SCO UNIX Won't Boot

Only in that adding all three up and multiplying by 1024 gives 4.1gb, adding 200mb for swap would suggest a 4.3gb disk drive. If there are two drives they are probably mirrored.

When the system boots, before the "boot:" prompt, you should see the following:
1 the amount of memory available.
2 a "<ctrl>A" prompt for the Adaptec host adapter.
3 a line For each SCSI device attached.
4 a "<ctrl>M prompt for the Mylex host adapter.
5 a list of System Disks and their state.
State will be either online, degraded, dead or offline
If there is only one system disk then u1 was always on the same drive as root.
If there is more than one system disk, and it is not online or degraded, then you have to determine why it is dead or offline, and resolve this.
You can rebuild a degraded system disk, or make an offline system disk online, by typing <ctrl>M and running the menu driven RAID configuration software.
If the system disk is dead, it means that enough drives of the cluster have failed to make recovery of the data impossible (eg both drives of RAID1 or more than 1 drive of a RAID5 configuration.)
Offline means that the disk is unavailable for use, marking it online is an option of the configuration software.
You should also have about 5 diskettes and several manuals for the Mylex controller.
There is also a command line version of this software with limited capabilities that you can run while the system is up.
/etc/dac960sh
and
/etc/dac960tk

There are man pages for both.

Update: I have managed to recover the missing filesystem via LoneStar software. The thing is everytime I manage to do a restore my filesystems shoot up to 100%.

I have a spare disk to add. Will expanding both root & u1 help & how will I be able to do it as well?

I managed to stick it out with the existing mirrored drives though.
Thanks in advnace for any more inputs.

bash-2.03# df -v
Mount Dir Filesystem blocks used free %used
/ /dev/root 3373036 2611284 761752 78%
/stand /dev/boot 30720 24672 6048 81%
/u1 /dev/u1 3373036 2611284 761752 78%

You cannot dynamically resize file systems.
The best solution is to acquire another (4th) disk and create a second mirrored system drive. These drives can be up to 33gb. Even though 72gb and greater drives are available, I can say from personal experience that timing issues arise.
After you have installed the two new drives, run the daccf configuration software to create and intialize a new system disk.
Then boot the system and run:
#mkdev hd
enter the scsi information should be adapter 0, channel 0, id 1, lun 1
in divvy use the entire disk for one file system, name it u2 for now.
Create two new directories /home and /u2 permissions same as /u1
When you have re-booted the system, start scoadmin and go to Filesystems, add the file system ( /dev/u2 to /u2), and mount now.
Copy all of /u1 to /u2
Go back into scoadmin ->filesystem ->modify and
modify /dev/u2 to mount to /u1 directory, and
modify /dev/u1 to mount to /home

Re-boot, and restore u1 from the tape.
Rebuild your lonetar emergency boot diskettes.
Later if you wish you can go into divvy and rename /dev/u1 to /dev/home, and /dev/u2 to /dev/u1

You will get a reasonably good performance gain, and some peace of mind, as well by also replacing the original 4.3gb disks with 18's or 33.

Update: Am still running out of space. I put in place a 17GB HDD to do jgt's suggestions with a hitch though. It seems that I can't rebuild my kernel as since /stand is out of space.

Would booting up in single user mode enable me to delete this? Appreciate more inputs again. Thanks.

-r--r----- 1 bin mem 3489806 Oct 24 15:52 unix
-rw-rw-rw- 1 root sys 1152419 Oct 23 06:02 __A2
-r--r----- 1 root sys 3489114 Oct 16 16:05 unix.orig
-r--r----- 1 bin mem 1989021 Jun 7 1999 unix.safe
drwxrwxr-x 3 bin auth 48 Jun 7 1999 etc
-r--r----- 1 bin mem 2014460 Jun 7 1999 unix.install
-r-------- 2 bin bin 6471 Jun 7 1999 bootos
-r-------- 2 bin bin 6471 Jun 7 1999 dos
-rwx------ 1 bin bin 2458 Jun 7 1999 sfmt
-r-------- 1 bin bin 72662 Jun 7 1999 boot
-r-------- 1 bin bin 52327 Jun 7 1999 link
drwxrwxrwx 2 root root 32 Jul 27 1998 lost+found

bash-2.03# df -v
Mount Dir Filesystem blocks used free %used
/ /dev/root 3373036 3316270 56766 99%
/stand /dev/boot 30720 24674 6046 81%
/u1 /dev/u1 3373036 3316270 56766 99%

%disk - - - type=S ha=0 id=0 lun=1 bus=0 ht=mdac

%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac

---------- Post updated at 09:09 AM ---------- Previous update was at 09:07 AM ----------

forgot to add this as well

bash-2.03# mkdev hd 10 0 0 1 1

Missing 'sdevice.d' file for host adapter driver '1'

Enter the prefix of the SCSI host adapter that supports this devi
or press <Return> for the default: 'alad'
Enter h for a list of host adapters or enter q to quit:

You can clean out some of the old kernels in /stand.
/stand is mounted in read only mode during normal operation.

Do this by:
#cd /
#umount /stand
#mount /dev/boot /stand
#cd stand
#rm (keep unix and unix.old, delete anything older, or gzip and copy to diskette)
#cd /
umount /stand
#mount -r /dev/boot /stand

We already habe successfullt removed those unwanted unix kernel files. We encountered another error in trying to do mkdev hd. Any more inputs? Thanks

%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac

bash-2.03# mkdev hd 10 0 0 1 1

Missing 'sdevice.d' file for host adapter driver '1'

Enter the prefix of the SCSI host adapter that supports this device
or press <Return> for the default: 'alad'
Enter h for a list of host adapters or enter q to quit:

The Host Adapter parameters will be automatically configured
\nDuring installation you may choose to overwrite all
or part of the present contents of your hard disk.
\nDo you wish to continue? (y/n) \c
y
idmkreg: can not locate or open 'Sdsk/device.dat'
\nThe hard disk installation program will now invoke /etc/fdisk.\nEntering 'q' a
t the following menu will exit /etc/fdisk,\nand the hard disk installation will
continue. \n\nIf you wish to exit the entire installation at this menu,\npress t
he <DEL> key.\n
/etc/fdisk: cannot open /dev/rdsk/2s0 for reading: No such device or address (er
ror 6)
/usr/lib/mkdev/hd: [: : integer expression expected
/etc/fdisk failed.
bash-2.03#

The default device that mkdev prompts for, is the last device used (so likely the tape was added last.)
Change the host adapter to mdac.
The scsi details are adapter 0 bus 0 scsid 1 lun 1

did what you suggested jgt....can't makedev the device. went to /dev. found these devices.

Appreciate more inputs, again. Thanks.

bash-2.03# ls -lt *hd1*
brw------- 1 sysinfo sysinfo 1, 64 Jun 8 1999 hd10
brw------- 1 sysinfo sysinfo 1, 79 Jun 8 1999 hd11
brw------- 1 sysinfo sysinfo 1, 87 Jun 8 1999 hd12
brw------- 1 sysinfo sysinfo 1, 95 Jun 8 1999 hd13
brw------- 1 sysinfo sysinfo 1,103 Jun 8 1999 hd14
brw------- 1 sysinfo sysinfo 1,111 Jun 8 1999 hd1a
brw-r----- 1 dos sysinfo 1,112 Jun 8 1999 hd1d
crw------- 1 sysinfo sysinfo 1, 64 Jun 8 1999 rhd10
crw------- 1 sysinfo sysinfo 1, 79 Jun 8 1999 rhd11
crw------- 1 sysinfo sysinfo 1, 87 Jun 8 1999 rhd12
crw------- 1 sysinfo sysinfo 1, 95 Jun 8 1999 rhd13
crw------- 1 sysinfo sysinfo 1,103 Jun 8 1999 rhd14
crw------- 1 sysinfo sysinfo 1,111 Jun 8 1999 rhd1a
crw-r----- 1 dos sysinfo 1,112 Jun 8 1999 rhd1d
bash-2.03# ls -lt *hd2*
brw------- 1 sysinfo sysinfo 139, 0 Oct 27 07:22 hd20
brw------- 1 sysinfo sysinfo 139, 15 Oct 27 07:22 hd21
brw------- 1 sysinfo sysinfo 139, 23 Oct 27 07:22 hd22
brw------- 1 sysinfo sysinfo 139, 31 Oct 27 07:22 hd23
brw------- 1 sysinfo sysinfo 139, 39 Oct 27 07:22 hd24
brw------- 1 sysinfo sysinfo 139, 47 Oct 27 07:22 hd2a
brw-r----- 1 dos sysinfo 139, 48 Oct 27 07:22 hd2d
crw------- 1 sysinfo sysinfo 139, 0 Oct 27 07:22 rhd20
crw------- 1 sysinfo sysinfo 139, 15 Oct 27 07:22 rhd21
crw------- 1 sysinfo sysinfo 139, 23 Oct 27 07:22 rhd22
crw------- 1 sysinfo sysinfo 139, 31 Oct 27 07:22 rhd23
crw------- 1 sysinfo sysinfo 139, 39 Oct 27 07:22 rhd24
crw------- 1 sysinfo sysinfo 139, 47 Oct 27 07:22 rhd2a
crw-r----- 1 dos sysinfo 139, 48 Oct 27 07:22 rhd2d

bash-2.03# divvy /dev/hd10
divvy: could not open temporary node: No such device or address
bash-2.03# divvy /dev/hd20
divvy: could not open temporary node: No such device or address

---------- Post updated at 12:24 PM ---------- Previous update was at 11:55 AM ----------

additional info:

Tue Oct 27 09:18:07 2009
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb

the command is
#mkdev hd

Also post contents of /etc/conf/cf.d/mscsi (don't need commented lines)

During POST, do you see a comment about 2 system drives found, or just 1
If you are only adding one drive, you still have to set it up as RAID 0.

POST says two drives installed. When taking a look at the last entry at /var/adm/messages though, it tells me a different thing?

Interface Systems, Inc. Cleo SNA PU 2.1 Installed
%cd-rom - - - type=IDE ctlr=sec cfg=mst dvr=Srom->w
%tape - - - type=S ha=0 id=2 lun=0 bus=0 ht=alad
%disk - - - type=S ha=0 id=0 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=1024 hds=128 secs=32 fts=sdb
mem: total = 786032k, kernel = 21740k, user = 764292k
swapdev = 1/41, swplo = 0, nswap = 782336, swapmem = 391168k
rootdev = 1/42, pipedev = 1/42, dumpdev = 1/41
kernel: Hz = 100, i/o bufs = 6652k

%cpu - 255 - unit=2 family=6 type=Pentium II (D)
%cpuid - - - unit=2 vend=GenuineIntel tfms=0:6:5:2
%fpu - - - unit=2 type=80387-compatible
Tue Oct 27 11:30:06 2009
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb

Looks like the second drive is installed twice???
Post the contents of /etc/conf/cf.d/mscsi

This is the boot disk (

%disk - - - type=S ha=0 id=0 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=1024 hds=128 secs=32 fts=sdb

This is the secondary disk
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb
%disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
%Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb

I don't think that it should print twice, but I no longer have a machine with the same config to look at.

When you add a disk, you must run mkdev hd twice.
The first time the drive is added to the kernel, the second, fdisk and divvy is run, you have to reboot between invocations.

  •   the licensees of The Santa Cruz Operation, Inc.  Such users have the   
    
  •   right to use, modify, and incorporate this code into other products    
    
  •   for purposes authorized by the license agreement provided they include 
    
  •   this notice and the associated copyright notice with any such product. 
    
  •   The information in this file is provided "AS IS" without warranty.     
    

*

  •   Scsi configuration master table                                        
    

*

  •   ha    = device name of SCSI host adapter driver e.g. ad                
    
  •   attach= device name of attached SCSI device driver e.g. Sdsk           
    
  •   number= host adapter number                                            
    
  •   ID    = controller number                                              
    
  •   lun   = logical unit number                                            
    
  •   bus   = host adapter bus number                                        
    

*
*ha attach number ID lun bus
*
wd Srom 1 0 0 0
mdac Sdsk 0 0 1 0
mdac Sdsk 0 1 0 0
alad Stp 0 2 0 0
alad Sdsk 0 10 0 1
mdac Sdsk 0 10 0 1
mdac Sdsk 0 1 1 0

---------- Post updated at 04:10 PM ---------- Previous update was at 04:08 PM ----------

how can i delete/remove this?

*
wd Srom 1 0 0 0 # CD drive
mdac Sdsk 0 0 1 0 # Boot Disk
mdac Sdsk 0 1 0 0 # delete this one
alad Stp 0 2 0 0 # tape drive
alad Sdsk 0 10 0 1 # delete this one, unless you added a disk to the
Adaptec controller with SCSI id 10
mdac Sdsk 0 10 0 1 #delete this one
mdac Sdsk 0 1 1 0 #this is the secondary drive

Edit this file with a text editor, and relink the kernel, and reboot.

Run mkdev hd and enter the same data for mdac 0 1 1 0
This should start fdisk and divvy.

I think I did it wrong, again. Here were the results:

What is the LUN of this device?
Press <Return> to use the default: 0
Select 0-7, or h for help, or q to quit: 0

You are about to add the following SCSI device:

Host
Adapter Adapter
Type Device Number ID LUN BUS
----------------------------------------------
mdac Sdsk 0 1 0 1

Update SCSI configuration? (y/n) y

The SCSI configuration file has been updated.
\nA new kernel must be built and rebooted before disk configuration can continue
. Would you like to relink at this time? (y/n) \c
y

    The UNIX Operating System will now be rebuilt.                          
    This will take a few minutes.  Please wait.                             
                                                                            
    Root for this system build is /                                         

idmkreg: can not locate or open 'Sdsk/device.dat'

*
wd Srom 1 0 0 0
mdac Sdsk 0 0 1 0

alad Stp 0 2 0 0

mdac Sdsk 0 1 1 0

mdac Sdsk 0 1 0 1
bash-2.03#

---------- Post updated at 09:03 AM ---------- Previous update was at 08:45 AM ----------

deleted the one in red & rebuilt the kernel.....tried this.....but this is pointing to a specific disk (in red)??

Enter the prefix of the SCSI host adapter that supports this device
or press <Return> for the default: 'mdac'
Enter h for a list of host adapters or enter q to quit:

Which 'mdac' SCSI host adapter supports this device?
Select 0-4, or enter q to quit: 0

The Host Adapter parameters will be automatically configured

What SCSI Bus is this device attached to?
Press <Return> to use the default:0
Select 0-n, or h for help, or q to quit: 0

What is the Target ID for this device?
Select 0-15, or h for help, or q to quit: 1

What is the LUN of this device?
Press <Return> to use the default: 0
Select 0-7, or h for help, or q to quit: 1
\nDisk already configured as disk number 1 (/dev/dsk/1s0)
\nDuring installation you may choose to overwrite all
or part of the present contents of your hard disk.
\nDo you wish to continue? (y/n) \c

---------- Post updated at 09:15 AM ---------- Previous update was at 09:03 AM ----------

I don't know if this is the right one? Would this be my new 17GB hard drive that I need?? Appreciate more. Thanks.

bash-2.03# divvy /dev/hd3a
+-------------------+------------+--------+---+-------------+------------+
| Name | Type | New FS | # | First Block | Last Block |
+-------------------+------------+--------+---+-------------+------------+
| | HTFS | no | 0 | 0| 15997967|
| | NOT USED | no | 1 | -| -|
| | NOT USED | no | 2 | -| -|
| | NOT USED | no | 3 | -| -|
| | NOT USED | no | 4 | -| -|
| | NOT USED | no | 5 | -| -|
| | NOT USED | no | 6 | -| -|
| hd3a | WHOLE DISK | no | 7 | 0| 15999999|
+-------------------+------------+--------+---+-------------+------------+
15997968 1K blocks for divisions, 2032 1K blocks reserved for the system

n[ame] Name or rename a division.
c[reate] Create a new file system on this division.
t[ype] Select or change filesystem type on new filesystems.
p[revent] Prevent a new file system from being created on this division.
s[tart] Start a division on a different block.
e[nd] End a division on a different block.
r[estore] Restore the original division table.

Enter your choice or q to quit:

That's it, create one new HTFS partition, then follow the instructions in scoadmin->filesystems to mount.

15997968 1K blocks for divisions, 2032 1K blocks reserved for the system     
                                                                             

Multiplies out to about 17gb

Thanks. I now want to remove the previous mount point /u1 from the system. How do I do it? I tried to define the new drive as /u1, got a message instead of /u1 in use. I have already removed the u1 directory. Am I missing something? Appreciate more. Thanks.

The divvy program looks at the file system names for all disk drives.
On disk 0 you already have a /dev/u1
On disk 1 you can add a /dev/newu1

/u1 is a directory that is filled with the data that resides on /dev/u1 when you mount the file system.

See post #43 for the details of moving the data from the old u1 to the new u1.

You are going to want a directory /u1. That is the filesystem tie point for your data, tied to either /dev/u1 from your old data or to /dev/newu1 for your new data and is required for your programs to find the stuff they need.

I would suggest that you create a directory called newu1 for the tie point for your new hard drive and use that for the transfer as JGT suggested from post 43. It is awfully easy to forget where you are when you are moving around in directories.

And as an alternative to copying stuff from directory to directory you could also create a backup directory (newu1/backup) and tar your /u1 to it, then when the /dev/newu1 is attached to /u1 you could tar it back off the /u1/backup directory that will show up with the filesystem change.

JGT may have more comments on this.

thanks. post 43? can't seem to find it? would you show me how? please & thanks.

---------- Post updated at 10:16 AM ---------- Previous update was at 09:19 AM ----------

I have made the new filesystems. Do I manually need to mount them? More inputs please. Thanks.

i[nstall] Install the division set-up shown
r[eturn] Return to the previous menu
e[xit] Exit without installing a division table

Please enter your choice: i

Making filesystems
bash-2.03#

---------- Post updated at 10:27 AM ---------- Previous update was at 10:16 AM ----------

Added 2 new mounts. Now to figure out the links..from /u1 to /flexxu1.

Mount Dir Filesystem blocks used free %used
/ /dev/root 3373036 2363442 1009594 71%
/stand /dev/boot 30720 22408 8312 73%
/flexxu1 /dev/flexxu1 26000002 815686 25184316 4%
/backup /dev/backup 4000000 125572 3874428 4%

---------- Post updated at 10:34 AM ---------- Previous update was at 10:27 AM ----------

Am I still missing something? Did a reboot & the 2 new filesystems created didn't mount automatically.

bash-2.03# df -v
Mount Dir Filesystem blocks used free %used
/ /dev/root 3373036 2363518 1009518 71%
/stand /dev/boot 30720 22408 8312 73%
bash-2.03# mountall
fsstat: /dev/boot mounted
Mounted /stand filesystem
bash-2.03# mount /dev/flexxu1 /flexxu1
bash-2.03# mount /dev/backup /backup
bash-2.03# df -v
Mount Dir Filesystem blocks used free %used
/ /dev/root 3373036 2363528 1009508 71%
/stand /dev/boot 30720 22408 8312 73%
/flexxu1 /dev/flexxu1 26000002 815686 25184316 4%
/backup /dev/backup 4000000 125572 3874428 4%
bash-2.03#

Ususal way for automatic mounting is "mkdev fs".

Have your directory structure finished first by creating directories flexxu1 and backup, both on root.

run "mkdev fs" twice, once for flexxu1 and once for backup. point both hardware attachements to /dev/xxx and the directories to /xxx, don't let users mount, and clean if dirty.

Next boot they will mount. Both will create lost+found directories.

You can get some instructions via "man mkdev", the built in manual system.

---------- Post updated at 11:51 AM ---------- Previous update was at 11:26 AM ----------

If you attempting the method I suggested this is how you would proceed:

on next login try "df -kv" to verify that the filesystems are in place

tar cvf /backup/u1backup.tar /u1 does a tar of u1
mkdev fs and remove the flexxu1 filesystem
mkdev fs and add the flexxu1 file system to your directory /u1
reboot the system
tar xvf /backup/u1backup.tar which will pull everything back into /u1
you now have your larger /u1 directory mounted on the new drive.

be aware that if the flexxu1 filesystem fails to mount for any reason you will be seeing old data at /u1.

You may at some time in the future want to unmount flexxu1 and wipe the old data out. But not now.