When SCO OpenServer5 boots you will not see the "disk" advertised in
the "hwconfig -hc" listing.
However, SCO OpenServer5 has support for a second IDE drive built in.
By running:
# divvy -N -P
0 0 29999 boot EAFS
1 30000 815851 swap NON FS
2 815852 19530997 root HTFS
6 19530998 19531007 recover NON FS
7 0 19539008 hd0a WHOLE DISK
we can see the structure of a normal disk.
The second disk can be accessed by checking:
# divvy /dev/hd10
+-------------------+------------+--------+---+-------------+------------+
| Name | Type | New FS | # | First Block | Last Block |
+-------------------+------------+--------+---+-------------+------------+
| | EAFS | no | 0 | 0| 29999|
| | NON FS | no | 1 | 30000| 815851|
| | HTFS | no | 2 | 815852| 19530997|
| | NOT USED | no | 3 | -| -|
| | NOT USED | no | 4 | -| -|
| | NOT USED | no | 5 | -| -|
| | NON FS | no | 6 | 19530998| 19531007|
| hd10 | WHOLE DISK | no | 7 | 0| 19539008|
+-------------------+------------+--------+---+-------------+------------
The filesystems shown may be not be named but can be with the "n"
name option in "divvy". For example, "u1".
+-------------------+------------+--------+---+-------------+------------+
| Name | Type | New FS | # | First Block | Last Block |
+-------------------+------------+--------+---+-------------+------------+
| | EAFS | no | 0 | 0| 29999|
| | NON FS | no | 1 | 30000| 815851|
| u1 | HTFS | no | 2 | 815852| 19530997|
| | NOT USED | no | 3 | -| -|
| | NOT USED | no | 4 | -| -|
| | NOT USED | no | 5 | -| -|
| | NON FS | no | 6 | 19530998| 19531007|
| hd10 | WHOLE DISK | no | 7 | 0| 19539008|
+-------------------+------------+--------+---+-------------+------------
Do not create a new filesystem at this point, ie. don't mark it
as "y" otherwise it will destroy your existing data.
quit at this point but enter "i" to install.
The "i" install will add a /dev/u1 node.
Mount the disk and filesystem with:
# mount /dev/u1 /mnt
If the disk is not mounting then you may need to run "fsck" first.
You should then be able to copy data, for example:
# mkdir /u1
# cd /mnt
# find . -print | cpio -pdumv /u1
Note: There is also the new -l option to divvy, which will show the labels
of any filesystems that are labelled - this can be really useful if
you don't remember what each division was previously used for.
It also adds an "l" interactive option to apply the labels as
division names. swap and root usually don't have filesystem labels,
swap because it isn't a filesystem and root because it is almost
always mounted, and filesystems can only be labelled when they
aren't mounted.
This feature is new to OpenServer 5.0.7.
If you not able to see the 'divvy' table then see this
Troubleshooting Section:
- Check the IDE disk is correctly jumpered.
Although you will not see the IDE disk initially in "hwconfig -hc",
any attempt to access it with the above commands should generate a
console message and in syslog to say it has been detected. eg:
Jul 25 12:47:47 d32 %disk 0x01F0-0x01F7 14 - type=W0 unit=1
cyls=9726 hds=255 secs=63
- Check if there is an active partition on the disk with:
# fdisk -f /dev/rhd1a
- Check with 'divvy':
I am getting:
# divvy /dev/hd10
Cannot stat device "/dev/hd10"
stat: No such file or directory
The device files for /dev/hd10 have been removed. No secondary drive
has been detected.
or I am getting:
divvy /dev/hd10
No valid division table.
Run divvy with either the -i flag if this device is to be initialized
with a new root file system, or the -m flag if it is to be
initialized with a number of mountable file systems.
This means the partition is active but the divvy table has been lost.
- Check the Disk Geometry:
# dparam /dev/rhd10
To see what disk bios geometry has been detected
or:
# dparam -w /dev/rhd10
To re-write the Master Boot Record and reboot.
If this still fails then refer to Technical Article 110737, "Filesystems on my SCSI
disk drive are no longer accessible after moving it to another
OpenServer 5 system, or after making various SCSI configuration
changes."
which also applies to IDE to try to reset the disk's geometry.
- If the disk is not recoverable then you may be able to run:
# divvy -i /dev/rhd10
There are two possibilities, if a root division table layout is
detected or not.
a) IF you run "-i" then, AND if a root division table layout is seen,
then you will see:
This device contains a valid disk division table. Your
additional (non-root) filesystems/divisions can be
preserved across the reinstallation. Do you want to
preserve your additional filesystems/divisions? (y/n) y
The layout of the filesystems and swap area is now prepared.
Do you wish to make any manual adjustments to the sizes or
names of the filesystems or swap area before they are created
on the hard disk? (y/n) y
You will then have the opportunity to prevent any new filesystems
from being created before you proceed.
If you attempt to write a new division table there when there already
is one then you may see:
'Cannot determine type of previously made root device: No such file
or directory.'
Also if you say "no" to preserve your filesystems then you may see:
'new root file system construction unsuccessful'
b) IF you run "-i" AND there is NO divvy division table detected then
it will write a new division table there.
You will be given the option to enter the swap and boot space
default allocations and a separate /u filesystem. You will have
the option to manually amend the filesystems before proceeding
and will have the option to abort with 'intr' at the:
i[nstall] and r[eturn] menu.
Use "divvy -m" or "mkdev hd", to create new filesystems neccesary.
c) IF you wanted to completed remove all partitions and start again
then either perform a low level format or run:
# dd if=/dev/zero of=/dev/rhd10
SEE ALSO:
divvy(ADM)
|