[Info-vax] TCPIP$SMTP SYSTEM-F-EXQUOTA, SYSTEM-F-ACCVIO after upgrade to V8.3

[Steven Schweda]

Richard B. Gilbert wrote:

> >    Note: "new".

> "New" is no guarantee that it works!

   Well, duh, again.  Did I saw "new", or "new and untested"?

>   The better vendors test the drives
> before they leave the factory and pack them sufficiently well that you
> might be able to load the box into a cannon of suitable size and fire it
> without damaging the drive.  Still, shit happens!  Last time I looked,
> disks were sold with a warranty so you may be able to get a replacement
>   or your money back!

   On my budget, "better vendors" are out of reach, and my
sources (Ebay) are more than one step from the factory.  When
I buy a disk drive, I usually get one claimed to be "new", and
then I normally run it through four passes of the "analyze"
function in the "format" program on a Sun Solaris system.
This typically wastes several hours, but on rare occasion, it
finds perhaps one new bad block.  If it does, I'll probably
run it for another eight passes.  Then the drive either goes
directly into service, or else sits around for months or
years, waiting for a failure or some new demand.  The old
drive in this case was in use for only about two years, which
may be the shortest useful life I've seen on a Seagate drive
(even among the junk I buy).

   On the bright side, the old disk was ODS2, and can be
traced back to an old VMS version, so, with its 36GB size, it
had a cluster size of 69 blocks.  When I initialized the new
one, I decided that the new default (8 blocks) might be an
improvement.  With some help from BACKUP /TRUNCATE, I figure
that that change accounts for most of the difference between
17% free (old) and 43% free (new).