I have a script which runs overnight to backup a database into a tar file. I then gzip the file and send it off to a tape device.

The command I'm using for the tape backup is

tar cvf /dev/rmt/0n /var/dbbackup/${FILE}.gz

I thought the n meant no rewind so the next file is copied to tape but does not overwrite what's already there. However it seems as though the old files on the tape device are being overwritten i.e. when I do

tar -tvf /dev/rmt/0

I can only see last night's file
-rw-r--r-- 100/100 102192 Jan 23 03:15 2004 /var/dbbackup/AS1.2004012303.bak.gz

What's up with my tar options?

One possible problem - you mention looking at the tape to see what's on it but you don't use the no rewind option. Anytime you don't use it, the tape will ,by default, rewind. Or anytime anyone doesn't use it the tape will rewind. Check that no one else is attempting to hit the drive - no other cron jobs, users,...

I've had a look but I can't find other jobs hiting this tape.

While investigating I failed the applications over to the secondary box and ran;
# shutdown -i 0
then
ok probe-scsi-all

This sent the box into a "fault bus" error message hissy fit resulting in me having to pull the power cable out the back. Any more suggestions?

Why would you shutdown the system? (I'm glad I didn't suggest that!)

If this is a SUN system, then sometimes after a shutdown it will hang by doing a probe-scsi-all. Normally all you have to do is a reset and break out (Stop-A) of the boot process to be able to run the proble-scsi-all.

But if you are getting error messages, you might want to see if you have your tape drive set up on the same id as a disk drive (which could cause errors).

Please post the system hardware, OS, and error messages you saw.

The system info is Sun Fire V120 (UltraSPARC-IIe 648MHz), with 1024 MB RAM.

The reason I did a probe-scsi-all was that mt - f /dev/rmt/0 status said no device found. I tried shutting down with

sync; init 0

and probe-scsi-all seemd to better like it.

ok probe-scsi-all
/pci@1f,0/pci@1/scsi@8,1

/pci@1f,0/pci@1/scsi@8
Target 0
Unit 0 Disk SEAGATE ST336607LSUN36G 0307
Target 1
Unit 0 Disk SEAGATE ST336607LSUN36G 0307

ok

and when I do a boot -r the following error is given

NG: /pci@1f,0/pci@1/scsi@8,1 (glm1):
SCSI bus DATA IN phase parity error
WARNING: /pci@1f,0/pci@1/scsi@8,1 (glm1):
SCSI bus DATA IN phase parity error
WARNING: forceload of misc/md_trans failed
WARNING: forceload of misc/md_raid failed
WARNING: forceload of misc/md_hotspares failed
WARNING: forceload of misc/md_sp failed
configuring IPv4 interfaces: eri0 eri1.

As far as I am aware nothing has changed, any ideas?