Re: Old Magnetic Tapes
From: John Laird (nospam_at_laird-towers.org.uk)
Date: 12/17/03
- Next message: Paul Repacholi: "Re: Old Magnetic Tapes"
- Previous message: Bob Koehler: "Re: BOOT, HALT, RESTART, 1, 2, 3"
- In reply to: Alan Adams: "Re: Old Magnetic Tapes"
- Next in thread: Paul Repacholi: "Re: Old Magnetic Tapes"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Date: Wed, 17 Dec 2003 18:08:19 +0000
On Wed, 17 Dec 2003 17:10:37 GMT, Alan Adams
<alan.adams@orchard-way.freeserve.co.uk> wrote:
>The last three months of my previous employment were spent attempting to
>recover the data from our archive tapes. They were 6250bpi, written in the
>main on TA78s, with two usable copies of each tape. (The third copy had been
>stored in an unheated Portakabin, and I didn't want to use them.)
>
>I was using Saveset Manager to copy to disk, then copied to DLT using the
>same software.
>
>I found that the first pass would read about half way through the tape, then
>the oxide build-up made the tape unreadable. A repeat run would go further,
>as the loose oxide had now been removed from the first half. With three
>passes, most of the data was recovered. The tapes were mostly 12 to 15 years
>old.
>
>All data from all passes was kept, and merged afterwards. This was because
>files successfully read on pass one could become unreadable on pass 2.
>
>I also found that the tape usually stuck solidly to the head just after the
>last file on the tape. This is because that part of the tape had never been
>written. The rest had had, generally one, pass of the head when it was
>created, and this removed a lot of loose oxide. It was normal to have to
>clean the heads after writing a new tape.
>
>I did all this recovery on a TSZ(05?) drive, and found it useful to disable the
>lid sensor, so I could open the lid after each failure, then clean the heads
>and continue from where the failure occurred, effectively making the
>operation single pass.
I had a similar experience about a week before I left one company, as they
were relocating offices. Someone, in tidying their pod, came upon a box of
seismic data tapes which they "had always meant to get loaded". And could I
manage it ? By this time, we had about 3/4 of a cluster relocated already,
and hardly any available disk space, cpu, or access to the larger machines
with Thruway-accessible tape drives. Moaning over lunch to a colleague, he
let slip someone had "found" a table-top tri-density HP SCSI 9-track drive
not in use. Did we want it ? Did we - I had two staff permanently loading
data via the error-prone operator "assisted" networked drives, and for two
years some *git* had exactly what we wanted in a cupboard...
Anyway, it was the absolute dog's wotsits, this HP unit. You had full
access to the tape path, heads, reels, any bad bits you could shunt the tape
back or forwards, give the head a quick clean, back on-line, issue a skip
command, try again. Fantastic piece of kit. Got through a bucket-load of
cleaning swabs and some of the tapes might never have been read again, but
we recovered a huge amount of data.
-- Coffin for sale. Lifetime guarantee. Mail john rather than nospam...
- Next message: Paul Repacholi: "Re: Old Magnetic Tapes"
- Previous message: Bob Koehler: "Re: BOOT, HALT, RESTART, 1, 2, 3"
- In reply to: Alan Adams: "Re: Old Magnetic Tapes"
- Next in thread: Paul Repacholi: "Re: Old Magnetic Tapes"
- Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
Relevant Pages
|
|
