Scott Gavin wrote:
 
<snip>
 
>Unfortunately, the number of disk failures that cause the loss of
>data may *double* from the customer's point of view (it could be
>worse than this because I hear about more controller failures than
>mech failures usually).
>
>So HP actually sees *fewer* drive failuers, whereas the customer
>perceives *more* (catastrophic) drive failures.  A quite interesting
>quality problem. :-)
>
>G.
 
Scott makes a great point. The customer's point of view is critical
as his/her confidence will have impact on future purchases.
 
I recently had my confidence shatered by the loss of four 2GB drives
on HP-UX systems within a two week period of time. A bad power pack
caused the loss of two drives on one system.
 
On the other system we lost the root drive and had to reinstall the
system. On the way up, we then lost one of the data drives. The
technical details are not available to me.
 
Even though these drives were physically replaced, I don't know
how HP officially accounts for these but our management perceives
this as a failure of _disk_ hardware.
 
Please forgive me for introducing HP-UX but I believe that the
same hardware applies on both sides of the "fence". I have been
working with MPE systems since 1979 and have been generally
impressed with the reliability of HP hardware (after easily
being converted from the IBM mainframe environment).
 
These latest incidents is reminiscent of the "night"-mares that
we had with the old HP7925 drives which were litteraly scrapped
as soon as more reliable hardware was available.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> John Schmid         < My opinions > voice: 612-733-9906     <
> Lead DBA            < do not not  >   fax: 612-736-7757     <
> 3M Company          < necessarily > email: [log in to unmask] <
> 3M Center 224-3S-25 < represent   >                         <
> St. Paul, MN 55144  < those of 3M >                         <
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~