SCO

[bronzini]

I've got an old Pentium II machine from 1995 running SCO. I'm trying to backup some information, but root is full. The backup only allows dumping to root, so I need to know which folders I might be able to temporarily move to another mount while I create the backups. I need to clear up 50-100 MB. I don't know hardly anything about SCO and I don't want to cause myself any headaches. Any suggestions? The other folders in root that I see are:

.odtpref
.scohelp
.xdt_dir
bat
bin
dev
etc
ibin
ipsraid
lib
lost + found
mnt
opt
pmd
rac
sbin
shlib
stand
udk
var

Thanks for you help in advance!

 

[PHV]

I don't think that rac is a system folder.

Hope This Helps, PH.
FAQ219-2884
FAQ181-2886

 

[motoslide]

I'm not sure why your backup only allows you to dump the information into the root filesystem.
Maybe you could see how much disk space is occupied in each of these directories. When "root" is full, there's usually a reason and it can often be solved by locating a very large log file, or previous backup error.

Common placed to look:
/usr/spool/lp/logs/
/usr/spool/mmdf/lock/home/
/dev/
/usr/spool/mail/
/tmp

I'm not suggesting you remove anything you find in these areas, but it could be helpful to explore their contents:

# du -s /usr/spool/lp/logs
(etc)

Can you use FTP to backup the data you want onto a different platform?



"Proof that there is intelligent life in Oregon. Well, Life anyway.

 

[edfair]

If you have been running since 1995 without looking at anything you probably have enough mail to be taking up the space. You can move the /usr/spool/mail/root file anywhere else and the system will create a fresh one.

If the backup is lone-tar, their log file would account for a good bit of the space you need. I can't remember where it resides, but it grows huge, and it will create a new one if you move the original.

It might help if we knew the version you are running, "uname -X" to find it, the alternate space available, "df -k -v" to find it, and the backup program you are using, "custom" will report the software installed and "crontab -l " (L) will report the scheduled tasks which are running.

Do you know what software (applications) you are running and where the data is stored. It is possible that you have two problems, the data has grown, and the file space to back it up has increased to match it. And if the data is on root you are at risk of losing it all if anything glitches.

And one other significant question, do you have an emergency boot floppy set?

Ed Fair
Give the wrong symptoms, get the wrong solutions.

 

[edfair]

Although there are 2 directories (bat & rac) that do not appear to be for the OS it is possible that they are strictly mount points to get access to alternate filesystems or folders used for data or programs and are not part of your problem.

I suspect that you could "cat /etc/default/filesys" and get a report of the file systems you have and their mount points. Depends on the OS, but from the age is probably OS5.0.0 or 5.0.2.

Ed Fair
Give the wrong symptoms, get the wrong solutions.

 

[bronzini]

Thanks for your help. Didn't quite solve my problem today, but I'll have to come back to it some other time. The aggrevating issue is that the backup utility was set up by the software company. They monitored the software remotely. We no longer subscribe, but our machines are getting sketchy, so we want to back it all up. But we have no admin rights within the database software, so we can't change any settings, including the backup utility from what I can tell. But thanks for your help!

 

[edfair]

Monitoring backup software remotely can be as simple as logging on as root via modem and reading and deleting mail messages. In the case of backupedge just opening mail shows the subject matter of whether the backup succeded or failed. Lone-tar is about the same, but it has been a while since I cleaned one of those out.
In the systems I monitor, whether I have administrative rights or not, the backups and the possible restores work independently of those rights. Root access gives me everything I need.
I hit each of the systems (that still have modems) once a week and delete backup messages older than a week and the ISO creation message and CDR burn message of any day when both complete successfully.
If you have SCSI drive hardware you could probably add an external hard drive, tape drive, or CDR and do essential backups to it. Depends on the controller you have.

Ed Fair
Give the wrong symptoms, get the wrong solutions.