There is no unused space on the disk.
Notice that root and stand and swap account for all the blocks on the drive.
Run fdisk and see if the drive is split into two partitions. If it is then u1 is probably the second partition.
If this is so, run mkdev hd and add the second partition/drive, do not create a "new" file system, or do any destructive bad track testing.
If there is no second partition then the data that u1 contained is gone.
If the drive is only 2.1gb in total then there are no other partitions.
# fdisk
1. Display Partition Table
2. Use Entire Disk for UNIX
3. Use Rest of Disk for UNIX
4. Create UNIX Partition
5. Activate Partition
6. Delete Partition
7. Create Partition
Enter your choice or 'q' to quit: 1
Current Hard Disk Drive: /dev/rhd00
\+-------------\+----------\+-----------\+---------\+---------\+---------\+
| Partition | Status | Type | Start | End | Size |
\+-------------\+----------\+-----------\+---------\+---------\+---------\+
| 1 | Active | UNIX | 1 | 130943 | 130943 |
\+-------------\+----------\+-----------\+---------\+---------\+---------\+
Total disk size: 131072 tracks (129 reserved for masterboot and diagnostics)
I do have a backup for this. would extracting it to root fix the problem? Just did a #tar tvf /dev/rct0 on it & contained a full backup on all the filesystems. Would restoring it be a fix to it?
Thanks.
First.
#tar tvf /dev/rct0 >/tmp/tapelist
Examine the contents of tapelist and then decide whether to:
#tar xvf /dev/rct0 /u1
or
#tar xvf /dev/rct0 u1
In both cases this will create a directory in the root file system called u1.
Also you can check /etc/default/tar, in all likelihood, "tar tvf /dev/rct0" can be shortened to "tar tv8" where 8 is used to substitute the appropriate line from /etc/default/tar.
There will be no difference in system operation whether u1 is a directory or a file system, other than not having to mount a u1 file system.
If you need other files off the tape, create a directory called oldroot, and extract the tape into it, (check for absolute file names on the tape.)
Offhand you may want to look at /etc/passwd, to find the uid numbers, and add the users back the same way,
/usr/spool/lp/admins/lp/interfaces for code added to the printer models.
/usr/spool/cron/crontabs
/etc/rc2.d
I did the following:
#tar tvf /dev/rct0 ------this showed all the filesystems were backed up in full
#tar xvf /dev/rct0 / -----tried to run this hoping that when it restores root, it fixes my missing filesystem domain.
Dunno if I am doing it right, appreciate more inputs. Thanks
You really don't want to restore from /. If you had problems at the point you did the backup you'll drag them back into your system.
You would be better served following JGT's advice and selectively restoring the /u1 (that was the one, wasn't it?) .
There may be a problem with JGT's advice about restoring stuff to /oldroot. It depends on how much space is left on root after u1 is pulled back in and how much is stored on the tape. Doing a full restore to oldroot is going to pull /u1 into oldroot too. You may need to do a partial restore to get just those things you may need.
I'll put in a plug here for backupedge or lonetar. They are a better backup method than tar due to verifications and reporting capabilities. I've used both but find lonetar's reporting to create more messages than I need.
good point. anyways, i've discovered that the backups that i have were run by a script using lone tar software. i've already ordered from them awhile ago. my current contingencies:
- re-install os, re-create filesystems.
- increase partition size of root, swap & u1.
- install lone tar & use a utility called airbag that is packaged with it to restore the last good bakup that I have.
or do this:
- add a new disk & create a partition soley for u1.
- restore using lone tar software.
my only hitch is as follows:
- I have no media, os licenses ( i guess & assume that restoring from backup will take care of this.
- I have no recovery/boot diskette's.
Am going for contingency no# 2 since I do not need any media.......
More inputs are highly appreciated.
I would go for alternative 1 with a different hard drive. Cleaner in the long run. Media, patches, and license are resolvable. No need to diddle with the division sizes, they would automatically generate what you need.
That being said, you might post your hardware specs. It is easy to say what you should do, based on our experiences, but you may be locked into something that would preclude the actions we would take.
here it is...reposting....
Ed is right, we need the output of "hwconfig -hc" or the contents of /usr/adm/messages, starting at the last boot, and the output of "swconfig".
There are inconsistencies.
For instance desktop systems are 2 user, but do not support HTFS file systems, they use DTFS instead.
The 2 user reference from the output of uname -X could be because the license manager was not running.
I missed the post about there being a mirrored drive.
Jedi, you mentioned in one post moving serial cables, is this a host or enterprise version of SCO. The host version does not have tcp, and all terminals are attached through serial ports. The enterprise version has tcp and all the LAN functionality.
You can find this out by running scoadmin and selecting license manager.
I think i zeroed out messages previously from earlier posts. I am currently adding a third standalone scsi hard disk to use for /u1. The system is still initializing @ 50%. Will post the hwconfig -hc as soon as it is done.
Just got the Lonetar software last night. Tried to FTP it from my PC to my SCO box.....unfortunately, there is not entry for /etc/ftpusers.....dunno if this is the right place to enable FTP in sco.
appreciate all the help. thanks.
How about trying "divvy /dev/u1".
Lacking a response from that, look in /dev for all devices starting with hd* to see what the system thinks is there.
Since df shows a /dev/u1 and divvy doesn't, I suspect that you may not be mirroring, but actually have /dev/u1 as the second drive. That is the only explanation I can come up with for the evidence I see.
You should see some indication of a private message I sent.
---------- Post updated at 11:46 AM ---------- Previous update was at 11:45 AM ----------
And just to make sure of the other things that may be important, how about listing the installed software that shows up when you run "custom".
tried to do this:
bash-2.03# divvy /dev/u1
divvy: could not open temporary node: No such device or address
bash-2.03# cd /etc/
bash-2.03# divvy /dev/u1
divvy: could not open temporary node: No such device or address
device address vec dma comment
====== ======= === === =======
kernel - - - rel=3.2v5.0.5 kid=98/07/02
cpu - - - unit=1 family=6 type=Pentium II (D)
cpuid - - - unit=1 vend=GenuineIntel tfms=0:6:5:2
fpu - 13 - unit=1 type=80387-compatible
pci 0xcf8-0xcff - - am=1 sc=1 buses=3
PnP - - - nodes=0
serial 0x3f8-0x3ff 4 - unit=0 type=Standard nports=1 fifo=yes
serial 0x2f8-0x2ff 3 - unit=1 type=Standard nports=1 fifo=yes
console - - - unit=vga type=0 12 screens=68k
adapter 0xd000-0xd0ff 5 - type=alad ha=0 bus=0 id=7 fts=sto
floppy 0x3f2-0x3f7 6 2 unit=0 type=135ds18
kbmouse 0x60-0x64 12 - type=Keyboard mouse
adapter 0x1f0-0x1f7 14 - type=IDE ctlr=primary dvr=wd
adapter -0x43 11 - type=mdac (v4.15) 2-chnl 16-ids ha=0
eeE0 0xb400-0xb43f 16 - type=EE PRO/100+ 00:90:27:2d:e3:be
epca 0x208-0x208 - - mem=0x000D0000 ports=32 ISA C/X V7.0.5
cd-rom - - - type=IDE ctlr=sec cfg=mst dvr=Srom->wd
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
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
disk - - - type=S ha=0 id=1 lun=1 bus=0 ht=mdac
Sdsk - - - cyls=8750 hds=128 secs=32 fts=sdb
Hope this helps. The new hard drive is in place. Do I need to run anything to get SCO to recognize it?? Thanks again.
The /etc/ftpusers file is a deny file, but ftp is not available with the host version.
You should be able to install the lonetar software directly from the media.
From your hardware list you have a 16gb (roughly) hard drive sitting at ID1 with a suspicion that it is running off the adaptec controller you have installed.
Your command "divvy /dev/u1" may have failed because the entry got corrupted. That is the divvy table entry in devices, there is also the hardware entry. They can also get corrupted or wiped.
I suggest that you get that drive up and running before you try to install another. If the drive stuff got corrupted in /dev, recreating it will bring your /u1 back to life. You just have to make sure to run the "mkdev hd" once, then run divvy to get things back in order. But don't do it until you've checked some of the other stuff here.
How about giving the results from "ls /dev/hd*. (L not 1)
Another thing to try is [CTRL] A while the BIOS is posting. That should bring up an adaptec bios with some diagnostics, one of which is a channel scan for attached devices.
If you are attaching a new drive into the existing chain, the ID should be 2 to 6 and no termination.
To make the new drive available you run "mkdev hd" and assign it to alad controller 1 with ID (whatever you plugged). That creates the /dev entry that you will use. Then reboot to get it attached and run mkdev again to create the division that you need and name it. During this run the drive will get formatted (along with media scan if you allow it) and afterwards will be available for "mkdev fs" to create the filesystem and attach it (if you wish to attach it immediately).
I burnt a copy on a cd-r(for lonetar). Then tried to mount the cd, still won't mount?
---------- Post updated at 12:09 PM ---------- Previous update was at 12:01 PM ----------
Would this help?
bash-2.03# ls /dev/hd
/dev/hd00
/dev/hd01
/dev/hd02
/dev/hd03
/dev/hd04
/dev/hd0a
/dev/hd0d
/dev/hd10
/dev/hd11
/dev/hd12
/dev/hd13
/dev/hd14
/dev/hd1a
/dev/hd1d
You have a Mylex DAC960 raid controller with presumably two mirrored disks. You have an Adaptec controller, with the tape drive attached to it. You have a CD-ROM attached to the secondary IDE channel as a master.
There are two cpu's.
If you have attached the new disk to the Mylex controller, then you have to first boot from a DOS diskette and run the daccf.exe configuration program to create a logical disk.
When you run divvy, the argument is a device, not a file system.
So either mkdev /dev/rhd0a, or /dev/rhd1a
should read divvy /dev/rhd0a, or /dev/rhd1a
divvy /dev/hd10
tell us what comes up.
---------- Post updated at 02:07 PM ---------- Previous update was at 02:03 PM ----------
JGT has experience with stuff I've never even seen, so answers to his questions will probably be more important than mine.
bash-2.03# divvy /dev/hd10
divvy: could not open temporary node: No such device or address
bash-2.03# cd /dev
bash-2.03# divvy /dev/hd10
divvy: could not open temporary node: No such device or address
bash-2.03# cd /etc
bash-2.03# divvy /dev/hd10
divvy: could not open temporary node: No such device or address
Because of the difference in sizes of the drives I had suspected that they were standalone, not mirrored, and that your /u1 was an attached division on the drive with ID1. That is still a possibility but it will take someone with experience with the controller to sort it out.
If they were standalone, and your drive1 had corrupted in certain ways, you could end up with problems like you are getting. The lack of a division separately identifed for use with /u1 was the reason I suspected it to start with.
I may kick in a thought or two down the road but JGT is the voice of experience from here out.
Good luck with it.
Found this file, created by lonetar software. the original filesystem on the SCO box I am working on. Would this be of help?
FILE MOUNTED FS SIZE
SYSTEM ON TYPE (1024 byte blocks )
/dev/root / HTFS 1686518 total blocks
/dev/boot /stand EAFS 15360 total blocks
/dev/u1 /u1 HTFS 2342912 total blocks