At 08:11 PM 10/12/96 -0500, Sean Donelan wrote:
>
>Stuff happens. No one can predict or plan for everything. But you can
>have procedures in place to combat the biggest problem in diasasters, poor
>communication. Partial knowledge about the real situation happens when
>the real information isn't made available. Techies sometimes get too
>wrapped up in the hardware, redudant fiber, generators, bomb-resistant
>shelters; and forget about keeping people informed.
>
Here's what the latest issue of RISKS Digest [18.52] had to say.
- paul
[snip]
----------------------------------------------------------------------
Date: Sat, 12 Oct 1996 13:48:41 -0700
From: "Peter G. Neumann" <Neumann(a)csl.sri.com>
Subject: Rats take down Stanford power and Silicon Valley Internet service
Two rats crawled through an underground cable conduit into a cabinet of
power switching gear adjacent to the Stanford University cogeneration plant,
and caused an explosion that cut off power to the Stanford area beginning
around 7:30pm on Thursday evening, 10 Oct 1996, and continuing until 3:30pm
Friday afternoon. The BBN Planet hub (Internet Point of Presence, or PoP)
at the Stanford University Data Center remained in operation for a few hours
on standby battery power, but then gave out around 9pm Thursday; it came
back up around 4:30pm, an hour after Stanford restored power. To name just
a few, Bay-Area BARNet users at Stanford, U.C. Berkeley, Apple, Sun,
Hewlett-Packard, Lawrence Livermore (partially), and SRI were cut off from
the Internet. The *Los Angeles Times* and *San Francisco Chronicle* on-line
sites were also off the air. Because I had no Internet access yesterday, I
held up RISKS-18.52 -- thus enabling me to include this item adding to our
RISKS archives collection of rodent-induced outages. (Long-time readers
recall that SRI alone has contributed four fresh-fried squirrels resulting
in power outages.) [Sources: On-line messages and a front-page *San
Francisco Chronicle*, 12 Oct 1996 item]
Evidently, the horse is out of the BARNet, and the rats found the weak lynx.
They sure put a ro-dent in the day for many BayAreans. Perhaps your mouse
will click on a tale of its own. At any rate, this is just one more saga in
the weak-link-in-the-infrastructure department. But I'm surprised that
power-system technology has not found a way to develop rodent-tolerant
circuits.
[With SysAdmins and others pacing the halls at SRI waiting for whatever,
Doug Moran remarked that keeping around a few fresh-frozen electrofried
rodents is allegedly standard practice for purveyors of power; it is then
very easy to have a fallback alibi when no other cause can be found.]
------------------------------
[snip]
