-----Original Message-----
From: John R. Jackson [mailto:jrj AT purdue DOT edu]
Sent: Wed 9/21/2005 10:35 PM
To: Lengyel, Florian
Cc: amanda-users AT amanda DOT org
Subject: Re: Wait for dumping

>mt -f /dev/nst0 compression off
>
>[by the way, the argument "off" is worth a small fortune in
>consulting fees, since it has to be guessed]

Yeah, I've been down that path, too.  In my case, I wanted to turn
compression *on* and absolutely nothing I tried worked.  And you're
absolutely right that the documentation for this particular area is,
shall we say, minimal.

So, compliments of the wonders of open source, I went to the actual code.
Turns out the "value" for any mt "command" (e.g. "fsf") is handled by a
common routine that uses atoi(), i.e. it has to be a number.  This makes
some sense because most of time you're doing things like "mt fsf 5".

Tried "mt compression 1" and everything got happy (and '0' turned it
back off again).

JJ

And here I was ready to believe that the argument to 
mt commands had to confirm to some obscure, machine-independent
"W3C MAGTAPE-XML" specification.* **


The epilogue to my war story is that, according to the manual,
the jaundiced amber tape changer LED on my Spectra Logic 2K wasn't amber
because the unit needed a cleaning tape. The manual didn't mention
that the unit would cease to function, even after power-cycling,
if it needed a cleaning tapes. According to the manual, an amber
led on the tape picker means that the drive isn't ready for use;
they suggest checking SCSI connections and terminations. No joy.
Of course, the connector and terminator were reversed, if you
believe the manual.  For some reason, the cleaning tape worked.

F

PS. The astericks refer to the following scholarly footnotes.

*One day, Motorola will invent a chip that requires XML in its data registers.

**I didn't actually find a reference to "compression off" in
Latvian--might have been German).