Damaged Tapes

[Martin Golding]

In <1KL&MK$@uzi-9mm.fulcrum.bt.co.uk> igb at fulcrum.bt.co.uk (Ian G Batten) writes:

>I have a friend (really!) who has done a very careless thing.  He
>inserted a QIC-11 tape which had some vital data on it which was write
>enabled into a tape drive and wrote a few kilobytes over the beginning
>of it.  Given that I have kernel and driver sources for V.3 machines and
>various other equipment scattered about, can anyone suggest anything
>which might drag some of the data off?  I've tried the obvious method of
>reading the tape to the end of media and then trying again using a
>non-rewinding open, but I simply get read failures.  I assume the
>problem is that I have a partial block following the double tape mark.

It's worse than that. Each block is numbered, the formatter (the part
of the drive that diddles the data) puts a header on each block with
all kinds of interesting stuff. The same hardware knows better than
to let you start reading in the middle of the tape, or read past the
eom marker block. Without direct access to the electronics, you're not 
likely to get anywhere. In addition, most of the serpentine drives use
erase bars; so you may have a data gap at the beginning or end of each
track.

On the other hand, it may be worth your time to contact the manufacturers
directly. They'll know exactly how the firmware works, and might be able
to tell you how to cheat it. ( I wrote the firmware for an obscure
cassette drive, and I _think_ I could trick the drive into reading
past the eom).

OB scolding: Don't your friend's tapes have write protect :-/ ?

Martin Golding    | sync, sync, sync, sank ... sunk:
Dod #0236         |  He who steals my code steals trash.
Blechtrottel CPU with many megs...
{mcspdx,pdxgate}!adpplz!martin or martin at adpplz.uucp