Well, that's the strange part. I'm confident it's not a conflict - when
there is a conflict lbtest returns an error and I can check for that.  The
library is returning an inventory, providing all info about what is in each
slot *except* for the barcode.... :(

Paul

On Mon, May 9, 2011 at 4:26 PM, Marcel Anthonijsz <marcel AT anthonijsz DOT 
net>wrote:

> Paul,
>
> No, I have not seen this before: I use IBM Atape and tapeutil -f /dev/smc0
> inventory without any problem.
>
> Make sure that TSM or another program is not using the library device at
> the
> same time or program around that.
> Maybe a previous perl script that hangs/waits until next timeout?
>
> Marcel
>
> 2011/5/9 Paul Fielding <paul AT fielding DOT ca>
>
> > I have a wierd issue I'm hoping someone has seen before.   I have a 3584
> > tape library attached to an AIX box.  I have a bunch of perl scripts that
> > deal with tape automation.  Part of this process uses lbtest to grab the
> > current contents of the IO door.   However, I've had it happen a few
> times
> > where things stopped working, and when I investigate I find that lbtest
> has
> > suddenly stopped returning barcodes.   All of the elements show FULL or
> > EMPTY, but no barcodes are listed, and barcode_len = 0.
> >
> > If I go to the library web gui, I see the barcodes.  And if I just leave
> > things alone, in 24h or so it seems to go back to normal.  Nothing I do
> to
> > the library (including rebooting it) seems to fix the problem
> immediately.
> >
> > Very very wierd.  Anyone seen this before?
> >
> > regards,
> >
> > Paul
> >
>
>
>
> --
> Kind Regards, Groetje,
>
> Marcel Anthonijsz
> T: +31(0)299-776768
> M:+31(0)6-2423 6522
>