Jon LaBadie <jon AT jgcomp DOT com> writes:

> On Wed, Apr 04, 2007 at 11:01:20AM +0200, Sebastian Henrich wrote:
>> > On Tuesday 03 April 2007, Sebastian Henrich wrote:
>> > >Hello,
>> > >
>> > >I'm using amanda 2.4.5 to backup my server. The tape is a HP DAT 72.
>> > >This drives stores 36 GB of uncompressed data. I switched the
>> > >hardware compression off and tried to use the software compression.
>> > >Now I have the problem that amanda tells me every time
>> > >
>> > You may be getting bit by something I ran into several years ago now.
>> > 
>> > If the tape is written with the hardware compression on, then the drive 
>> > will detect that and turn the compression on even if you think its turned 
>> > off.
> ...
>> 

>> I diabled the hardware compression before I used the tapes for the
>> first time like it's described in the docs. mt shows me that
>> hardware compression is off after inserting a tape. So I think that
>> it's really disabled.

> The switch may not occur unless the drive actually attempts to read
> the tape.  Amanda does an initial read before any attempt to write
> to a tape to check if it is actually an amanda tape and is the
> expected amanda tape.

I had this problem for years (on Linux); then someone on this list
mentioned mt-st and stinit. Now this works for me reliably even in
this case.

Some older versions of stinit (or/and Linux?) didn't work. Currently I
use:

  # stinit --version
  stinit v. 0.9b
  # uname -r
  2.6.8.1-common
  # cat /etc/stinit.def 
  {buffer-writes read-ahead async-writes}

  manufacturer=QUANTUM model=SDLT600 {
  mode1 blocksize=0 compression=0
  }


        Sven