Should we reuse mag tapes?

[falk at uxc.cso.uiuc.edu.UUCP]

In article <1058 at megatest.UUCP> jacobo at megatest.UUCP (Jacobo Bulaevsky) writes:
>
>   The question is:  Is it safe to reuse mag tapes  for  important  data  or
>   should we use brand new tapes when storing critical information?
>
>   Would someone have a strong feeling, based on experience or data, one way
>   or  the other?  I know that the whole idea of using mag tapes is that you
>   can reuse them a good number of times,  but  I  have  also  heard  people
>   saying that valuable info should go on virgin tapes.
>   
In my experience, brand-new tapes tend to have *more* problems than "slightly
used" ones. At a former job, I ended up doing a rather large "study" of this
phenomenon after we began experiencing problems with recovering backups from
once-written tapes. One of the things that I found was that brand new tapes
tend to be dirtier than slightly-used ones (excess iron-oxide that came off
in the tape drives), so I instituted a policy of cleaning each new tape before
use. That eliminated some of the potential failures, but not all. By trial
and error it was noted that tapes that had been written and read at least once
before had a better "recovery rate" than once written tapes- not sure why
except perhaps that you are weeding out the bad apples at that point. If you
think about it, would *you* want to trust valuable data to an untested medium?

As far as maximum usage, theoretically, you ought to be able to reuse tapes into
the hundreds or thousands of times. Improper handling, poorly adjusted tape
drives, environmental disturbances, etc., guarantee that number to be reduced
significantly, probably down to 50-100. While we were going thru this mess,
before we isolated the problem, I had some tape manufacturer reps come down
and they told me that they do verification testing of some tapes from each
batch by writing and reading tapes n-times until they fail, hence, the above
numbers.

         Connie Falk

UUCP:	 {ihnp4,pur-ee,convex}!uiucdcs!uiucuxc!falk
ARPANET: falk%uiucuxc at a.cs.uiuc.edu	 CSNET:	 falk%uiucuxc at uiuc.csnet
ICBM:	 40 07 N / 88 13 W
US Mail: Univ of Illinois, CSO, 1304 W Springfield Ave, Urbana, IL  61801
AT&T:	 217-333-8050