I have heard many stories like this over the years, but they almost always
occurred with DDS-2 drives.  Did your misadventures include DDS-2 or DDS-3?

Denys
-----Original Message-----
From: HP-3000 Systems Discussion [mailto:[log in to unmask]] On Behalf
Of Greg Stigers
Sent: Wednesday, October 26, 2005 8:49 AM
To: [log in to unmask]
Subject: Re: [HP3000-L] DDS back online after :STORE

At this point, should this thread be changed to DDS horror stories? I have 
long been a fan of moving the tape to another drive, and VSTORing on that 
other drive. This can help catch head drift or other forms of drive failure.

In my previous shop (I was given the opportunity to move on last August, 
speaking euphemistically), in just a few months, VSTORing maybe twice a 
week, we had two or three failures that required some investigation, and had

a tape drive replaced once or twice in that time. Come to think of it, we 
had an outright drive failure or two before that, where STORE failed.

Of course, VSTORE will not prove with certainty that your tapes are good and

that you could perform a full restore. See DDS horror stories for that and 
more. The truly ambitious can restore everything elsewhere if you have that 
kind of disc space, or restore to NULL, to confirm that another drive thinks

it can read the whole tape. This will of course consume some amount of 
resources; it was not an issue on our N-class. And IIRC, even that doesn't 
prove that what one gets is what one had. I think the DDS horror stories 
thread(s) recounted a failure with an IMAGE database that appeared to 
successfully restore, but could not be used.

One argument against doing any of this is that the drives are rated for some

duty cycle, and this can exceed it. It will probably wear out the drives 
that much faster. I never thought that was a very good argument. My 
reasoning is that should I be unable to restore data for some reason, I 
would rather be able to say that I took all reasonable precautions to make 
sure that I had good backups, than to have to explain why I could have 
detected the ultimately fatal flaw, but didn't.

Funny thing is, I think that these concerns apply to any backups on any 
platform.

Greg Stigers 

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *