We just went from HP-UX 10.20 to 11.00.
All the bugs are out, except for a small detail...the old 9-track isn't working.
We receive large amounts of data on reel-to-reel so I kinda need it fixed. In the meantime, I'm trying to see if they can send me a DAT tape instead.
I'm pretty sure the drive uses 1600 compression. So I created a custom device file for it, but to no avail. The error I get is, Check Ring, Privileges etc.
The user is root, so I can't see it being privileges. Argh.
The old device file was called /dev/rmt/1600m. I can't seem to duplicate that file. I tried everything I can think of.
Does anyone know of any tricks to share?
Run "ioscan -fnC tape" and post the results.
The device shows up, but for some reason, it's unreadable.
tape 1 10/4/8.4.0 tape2 CLAIMED DEVICE HP 7980S
/dev/diag/rmt/c1t4d0 /dev/rmt/c1t4d0BESTb
/dev/rmt/1m /dev/rmt/c1t4d0BESTn
/dev/rmt/1mb /dev/rmt/c1t4d0BESTnb
/dev/rmt/1mn /dev/rmt/c1t4d0D1600
/dev/rmt/1mnb /dev/rmt/c1t4d0wD1600b
/dev/rmt/c1t4d0BEST
I don't have an 11.0 system with a tape2 driver but the man page implies that this should work. How were the 1600 special files created? What does
lsdev /dev/rmt/c1t4d0D1600
and
lsdev /dev/rmt/c1t4d0wD1600b
report?
I tried creating the special device files for D1600 compression, but those don't seem to work.
I really don't know how the previous dev file was created because it was just "always there". It was lost when I re-loaded unix 11.0.
The program that needs to access it has an entry for the device file I can change, and also has this entry:
Device_6=48~48~MT~M/UX~~("avl":0:2048)~~9 track 6250
I don't know if this needs to be changed.
I really need to see what you get from the commands
lsdev /dev/rmt/c1t4d0D1600
and
lsdev /dev/rmt/c1t4d0wD1600b
Opps I don't mean lsdev..I mean lssf. Sorry about that.
Here's what I'm getting:
# lssf /dev/rmt/c1t4d0D1600
tape2 card instance 1 SCSI target 4 SCSI LUN 0 at&t 1600 bpi 9 track at addr
10/4/8.4.0 /dev/rmt/c1t4d0D1600
and
# lssf /dev/rmt/c1t4d0wD1600b
tape2 card instance 1 SCSI target 4 SCSI LUN 0 berkeley wait 1600 bpi 9 track at
address 10/4/8.4.0 /dev/rmt/c1t4d0wD1600b
I don't see anything wrong with your configuration. Your special files are right. You have the right driver. The kernel can see the 7980.
Pull the drawer out and watch the tape. Does it move? Can you unload the tape with "mt offl"? Have you tried another tape?
I'm pretty sure that you should be able to read a 6250 tape with a 1600 device file. But I don't see an explicit statement to that effect. You might try making a 6250 device file and see if that works any better.
mksf -v -C tape -I 1 -b D6250 -n
should do it.
The only other thing that I suggest is to check for patches to the tape2 driver.