[Bioclusters] How do people deal with temperature excursions (A/C failures)?

[George White]

Our lab has a homebrew ROCKS cluster in small computer room that relies on
a single air conditioner. My experience has been that the poweroff(8)
command can't be trusted, so you often end up having to turn off the power
manually.  We had hardware damage when the A/C tripped the circuit breaker
recently.  Now we want to add another cluster and a file server so I'm
anxious to have a reliable way to shut systems down when they get too
warm.  How do other labs deal with this problem?

--
George White <aa056 at chebucto.ns.ca> <gnw3 at acm.org>
189 Parklea Dr., Head of St. Margarets Bay, Nova Scotia  B3Z 2G6