Duty-cycle on mini's (was Re: Compiler patches ...)

Garance A Drosihn drosih at rpi.edu
Sun Jul 10 23:27:44 GMT 2005


At 1:00 AM -0700 7/10/05, Lyndon Nerenberg wrote:
>>>Are they really designed to run at full bore at the time?
>>
>>You'd hope so, and that the fans blowing full speed would allow
>>this to be the case. I've not had this happen on mine, so perhaps
>>it's a ventilation issue ?
>
>I can't speak to "designed" for flat-out operation, but my mini
>runs 24x7 without complaint.

The background for some of this discussion is that my Mac-mini died
after running FreeBSD for about five months.  During that time, the
machine was up 24x7 except for a few days in February and March.
I'll guess that 98% of the time that it was running, it was running
FreeBSD/PPC.  The other 2% was running MacOS 10.

The symptom I saw was that the mini would start up okay, and then
after some amount of time it acted like the hard disk had been
disconnected.  Needless to say, the machine couldn't do much after
the hard disk disappeared.  If I just-rebooted the machine at that
point, then even open-firmware could not find the hard disk.  If I
powered it down and waited an hour or so, then it would boot up fine.

Initially the problem only happened on FreeBSD/PPC, but within a
week it was also happening quite regularly on MacOS 10.

So, several people (including me) wonder if maybe FreeBSD was
doing something wrong with the fans.  Either running them too much,
or not enough, or something else.  Two important notes on this
failure:
    1) RPI has spells where the air-conditioning to offices in the
computer center doesn't work well.  At the time of the failure, my
office had been between 80 and 85 Fahrenheit for a few days straight.
(this is partially because I have eight computers, two UPS's, and a
refrigerator in here...).
    2) When I bought it in to be fixed, the guy at the Apple store
said he had seen a few similar failures -- and none of those were
running FreeBSD.  But he hadn't seen all *that* many failures.  I
had the impression they had seen between five and ten.

Whatever the problem was, Apple was able to fix it without replacing
the hard drive or the motherboard.  The mac-mini has been working
fine for me since then.  I have the vague feeling that the fans seem
to run a bit more now than they used to, but I can't say that I was
ever really paying all that much attention to them.

So, I wouldn't read too much into the failure of this one Mac-mini,
but it would be interesting to hear if anyone else has a similar
problem.

-- 
Garance Alistair Drosehn            =   gad at gilead.netel.rpi.edu
Senior Systems Programmer           or  gad at freebsd.org
Rensselaer Polytechnic Institute    or  drosih at rpi.edu


More information about the freebsd-ppc mailing list