28 Jul
2004
28 Jul
'04
9:16 a.m.
Teo Zenios wrote:
----- Original Message -----
From: "Doc Shipley" <doc(a)mdrconsult.com>
> My department had always followed those
guidelines faithfully.
>Still, soon after I took over the subnet, the very first time they were
>asked by the state to provide archived files we were unable to produce.
> All the *tapes* had been stored properly, but all the machines with
>compatible *drives* had been surplussed long since, and the backup
>software junked. It was my first experience in data recovery and
>migration, and one that I've since learned is very common. Perversely,
>I seem to have developed a taste for it. :)
You would think the drives making the backups and the
original software
would have been removed and placed in the vault with the media. Did some of
the older tape drives have problems reading tapes from different machines
because of head alignment?
No problems with any of the 200-300 tapes I migrated, over two years'
time. And yes, you would think, unless you stop to think that the IT
for that department had been under the management of an archeology
student who "knew a lot about computers". They just never equated the
tapes in storage with the hardware going out the door.
That scenario is a *lot* more common than it ought to be.
Another one is that of an architect for whom I built a file server
and backup server a few years ago. He had an Iomega 800MB parallel port
tape drive (Travan 1, were they?) and insisted on using it, and the
Iomega software, for the whole office backups.
I begged him through 3 years and a couple of major upgrades to spend
a few hundred dollars on a good refurbed SCSI tape drive. DDS2 or an
Exabyte 8505 would have done him just fine. He couldn't afford "all
that money to buy something I already have". He also never had the
money to pay me to test the DR plan or backups. He had to get back a
trashed or deleted drawing now and then, and considered that
verification of his backups.
About 2 years ago he called and he had crashed a disk, and he was
having trouble getting his files back off of tape. It turned out that
he'd been cleaning the tape drive as instructed, and rotating tapes on
schedule. Rotating the same 10 travan tapes for 4.5 years.
There were places on those tapes you could see through. We didn't
get any data back that was less than 6 months old. It nearly put him
out of business.
Doc