Can anyone enlighten me as to what the thinking was as to why someone might
have designed the "Service / Normal" switch onto the back of a HP3000 928
CPU? What purpose does it serve except to totally screw things up?

We always had ours set to "Service". It has been that way for so long, I
forgot it even existed or what it did. Somehow it got flipped to the
"Normal" position in the last few weeks.

We do most of our systems administration remotely. The physical computer is
way off in another building. So last night I went to do a shutdown and
reboot "START NORECOVERY" because we were having a problem with some IO
configurations which refused to completely apply themselves through
IOCONFIG.

So, with this switch mistakenly in the "Normal" position, I can happily do a
"ctrl-A" then SHUTDOWN. So the computer is now shutdown. But wonders of
wonders with the switch in the "Normal" position, I can not do a "Ctrl-B". I
can not get the boot screen up on the console. I can't get the system back
up remotely any way I know.

Total panic time. Did I mention this is a critical production computer?
Every extra minute it is down costs the mill lots of money. Every extra
minute beyond what I have scheduled has the potential of messing up paper
production.

So I charge off to the building with the computer room. Since I don't
remember the *#!*&%#@ Service/Normal switch I power down the 928 and power
it back up.

More minutes of major panic since the Service/Normal switch in normal
position appears to prevent the display of the CPU self test on the bottom
of the console from appearing. It looks at first as if power cycling the CPU
is still not causing it to restart.

Finally the boot menu appears. With heart pounding, blood pressure ready to
blow, I reboot the system.

So why the $#@!^* would anyone ever want this stupid switch in the "Normal"
position? Maybe if it prevented both "Ctrl-A" shutdown and "Ctrl-B" boot. I
would make sense as some sort of security or preventing an unintentional
shutdown. But by only preventing a reboot, but still allowing a shutdown, it
only means one can screw things up. I don't get it.

Timothy Atwood
Holtenwood Computing
http://www.holtenwood.bc.ca/computing/
for Domtar Vancouver Mill
(Opinions expressed are mine and do not reflect Domtar)

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