Where i can check a log about the last reboot time of myfreebsd box

[Bill Vermillion]

While normally not able to pour water out of a boot with
instructions on the heel, on Wed, Aug 25, 2004 at 17:53  
our dear friend Carlos David uttered this load of codswallop:

[Lucretia Deletia hacks away a lot of excess txt .. wjv]


> From: "Bill Vermillion" <bv at wjv.com>

> > > owner-freebsd-isp at freebsd.org wrote:

> > > > Hi, im having troubles with a freebsd box, suddenly this fbox
> > > > reboots many times on a day, every hour at least, :(, i dont
> > > > know what could be happening, this fbox was working great, i'd
> > > > like to know if a log on the system could helpme to check if a
> > > > program or process is making some bad things.. With this box
> > > > i share internet to many clients.. am worried :(. thanks in
> > > > advance.

[someone else said].

> > > A normal running system that suddenly starts rebooting its self is
> > > more that likely an overheating problem caused by a thick layer of
> > > dust on the motherboard chips or early signs of a failing power
> > > supply.  You will not find any log error messages about this.

> > If he had looked more carefully at the log he posted, a good many
> > of the 'reboot' lines had one right before it that said 'crash'.
> >
> > So there was at least an error indication, but he didn't seem to
> > ask about it - just the 'reboot' lines.  Both of your suppositions
> > fit the problem.  I thought it was mildly interesting in that
> > at least some of the reboots were preceded by a 'crash' line.
> >
> > That would indicate to me that it really is heat or power, as
> > it is inconsistent in it's reboots.


> I attach my last output command..., yessssss i have many crash lines
> then is problem of overheating................. :(

I'll not inlcude the log file you attached.  I see many that were
just 'reboot' with nothing else associated with them.   

Then there are those that are rebeoot followed by a 'crash' line.
[remember the log is listed in reverse order].

These reboots are sometimes coming two minutes apart.  To tell if
it is heating about the only thing you can do is turn the machine
off for an hour or so so it cools down and see if it takes longer
to reboot.

It could also be a memory problem.  In the old days when everyone
ran parity memory - the system usually could write an error message
to the screen before it totally went away.  But on systems that
didn't support that the system usually just quit.

You could have a heat problem.  Is the case warm to the touch?
Open the case and make sure the CPU fan is turning.  I've had those
fail before.

If that check out, I'd suspect power supply and/or memory. Power
supplies are pretty cheap any more, and because they are cheap I've
noticed they seem to fail more often than the ones we used to have.

The other place could be memory.

You will have to check each of these yourself as they are all
physical problems.   If it's not heat, I'd suspect power supply
first.  That's based on past experiences with having only 1 memory
stick fail.  But then again I only buy good brand name memory.

When I had one system fail with what looked like a memory problem I
took it to a dealer who had a HARWARE MEMORY TESTER. [do not trust
SW tests].  He looked at the stick and said, that's good memory,
it's probably not bad.  He put it in the testing device, and it was
bad.  He commented that he didn't think he'd seen a bad stick of
memory from that company before.  But it can happen to all vendors.
Name brands usually fail far fewer times.

Lot's of luck.  Trouble shooting intermittent problems is
not the most fun I've ever had. [And before computers I worked in
broadcast and recording - as a chief engineer in both places.  I
don't like soldering, and I don't like voltages that can kill you
instantly - that's why I like computers.  I'm bigger than they are
and they know it]. :-)

Bill
-- 
Bill Vermillion - bv @ wjv . com