero your partition map
> before trying this experiment.

Yes, I used a previously partitioned disk, but I use the same method
with clean discs with no problems. Ok, I'll try with a dd'ed one and
I'll make you know.


Bye.
    Giuliano Pochini ->)|(<- Shiny Network {AS6665} ->)|(<-


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Giuliano Pochini]

> > [root@Jay linux]# /sbin/pdisk -i
> > Top level command (? for help): e /dev/sdb
> > Edit /dev/sdb -
> > Command (? for help): i
> > map already exists
>   ^^^ seems to suggest that you did not use dd to zero your partition map
> before trying this experiment.


[root@Jay Giu]# dd if=/dev/zero of=/dev/sdb bs=1048576 count=100
100+0 records in
100+0 records out

[root@Jay Giu]# /sbin/pdisk -i
Top level command (? for help): e
Name of device: /dev/sdb
pdisk: No valid block 1 on '/dev/sdb'
Edit /dev/sdb -
Command (? for help): I
A physical block is 1 bytes: 2048
A logical block is 512 bytes: 2048
size of 'device' is 310352 blocks (2048 byte blocks):
new size of 'device' is 310352 blocks (2048 byte blocks)
Command (? for help): s 3
Command (? for help): p

Partition map (with 2048 byte blocks) on '/dev/sdb'
 #:                type name   length   base   ( size )
 1: Apple_partition_map Apple       3 @ 1
 2:          Apple_Free Extra  310348 @ 4      (606.1M)

Device block size=2048, Number of Blocks=310352 (606.2M)
DeviceType=0x0, DeviceId=0x0

Command (? for help): r 2 1
Command (? for help): p

Partition map (with 2048 byte blocks) on '/dev/sdb'
 #:                type name   length   base   ( size )
 1:          Apple_Free Extra  310348 @ 4      (606.1M)
 2: Apple_partition_map Apple       3 @ 1

Device block size=2048, Number of Blocks=310352 (606.2M)
DeviceType=0x0, DeviceId=0x0

Command (? for help): w
Writing the map destroys what was there before. Is that okay? [n/y]: y
The partition table has been altered!

Command (? for help): q
Top level command (? for help): q
[root@Jay Giu]# /sbin/mke2fs -b 2048 -m 0 /dev/sdb1
mke2fs 1.19, 13-Jul-2000 for EXT2 FS 0.5b, 95/08/09
Filesystem label=
OS type: Linux
Block size=2048 (log=1)
Fragment size=2048 (log=1)
77824 inodes, 310348 blocks
0 blocks (0.00%) reserved for the super user
First data block=0
19 block groups
16384 blocks per group, 16384 fragments per group
4096 inodes per group
Superblock backups stored on blocks:
	16384, 49152, 81920, 114688, 147456

Writing inode tables: done
Writing superblocks and filesystem accounting information: done
[root@Jay Giu]#


It works fine. Are you sure your HD is ok ?

Bye.

** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/

Re: 9.1 Gbyte SCSI Drive

[Rick Cook]

Hello all,

Thanks for all of the suggestions. I'm not sure which one actually fixed
things, however, my 9.1 Gbyte drive is now happily running inside my 7300/200.

A couple of things different about this successful attempt to get it working.
It is SCSI ID 2, it is not the drive providing bus termination, it is running
in an already installed and running Debian environment, and the system is
running a 2.2.19 kernel.


Thanks again,

Rick
--
rcook@ntlug.org


** Sent via the linuxppc-dev mail list. See http://lists.linuxppc.org/