help with damaged partition

davexnet

unread,

Apr 4, 2003, 3:32:40 PM4/4/03

to

Hello,
I have damaged the slave hardrive by attempting to install
Linux and then using partition magic. It's a 20GB harddrive, and it
had a 12 GB data partition (fat32), then 8 gb extended. Within the 8gb
extended there was 2 GB of free space, followed by a fat32 data
partition (6gb)

I started the linux (slackware 9) setup, and created a
swapfile and linux partition within the 2gb of free space.
Did *not* install LILO. After it was done, I used Partition Magic 7
(from floppy) to take a look. Before the main PM screen appeared,
some errors on disk 2 were reported. 2 or 3 instances of CHS not
matching, and did I want to fix (replied yes), then it reported an
error in the EPBR, did I want to fix - replied yes. PM then hung.
PM will no longer initialize with this drive in the PC.

I downloaded "findpart" from Svend Olaf Mikkelsen's
website http://inet.uni2.dk/~svolaf/utilities.htm
and after making this the only drive in the PC, I
ran the program - here's the output:

Findpart, version 4.32.
Copyright Svend Olaf Mikkelsen, 2003.

Searches for partitions type 01, 04, 06, 07, 0B, 0C, 0E, 82, 83,
plus Fdisk F6 and Lilo sectors. Information based on bootsectors
is marked B. If the disk is larger than supported by BIOS, the
supported part of the disk is examined. Disks are numbered from 1.

OS: DOS 6.00

Disk: 1 Cylinders: 2489 Heads: 255 Sectors: 63 MB: 19524

-PCyl N ID -----Rel -----Num ---MB -Start CHS- --End CHS-- BS CHS
0 - 0C 63 23615487 11530 0 1 1 1469 254 63 B OK
1470 1 82 63 530082 258 1470* 1 1 1502*254 63 OK
1470 2 05 0 530145 258 1023 0 1 1055*254 63 +0 NB
0 - 0B 23615613 12289662 6000 1470 1 1 2234 254 63 B OK
Fdisk F6 sector 1476 1 1
1503 1 83 63 3550302 1733 1503# 1 1 1723*254 63 OK R OK
1724 1 0B 63 12289662 6000 1724* 1 1 2488*254 63 OK OK
0 - 0B 27696123 12289662 6000 1724 1 1 2488 254 63 BU OK
Fdisk F6 sector 2280 0 1
Fdisk F6 sector 2280 1 1
Fdisk F6 sector 2324 0 1
Fdisk F6 sector 2324 1 1

-----FAT CHS -Size Cl --Root -Good -Rep. Maybe --Bad YYMMDD DataMB
0 1 33 11527 8 --- 11527 0 0 0 980101 8636
1470 1 33 11981 4 996 11981 0 0 0 980101 4418
1724 1 33 11981 4 996 11981 0 0 0 980101 4418

Partitions according to partition tables on first harddisk:

-PCyl N ID -----Rel -----Num ---MB -Start CHS- --End CHS-- BS CHS
0 1 0C 63 23615487 11530 0 1 1 1469*254 63 OK OK
0 2 0F 23615550 16370235 7993 1470* 0 1 2488*254 63 OK

1470 1 82 63 530082 258 1470* 1 1 1502*254 63 OK
1470 2 05 0 530145 258 1470* 0 1 1502*254 63 OK

1470 1 82 63 530082 258 1470* 1 1 1502*254 63 OK
1470 2 05 0 530145 258 1470* 0 1 1502*254 63 OK
Only 64 partitions can be listed.

Not sure what to make of it, but I'm guessing that the
extended partion is damaged.
Appreciate any help in unraveling this messs.

Dave

Svend Olaf Mikkelsen

unread,

Apr 4, 2003, 5:43:25 PM4/4/03

to

On Fri, 04 Apr 2003 20:32:40 GMT, davexnet
<davexnet...@yahoo.com> wrote:

>Hello,
>I have damaged the slave hardrive by attempting to install
>Linux and then using partition magic. It's a 20GB harddrive, and it
>had a 12 GB data partition (fat32), then 8 gb extended. Within the 8gb
>extended there was 2 GB of free space, followed by a fat32 data
>partition (6gb)
>
>I started the linux (slackware 9) setup, and created a
>swapfile and linux partition within the 2gb of free space.
>Did *not* install LILO. After it was done, I used Partition Magic 7
>(from floppy) to take a look. Before the main PM screen appeared,
>some errors on disk 2 were reported. 2 or 3 instances of CHS not
>matching, and did I want to fix (replied yes), then it reported an
>error in the EPBR, did I want to fix - replied yes. PM then hung.
>PM will no longer initialize with this drive in the PC.
>
>I downloaded "findpart" from Svend Olaf Mikkelsen's
>website http://inet.uni2.dk/~svolaf/utilities.htm
>and after making this the only drive in the PC, I
>ran the program - here's the output:

>Not sure what to make of it, but I'm guessing that the

>extended partion is damaged.
>Appreciate any help in unraveling this messs.
>
>Dave

The partition tables are cyclic. The explanation that you can boot DOS
6.00 is that it that it does not know extended type 0F, and/or that it
is not MS DOS.

It is assumed that nothing was changed since Findpart was run. And it
is assumed that the disk is still disk number 1.

You can download davex1.bat in

http://inet.uni2.dk/~svolaf/davex1.zip

and run davex1.bat, which contains:

set findpart=edit
findpart 1 1470 2 - 05 1503 0 1 1723 254 63 1470 2489 255 63 26
findpart 1 1503 2 - 05 1724 0 1 2488 254 63 1470 2489 255 63 26
set findpart=
findpart 1 table fp1-1.txt

--
Svend Olaf

davexnet

unread,

Apr 4, 2003, 11:13:20 PM4/4/03

to

On Fri, 04 Apr 2003 22:43:25 GMT, svo...@inet.uni2.dk (Svend Olaf
Mikkelsen) wrote:

>>Dave
>
>The partition tables are cyclic. The explanation that you can boot DOS
>6.00 is that it that it does not know extended type 0F, and/or that it
>is not MS DOS.
>
>It is assumed that nothing was changed since Findpart was run. And it
>is assumed that the disk is still disk number 1.
>
>You can download davex1.bat in
>
>http://inet.uni2.dk/~svolaf/davex1.zip
>
>and run davex1.bat, which contains:
>
>set findpart=edit
>findpart 1 1470 2 - 05 1503 0 1 1723 254 63 1470 2489 255 63 26
>findpart 1 1503 2 - 05 1724 0 1 2488 254 63 1470 2489 255 63 26
>set findpart=
>findpart 1 table fp1-1.txt

Svend - thank you very much for your help.
Partition Magic is now initializing fine on this disk.

You are right about DOS. All of the MSDOS bootdisks I had
(Partition magic, windows 98 startup, windows 2000 cdrom boot)
were hanging. I was lucky to have one that used Caldera DR-DOS.
Ironically, turned out to be MAXDIAG - Maxtors diagnostic program.
That's how I was able to run your program.
The partition table looks like this now:

OS: DOS 6.00

Disk: 1 Cylinders: 2489 Heads: 255 Sectors: 63 MB: 19524

Partitions according to partition tables on first harddisk:

-PCyl N ID -----Rel -----Num ---MB -Start CHS- --End CHS-- BS CHS
0 1 0C 63 23615487 11530 0 1 1 1469*254 63 OK OK
0 2 0F 23615550 16370235 7993 1470* 0 1 2488*254 63 OK

1470 1 82 63 530082 258 1470* 1 1 1502*254 63 OK

1470 2 05 530145 3550365 1733 1503 0 1 1723 254 63 OK

1503 1 83 63 3550302 1733 1503# 1 1 1723*254 63 OK R OK

1503 2 05 4080510 12289725 6000 1724 0 1 2488 254 63 OK

1724 1 0B 63 12289662 6000 1724* 1 1 2488*254 63 OK OK

Can you tell me what may have happened? What is cyclic?

Dave
PS It looks like I didn't allocate enough for the linux partition
According to PM7,it's size is 1.733 GB, and 100% of it is used.
Yet the install looked as if it finished normally.
Part of the problem?

Svend Olaf Mikkelsen

unread,

Apr 5, 2003, 7:55:38 AM4/5/03

to

On Sat, 05 Apr 2003 04:13:20 GMT, davexnet
<davexnet...@yahoo.com> wrote:

>Svend - thank you very much for your help.
>Partition Magic is now initializing fine on this disk.
>
>You are right about DOS. All of the MSDOS bootdisks I had
>(Partition magic, windows 98 startup, windows 2000 cdrom boot)
>were hanging. I was lucky to have one that used Caldera DR-DOS.
>Ironically, turned out to be MAXDIAG - Maxtors diagnostic program.
>That's how I was able to run your program.
>The partition table looks like this now:

>Can you tell me what may have happened? What is cyclic?

>
>Dave
>PS It looks like I didn't allocate enough for the linux partition
>According to PM7,it's size is 1.733 GB, and 100% of it is used.
>Yet the install looked as if it finished normally.
>Part of the problem?

The second entry in an extended partition table points to next
extended partition table. In this case the partition table at cylinder
1470 pointed to itself.

The problem may have been that a Linux tool did not make standard
partition tables, and that Partition Magic was not able to handle that
correctly.

The Linux partition uses the Reiser file system. Partition Magic may
not know that file system.
--
Svend Olaf

davexnet

unread,

Apr 5, 2003, 8:19:13 PM4/5/03

to

On Sat, 05 Apr 2003 12:55:38 GMT, svo...@inet.uni2.dk (Svend Olaf
Mikkelsen) wrote:

>The second entry in an extended partition table points to next
>extended partition table. In this case the partition table at cylinder
>1470 pointed to itself.
>
>The problem may have been that a Linux tool did not make standard
>partition tables, and that Partition Magic was not able to handle that
>correctly.
>
>The Linux partition uses the Reiser file system. Partition Magic may
>not know that file system.

Thanks svend - very interesting.
Partition Magic correctly identifies the linux swap file (type 82)
and the linux partition (type 83).
Those partitions were created with the Linux Fdisk during the setup.
The linux partition turned out to be too small.
I could easily allocate more space and retrace my steps....
but I'd be afraid it would happen again.
Need to think about this....
Dave.