On Thursday 23 October 2003 05:48, Tony wrote:
>Hi all,
>
>I am still having trouble trying to label tapes and Amanda
>recognising them.
>
>I label a tape:
>
>bash-2.05a$ /usr/sbin/amlabel daily daily01fri
>rewinding, reading labelamlabel: strange amanda header: "AMANDA:
>T"
>, not an amanda tape
>rewinding, writing label daily01fri, checking label, done.
>
>I then check the tape:
>
>bash-2.05a$ dd bs=32k if=/dev/st0
>AMANDA: TAPESTART DATE X TAPE daily01fri
>
>1+0 records in
>1+0 records out
>
>So I then run amcheck:
>
>bash-2.05a$ /usr/sbin/amcheck daily
>Amanda Tape Server Host Check
>-----------------------------
>Holding disk /tmp: 521540 KB disk space available, that's plenty
>amcheck-server: strange amanda header: "AMANDA: T"
>ERROR: /dev/nst0: not an amanda tape
> (expecting tape daily01tue or a new tape)
>
>
>If I continually run amcheck, about maybe 5% of the time I will
>get the response "Tape daily01fri label ok", to indicate that it
>was actually able to read the label on the tape. This is VERY
>strange !
>
>I have downloaded and installed 2.4.4p1 and it is a RH7.3
>machine.
>
>As an aside, I did manage to work out the stinit stuff and the
>drive is no longer using hardware compression. I did another
>tapetype and got the following:
>
>define tapetype dds4 {
> comment "produced by tapetype prog (HW compresssion off)"
> length 19503 mbytes
> filemark 0 kbytes
> speed 2693 kps
>}
>
>which is a lot more like what it should be.
>
>
>Any suggestions on why it is not recognising the label ?
>
The only idea I keep coming up with is that for DDS tapes, the status
of the tapes compression is recorded in a hidden header on the tape.
Simply turning off the compression and relabeling the tape will turn
it back in in the initial label read, so you wind up with the
compression on regardless. The workaround I've used is to read out
the label with dd and save it to a scratch file. Rewind the tape and
issue the compression off commands, then dd the label file back to
the tape useing the non-rewinding device. This will not normally
move the tape, or flush the buffers as its not a big enough write,
and you'll only succeed if you force the drive to flush its buffers
which will then record the compression off state in the header. So
one must, in addition to dd'ing the label back, follow that up with
another dd from /dev/zero thats big enough to force the buffer flush,
use at least a meg of /dev/zero more than the drives rated buffer
size.
I have a script around here someplace, and which has also been posted
to this list on 2-3 occasions, that does all this in one swell foop
per tape MOST of the time.
>Thanks,
>Tony.
>
>
>____________________________________________________________________
>____ Want to chat instantly with your online friends? Get the FREE
> Yahoo! Messenger http://mail.messenger.yahoo.co.uk
--
Cheers, Gene
AMD K6-III@500mhz 320M
Athlon1600XP@1400mhz 512M
99.27% setiathome rank, not too shabby for a WV hillbilly
Yahoo.com attornies please note, additions to this message
by Gene Heskett are:
Copyright 2003 by Maurice Eugene Heskett, all rights reserved.
|
