Date: Wed, 5 Nov 2003 11:22:35 -0500 From: "J. Seth Henry" <jshamlet@comcast.net> To: freebsd-questions@freebsd.org Subject: Re: freebsd-questions Digest, Vol 33, Issue 6 Message-ID: <200311051122.35429.jshamlet@comcast.net> In-Reply-To: <20031104184220.5BB1916A4E0@hub.freebsd.org> References: <20031104184220.5BB1916A4E0@hub.freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
Actually, I have noted this same issue, and both points are correct.
I have a Compaq IA-1 internet terminal which I converted into an X terminal=
=2E=20
The hardware (was) unmodified, and ran WinCE with no active cooling at all.=
=20
The little machine was perfectly stable, and in fact was designed to never=
=20
completely power down - but instead enter a sleep state when the power butt=
on=20
was pressed.
I then ran Midori Linux on the system (my first attempt at an X terminal), =
but=20
the X server and mouse driver had some serious issues. Nevertheless, the=20
hardware didn't lock up or crash.
I then loaded FreeBSD 4.8-REL on the box, and it started locking up right a=
nd=20
left. Eventually, I added a cooling fan/ heatsink to the AMD K6-2 CPU, and=
=20
the lockups were greatly reduced. (they still occur, but only under duress)=
=2E=20
Keep in mind, all I changed was the OS - the bus clock and multiplier didn'=
t=20
change. Eventually, I installed a K6-III+ mobile processor, and now I very=
=20
rarely get lockups - but it does run noticeably warmer.
This isn't just on AMD hardware, either. I have a dual PIII server that=20
suffers the same problem. It runs at least 4-5 degC cooler under RedHat Lin=
ux=20
(doing the same chores) than it does under 4.8-REL. However, it is a much=20
better built system than the IA-1, and doesn't crash, although the cooling =
is=20
so loud that I've been tempted to put Linux back on it just to get rid of=20
some of the fans.
The fact is, FreeBSD, for some reason, causes hardware to run hotter. Perha=
ps=20
it is a difference in the idle routine, perhaps it is more "active" about=20
checking hardware - I'm no kernel expert. However, this *IS* an issue. For=
=20
those who don't believe the OS can drive the power requirements of a system=
,=20
think again. I installed a SmartUPS on my network, and monitored the load o=
n=20
the power supply. Yep - it increased running FreeBSD versus Windows2k.
However, I feel that FreeBSD is, overall, a superior operating system. Ther=
e=20
is no way I would go back to the hell that is linux - much less Win2k. As=20
such, I just do a little more homework when buying hardware.
Regards,
Seth Henry
On Tuesday 04 November 2003 13:42, freebsd-questions-request@freebsd.org=20
wrote:
> Date: Tue, 4 Nov 2003 09:45:51 -0500
> From: Paul Mather <paul@gromit.dlib.vt.edu>
> Subject: Re: Overheating attributed to Freebsd --sysctl variables
> =A0=A0=A0=A0=A0=A0=A0=A0notavailable--
> To: freebsd-questions@freebsd.org
> Message-ID: <20031104144551.GA55894@gromit.dlib.vt.edu>
> Content-Type: text/plain; charset=3Diso-8859-1
>
> On Mon, 3 Nov 2003 21:07:45 -0700 (MST), Technical Director
> <trodat@ultratrends.com> wrote:
>
> =3D> Forgive me for saying:
> =3D>
> =3D> If this system is borked with FreeBSD due to the cpu's not cycling
> =3D> 'down', then use a different operating system. FreeBSD is not
> responsible =3D> for your trouble if you can solve the problem by moving =
on.
> Doing so and =3D> solving the problem is more important than holding the =
OS
> and the =3D> contributors to it accountable to something so seemingly far
> fetched. =3D>
> =3D> One way to test overall integrity of your hardware is to boot to bios
> and =3D> leave it. Does it bake out on you? Then there is definitely
> something =3D> wrong with your hardware, perhaps a fan is spinning less r=
pms
> than when =3D> new.
> =3D>
> =3D> In my humble opinion this is probably not associated with the OS, bu=
t,
> =3D> that doesn't solve 'your' problem. So besides seeing it for myself I
> can't =3D> see an absolute need to use FreeBSD, in your words the problem,
> and not =3D> use some other [$]NIX.
> =3D>
> =3D> One last thing, if your CPU's are baking out and crashing, are you n=
ot
> =3D> nervous that under load this will happen no matter what the OS? Twea=
king
> =3D> system variables will not help you if your server is working ultra-h=
ard,
> =3D> at some point you will reach a mark that your system should still be
> able =3D> to do which currently it can't.
> =3D>
> =3D> I doubt hardware manufactuers put out equipment that can't run at 10=
0%
> at =3D> least.
>
> FWIW, I doubt the accuracy of that last paragraph, and don't think
> this is "so seemingly far fetched" at all. :-)
>
> I have a related problem. =A0In my case, it's a borrowed laptop on which
> I installed FreeBSD 5.1-CURRENT (quite a while ago, but last
> {build,install}{kernel,world} was circa July 2003). =A0Also installed on
> the system is Windows 2000 Professional. =A0The related problem I have
> is that I can fairly easily get the laptop to power off due to
> thermally-initiated shutdown using FreeBSD (complete with "current
> temperature has exceeded system limits" type messages on the console
> beforehand), but can't seem to do so via Win2K. :-(
>
> Now I know that in a sense this is apples and oranges, because I don't
> do precisely the same things under both operating systems. =A0But, it
> seems that high-CPU/system activity under FreeBSD will ultimately lead
> to a thermal shutdown, but not on Win2K (no so far as I've been able
> to manage, anyway). =A0This is inconvenient, to say the least. =A0For
> example, a FreeBSD buildworld or buildkernel will not complete; it'll
> get part way through before the machine becomes too hot and shuts
> itself down. =A0Similarly, building "big" ports like Mozilla won't
> complete, which makes portupgrade a bit of fun. =A0Needless to say, this
> system doesn't get updated much. :-)
>
> Now I'm not saying the machine doesn't become physically hot when
> running Win2K, too. =A0It does (e.g., when playing CPU-intensive games,
> etc.). =A0But somehow, Win2K is able to manage things so that the system
> does not become so hot that the shutdown kicks in.
>
> So, I'm wondering if there's some sysctl or other knob that can be set
> in FreeBSD that will ameliorate this problem. =A0(I thought
> laptop/mobile CPUs generally were able to step down to lower clock
> speeds to conserve power/run cooler, for example.) =A0If I could do
> system rebuilds and port builds without having to restart that'd be a
> big improvement! :-)
>
> Unlike the original poster, this is an Intel-based system, not Athlon.
> It's a Gateway Solo 450 laptop. =A0If I didn't know better, I'd think
> that Gateway "engineered" (pah!) this system so it would run Windows
> "okay" and that's it as far as they're concerned. >;-) =A0FWIW, attached
> at the end of this message is a copy of /var/run/dmesg.boot in case
> anyone can suggest something to help.
>
> Cheers,
>
> Paul.
>
> PS: I'm glad I'm only borrowing this laptop and didn't buy it!! =A0The
> owner of the laptop only uses Windows, so this is only a problem for
> me running FreeBSD.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200311051122.35429.jshamlet>