3b2's and elec. codes

[/violet_g/dean]

In article <3800010 at snail> carroll at snail.CS.UIUC.EDU writes:
>Doesn't anybody care about file system integrity any more?  It is exceedingly
>rare that power must be cut off without even time for a sync.  I can't even
>think of such a case off hand.

This discussion brings to mind an interesting "feature" on the disk
drive cabinet in the Data General MV series.  There is a rocker switch
for the cabinet's power on its top front edge (right where you might
bump it accidentally).  But that's OK (one thinks) because there is an
interlock switch on the CPU cabinet that apparently disables the drive
cabinet's power switch, so it can't be accidentally shut off during
operation.  Does it do that?  No.  Why not?  The word we got was that
DG originally designed it that way, but then ran afoul of an electrical
code regulation stating that EACH cabinet must have a power switch that
will allow it to be easily and instantly turned off in the event of an
emergency.

In our lab that meant that we had to build little plexiglass protectors
to keep the switches safe.  Sigh...  (which didn't help when a ladder
fell on the drive cabinet, but that's another tale).

-Dean (dean at violet.berkeley.edu)
  (but the story comes from my days as uw-nsr!dean at uw-beaver.arpa)