Re: [Bacula-devel] 2.1.10 beta: no appendable volume found problemwith multidrive autochanger

[Kern S. <ke...@si...>]

On Friday 25 May 2007 21:17, David Boyes wrote:
> 
> > > > In
> > > > non-autochanger configurations, the StorageId is not significant
> since
> > > the
> > > > volume can be mounted on any device with the same Media Type.
> > >
> > > Would this argue for defining manual drives as an autochanger with a
> > > special/null changer device? That would allow the device selection
> code
> > > to be completely consistent regardless of device type or
> configuration.
> > 
> > I don't understand the problem you are trying to solve.  
> 
> The question is that manual tape drives are treated as a special case
> and there doesn't need to be a special case if we choose so.  A manual 
> tape drive can be thought of as an autochanger with a null changer
> device and a maximum of one slot (the drive itself) and a magazine size
> of 1. Using this definition would allow treating manual drives and
> robotic-assisted drives the same way conceptually and in the
> documentation. 

Possibly it would simplify the documentation, but it would complicate the 
configuration for users without autochangers -- they would have to add and 
Autochanger resource and all the various minor complications it adds.

Bacula already treats standalone drives and autochangers pretty much the same, 
but there is one major distinction -- a limitation, if you will, for 
autochangers.  As Arno so eligantly put it, we must restrict  "any attempts 
to throw or teleport tapes from one changer to the other :-)", which is not 
the case for a standalone tape drive.

> 
> Perhaps there isn't a problem, but unnecessary special cases are a
> concept that bothers me, so I asked the question. 

Good question, but I'm not sure there is much for me to do.