|
Home > Archive > Data Storage > January 2006 > how to find out the "lifetime" of DD5-tapes? NOT MTBF..
You are viewing an archived Text-only version of the thread.
To view this thread in it's original format and/or if you want to reply to
this thread please [click here]
| Author |
how to find out the "lifetime" of DD5-tapes? NOT MTBF..
|
|
| hardware55@bluemail.ch 2005-12-30, 7:47 am |
| Hello,
some advanced backup-solutions do allow to define the
"tapelifetime" in order to be able to use/take a new tape
from the tape-recycling pool before the tape in questions
really gets broken(e.g. too many read/write-errors).
It's quite easy to find out the MTBF of different tapedrive-
types but not the lifetime of the tape/media itself..
(of course that's not the case with high-end drives like LTO
which claims 10'000 load/unload cycles and 20'000 short
section read passes).
Any tip is appreciated very much. Thank you!
John
Can somebody point me to such
load
| |
| Paul Rubin 2005-12-30, 7:47 am |
| hardware55@bluemail.ch writes:
> some advanced backup-solutions do allow to define the
> "tapelifetime" in order to be able to use/take a new tape
> from the tape-recycling pool before the tape in questions
> really gets broken(e.g. too many read/write-errors).
Some drives let you read the error count registers but the commands
are drive specific. If you're asking how many times you can re-use
DDS tapes, the answer is: not very many.
| |
| Rob Turk 2005-12-31, 2:47 am |
| "Paul Rubin" <http://phr.cx@NOSPAM.invalid> wrote in message
news:7xhd8q6ejc.fsf@ruckus.brouhaha.com...
> hardware55@bluemail.ch writes:
>
> Some drives let you read the error count registers but the commands
> are drive specific.
The problem isn't so much to get the error counters (Log sense page 2 and 3
on almost all tape drives), but what conclusions to draw from them. If you
write 10GB to tape and the Total Corrected error counter is at 1000, is that
good or bad?? These are good for trend analysis when using known references,
but it's very hard to claim that a particular drive or tape is bad just
because it clocked a certain number of soft errors. Vendors will not provide
hard numbers of <x> % error rate as a cut-off point when you need to throw
the tape away.
Rob
| |
| Al Dykes 2005-12-31, 7:46 am |
| In article <I5rtf.3008$zc1.1352@amstwist00>,
Rob Turk <_wipe_me_r.turk@chello.nl> wrote:
>"Paul Rubin" <http://phr.cx@NOSPAM.invalid> wrote in message
>news:7xhd8q6ejc.fsf@ruckus.brouhaha.com...
>
>The problem isn't so much to get the error counters (Log sense page 2 and 3
>on almost all tape drives), but what conclusions to draw from them. If you
>write 10GB to tape and the Total Corrected error counter is at 1000, is that
>good or bad?? These are good for trend analysis when using known references,
>but it's very hard to claim that a particular drive or tape is bad just
>because it clocked a certain number of soft errors. Vendors will not provide
>hard numbers of <x> % error rate as a cut-off point when you need to throw
>the tape away.
>
>Rob
>
>
Agreed. When I ran a pool of tapes I made a practice of rotating the
tapes with error rates in the top few percent into non-critical
uses. Since most tapes had near-zero errors and a few had a distinctly
higher rate I figured I was pruning out the lemons.
It's not like I was throwing out those tapes. It wasn't costing me
any money.
--
a d y k e s @ p a n i x . c o m
Don't blame me. I voted for Gore.
| |
| David A.Lethe 2006-01-04, 2:53 am |
| On Sat, 31 Dec 2005 09:14:36 +0100, "Rob Turk"
<_wipe_me_r.turk@chello.nl> wrote:
>"Paul Rubin" <http://phr.cx@NOSPAM.invalid> wrote in message
>news:7xhd8q6ejc.fsf@ruckus.brouhaha.com...
>
>The problem isn't so much to get the error counters (Log sense page 2 and 3
>on almost all tape drives), but what conclusions to draw from them. If you
>write 10GB to tape and the Total Corrected error counter is at 1000, is that
>good or bad?? These are good for trend analysis when using known references,
>but it's very hard to claim that a particular drive or tape is bad just
>because it clocked a certain number of soft errors. Vendors will not provide
>hard numbers of <x> % error rate as a cut-off point when you need to throw
>the tape away.
>
>Rob
>
If your tape drive supports the TapeAlert feature, then one of the
defined functions in the spec will report not only when the tape needs
to be replaced, but also when the tape is nearing end-of-life.
There are a few caveats, like having a tape drive that uses this
optional parameter in its implementation of TapeAlert, and having some
software (such as ours -- shameless plug ;) that decodes and reports
it, but you can certainly write your own application.
To see if the TapeAlert mechanism an the media-related warnings are
impmlemented for your particular drive, then consult the
documentation.
David Lethe
http://www.santools.com
|
|
|
|
|