On Wednesday 03 December 2008 8:08:51 am you wrote:
> On Wed, Dec 3, 2008 at 8:25 AM, Chris Hoogendyk <hoogendyk AT bio.umass DOT 
> edu> 
wrote:
> > hmm. I haven't noticed that behavior. Is there some trick to making it
> > work?
>
> just havereader=1 in changer.conf
>
> > I've had no particular problem, but Amanda does seem to cycle through on
> > my system. It does keep the barcodes listed in changer-barcodes. Maybe my
> > library doesn't have that quick access feature? On the other hand, I
> > always keep the tapes in order, so the issue rarely comes up.
>
> That's probably why you're not seeing it -- it's hard to distinguish
> between loading the "next" tape, and loading the tape with the next
> desired label.
>
> If you look at your changer debug file, you can verify that the loads
> are by barcode.
>
> Dustin

Hi Dustin:

I have an ADIC Scalar24 with barcode reader enabled, running amanda-2.6.0p1, 
and it doesn't actually look like that is true for me.

Here is the output from an amcheck session in which amanda had to shuffle 
through several slots to find the next tape:

slot 12:read label `PhysicsSet117', date `20081111201501'.
Tape with label PhysicsSet117 is still active and cannot be overwriten.
slot 13:read label `PhysicsSet120', date `20081110201502'.
Tape with label PhysicsSet120 is still active and cannot be overwriten.
slot 14:read label `PhysicsSet309', date `20081031201502'.
Tape with label PhysicsSet309 is still active and cannot be overwriten.
slot 15:read label `PhysicsSet310', date `20081101201502'.
Tape with label PhysicsSet310 is still active and cannot be overwriten.
slot 16:read label `PhysicsSet135', date `20081031201502'.
Tape with label PhysicsSet135 is still active and cannot be overwriten.
slot 17:read label `PhysicsSet134', date `20081030201502'.
Tape with label PhysicsSet134 is still active and cannot be overwriten.
slot 18:read label `PhysicsSet124', date `20081029201502'.
Tape with label PhysicsSet124 is still active and cannot be overwriten.
slot 19:read label `PhysicsSet136', date `20070722'.
NOTE: skipping tape-writable test
Tape PhysicsSet136 label ok
Server check took 591.688 seconds

It looks like amanda ignored the barcodes and actually read the amanda labels 
from each tape in succession.  Otherwise it surely would not have taken ten 
minutes.

I have havereader=1 in changer.conf
There is a valid and correct changer-barcodes file

'mtx status' reports the barcodes for all tapes in the changer, so the changer 
itself is aware of the barcodes.  

But somehow amanda chooses not to use that information.  She just goes from 
tape to tape, physically reading the on-tape labels, as though there were no 
barcodes.  I would have expected that amanda would determine the correct slot 
based on the barcodes, and then would have verified the on-tape label.  But 
maybe I'm confused about what the expected behavior is.

This example was from 'amcheck' but 'amrecover' does the same thing.

-- 
--------------------------------------------------------------
Jeffrey Anderson                        | jdanderson AT lbl DOT gov
Lawrence Berkeley National Laboratory   | 
Office: 50A-5104E                       | Mailstop 50A-5101
Phone: 510 486-4208                     | Fax: 510 486-6808