According to the gospel as written by Stan Sieler:

> I would hope that all DDS drives have a read-after-write function ...
> but that doesn't remove the desirability of running VSTORE (or something)
> on a *DIFFERENT* drive.  It's vital to validate that your backups
> are readable on different drives, otherwise you could be making backups
> that are readable only by your drive...and if the drive fails, you might
> have a year's worth of useless backups.  (Yes, this has happened!)

<WAR STORY>

This just in...  I tried to restore a 4mm DAT this morning and received
the friendly message:

RESTORE ENCOUNTERED ERROR READING MEDIA DIRECTORY ON LDEV 8
(S/R 1281)

This occurred on a tape that :STORE reported as having completed
*successfully*!!

The session that was using the tape is now hung.  It cannot be aborted
using any method; the :SHOWPROC command shows no processes running for
this session; the I/O for LDEV 8 cannot be aborted.  In short, I will
have to shut down and restart a PRODUCTION SYSTEM (> 1 hour on a 950) in
order to "kill" the I/O on the tape device.

</WAR STORY>

<P*SSED OFF MODE>

After 20 years of working with HP-3000 systems and various iterations of
MPE, I find it unbelievable that STORE/RESTORE is still the weakest link
in the entire OS.  You would think that a data backup routine would be
the most rock-solid part of any OS (unless CSY considers a user's data
to be of little value to the user!).  VSTORE does not always report the
outcome of its operation correctly.  I have seen, on numerous occasions,
files that were "verified" by VSTORE fail to restore.  I have seen, on
numerous occasions, tapes that MPE claimed were stored successfully
become worthless trash when I tried to restore them.  This leads me to
the question:  are any of my backups useable or has STORE/RESTORE been
lying to me all along?  If so, then I have a *SERIOUS* data integrity
problem on my hands that needs to be fixed *QUICKLY*!!

Since we've switched over to an unattended backup several years ago,
we've tried various flavors of 8mm tape drives (there aren't that many
vendors around that will support a 3000/MPE system) and, most recently,
a remanufactured HP 4mm SCSI drive.  My experiences with unattended
backup has convinced me that HP never intended the 3000 to support an
unattended backup scheme.  Rather, they appear to be committed to using
9-track, 6250 bpi roomfulls of tapes and a semi-trained gorilla (you
know -- the one that slams your Samsonite luggage around) to sit and
watch tapes spin and get up out of his chair every 15 - 20 minutes to
change tapes.  Now, I don't intend any slight or slur against any
operator -- the ones we've had in the past were well-motivated,
dedicated, professional individuals who did their jobs conscientiously.
My point is, except for the largest DP shops, who can afford a person on
the payroll to do nothing more than change tapes and pull paper from a
printer??  C'mon, HP, wake up and smell the 90s:  downsizing, personnel
slashing, reduction of labor costs, etc.  And to think that the 90s are
almost over!!

And now this totally ridiculous business of having to completely shut
down and restart a *PRODUCTION SYSTEM* to kill off a hung session.
Folks, Un*x has had the capability of killing an individual process
since its inception many years ago.  Why not MPE??  Why is an OS so
poorly designed that you have to sacrifice the work of ALL users to get
rid of ONE offending session.  It's one degree of inconvenience to have
to reboot a $1,200 personal computer running Trash '95; it's another
world of inconvenience to have to phone 50 users and listen to the
torrent of invective when I tell them, "I have to restart the system
-- you'll be down for at least an hour!" for a system that is supposedly
running a "mission-critical" operating system.  Along these same lines,
I cannot count the number of times I've been told by an HPCE who just
responded to a minor disk drive problem, "You may have to re-install
your system" with about as much alarm as telling me that, "You may have
to make a pot of coffee!"  Re-install my system??  How would you like to
tell your users (especially users facing a deadline of *TODAY*) that
they are going to lose their system for 8 - 10 hours.  (Fortunately, in
defense of the hard-working HPCEs who service our site, they have always
to date been able to make a heroic effort to repair our disk drives
without having to re-install the system.)  Does HP really understand the
terms "mission-critical", "high availability" and "24 x 7" or are they
just thrown on advertising to look impressive?

</P*SSED OFF MODE>

Additional note:  In the 45 minutes it's taken me to compose, proofread
and send this email, my HP system still thinks that there is a session
logged in that "owns" the tape drive.  Jeez!!!

Oh well, I guess that's why God invented weekends!!  I suppose I should
just grin and say a few hehehe's like Art.  Thanks a bunch for letting
me get this off my chest!!

Tom Madigan
Christopher Newport University
1 University Place
Newport News, VA  23606-2998

<DISCLAIMER>

My opinions belong to me and me alone.  The opinions, if any, of my
employer may not necessarily agree with those that I have.  For those
rare instances where they do agree, they agree by accident and not by
design.  In other words, I share these opinions for the purposes of
discussion only; I am speaking for myself only and not in any official
capacity whatsoever.  So there!!

</DISCLAIMER>