🚀 go-pugleaf

In csu.cs Michael Dwyer <mdwyer@holly.ColoState.EDU> wrote:

: Their legitimacy is apparently still in debate.  I don't think I would
: call it a game, however...

What I am calling a game is not the USE of the processor, but the rude
attitude that "Hey, they're my machines; I paid for 'em; so I can leave a
12 day batch job running at a nice level of -10 (note this is Linux nice
levels) which will screw up everyone else's work in the process and it's
OK because..." I can't even mock that attitude; I don't know the
reasoning. The point is, they designed a task scheduler for a reason;
let's use it ;) 

There's a mighty difference between CPU usage at 3 am and CPU usage at 3
pm.

: The root-wielding sysadmin is a busy man who cannot possibly monitor
: every single CS machine in the lab.  Instead, it would appear that
: students take initiative and simply power-cycle the machines. (Ouch!)

Perhaps, but we have it down to a science over here; perl scripts grep out
processes at more than 60% of the CPU running for more than 5 minutes
and could do anything: renice them, notify an admin, whatever.

Of course, maybe at college, they think you guys are responsible enough to
do that on your own (yeah right! :-P )

Later,
Paul
  ------------------------------------------------------------------------
  J. Paul Reed                        preed@verinet.com || paul@619pro.com
  Well, you know, actually, there's nothing I can think of that I'd rather
  do, than drop napalm for a living.   --Chuck Norstadt,The Man w/o a Face
     Geek Code and various other frivolities at www.verinet.com/~preed