> > Check out vmphyinv; I've used it to inventory large numbers
> Is there a way to run it non-interactively, but where it
> does *not* update the database? The -non_interactive
> appears to apply no matter what. Until I get familiar
> with it, I'm not ready to trust it.
No data--that wasn't a consideration when I've used it. There are maybe
ten pages on vmphyinv in the 6.0 Media Manager docs (not that I RTFM,
either!).
> I have a tape in my robot that has (examples here) EVSN 5
> and RVSN 6. My records show I've sent tape 6 (obviously
> EVSN 6) offsite. It may or may not have valid data. I'll
> have to recall it to check.
>
> With vmphyinv, is it going to try to remap RVSN 6 to barcode
> 5? I'm not sure I want it to do that (in case the EVSN volume
> 6 has RVSN 6 on it, and it contains other valid data).
Since I don't know what happens automatically without reading up and
testing, I'd probably fall back on
o I have had tapes with different internal and external SNs and
NetBackup handles that. Or at least my memory says so, from two
situations: 1) damaged barcode on an existing volume--seems that
there's a technote on this, but that would be a reason to use vmchange
-barcode (below). 2) importing a tape that has barcode 1 and recorded
SN 2. (If I've dreamed this up, someone please correct me.)
o vmchange -barcode will let you assign a different barcode to a tape
I might be overlooking some operational aspect, but I believe you can
make changes to the E/RVSN in volmgr so that you can make it work with
the mismatches you now have--at least long enough to, as you proposed,
dup your data and relabel the problem tapes.
|