In article Tsn6h.1215$8u1.207@trndny04, ehsjr wrote:
[...]
I'll give you five examples:
1) Peat Marwick Mitchell datacenter, early 70's
An airconditioner failure caused DASD (2314) data
errors at exactly 94 degrees on their wall thermometer.
Ran fine at 93.

Reminds me of an incident that occurred in the late 80s/early 90s when I
worked for the Navy. I managed a Tandem TXP system that shared a computer room
with a Honeywell 66. One holiday weekend, the air conditioning system failed
in the wee hours of Saturday morning after the second shift operators had gone
home. (There was no third shift.) Monday being a holiday, the problem wasn't
discovered until the first shift operators arrived at about 6am Tuesday to
find the data center at about 110 degrees. The Honeywell had gone down only
about three hours after the air conditioning did... but the Tandem was still
up. The DASD cabinets were painfully hot to the touch, and one of the drives
had gone down -- but since Tandem uses mirrored drives, and the mirror was
still ok, it did no harm. I measured the exhaust air at the back of the
processor cabinet at 134 degrees... but the Tandem was still up.

--
Regards,
Doug Miller (alphageek at milmac dot com)

It's time to throw all their damned tea in the harbor again.