Random Reboot on a P4B

  • Thread starter Thread starter Victor
  • Start date Start date
V

Victor

Hello,

I've got a P4B about 1.5 years old. It works quiet well...yet in the
past two weeks it will randomly reboot without any notice. Also, in
the past few months it will not shutdown, it restarts instead (was
bothered but not concerned).

When it is in the process of booting, it has also rebooted, even
before it has reached the part where win2000 starts loading.

Otherwise, no lock ups or otherwise issues.

Any thoughts on this would be appreciated.

I've cleaned the machine completely.

One suspicion I have is running two maxtor drives master/slave on the
primary IDE drive, that are different models. One is ATA/100 and the
other is ATA/133 both maxtor.

I've replaced the ide cable with a new one. Updated the bios also,
reset the bios. No change.

Thanks in advance.

Victor
 
Hello,

I've got a P4B about 1.5 years old. It works quiet well...yet in the
past two weeks it will randomly reboot without any notice. Also, in the
past few months it will not shutdown, it restarts instead (was bothered
but not concerned).

When it is in the process of booting, it has also rebooted, even before
it has reached the part where win2000 starts loading.

Otherwise, no lock ups or otherwise issues.

Any thoughts on this would be appreciated.

I've cleaned the machine completely.

One suspicion I have is running two maxtor drives master/slave on the
primary IDE drive, that are different models. One is ATA/100 and the
other is ATA/133 both maxtor.

I've replaced the ide cable with a new one. Updated the bios also, reset
the bios. No change.
Click to expand...

The problem could be from a number of sources, software or hardware.
Hardware connectors, such as the CPU socket, DIMM slots, or AGP slots
could be dirty or making poor contact. It'd be worth a quick inspection
of the motherboard capacitors (as well as the video card), looking for
"bulging" caps or leaky brown ooze underneith them. To rule out hardware
faults, try the following:

First - Get memtest86, and run it to ensure the memory is stable. Let it
loop for over an hour as well. If you have failures, there is some sort
of fault in the memory system. Try testing individual sticks after a
failure to identify which is the problem.

Next - Get CPUBurn4, and let it run for over an hour. If it fails (or
reboots), the problem may be with the motherboard/CPU/memory/etc. It will
not be so easy to isolate the source of this fault. Here you might
substitute Prime95 or Seti, or run them in series/tandem.

Run some sort of video intensive test/loop. 3dmark provides some good
tests, and I like to run timedemo loops from Quake3 (but any similar game
like Unreal/etc will do). Failures will implicate a number of sources
including motherboard/CPU/memory, and particularly the video card.

If all the above pass, then I'd be suspicious of a software/driver issue.
I've had these before, but not so much with win2000...

--
We HAVE been at war with Iraq for 13 years now, bombing their
country on at least a weekly basis.
"U.S.-led sanctions have killed over a million Iraqi citizens,
according to UN studies" - James Jennings
3,000+ innocent Iraqi civilian casualties can't be "wrong"...
 
2 weeks you say. Got anti-virus software running? Do you have the latest
flavour of patches to Explorer if you use it.
Your problem sounds suspiciously like some of the viruses that are currently
popular.
Win 2K is not a particularly friendly OS. If you have the chance I would go
to XP it is a lot more stable and less error prone though it still gets
itself in a knot occasionally. If you have to stick with W2K try repairing
it observing all the good things like backing up critical data etc.
My experience with Microsoft is that reloading the software seems to cure
most of the problems till the next time
 
Also,

I neglected to mention this, but it's worth using the Asus Probe utility
or MBM (MotherBoard Monitor) to get a feel for the system and CPU
temperatures. Also check in the bios when applicable (boot-up, reboot,
etc).

Measure temperatures at start-up, how they are after memtest, how they are
idle in OS, how they are during cpuburn, how they are after video looping.
These should give some good clues as to whether there is a heat issue or
not...

I'd expect the following sorts of temperatures:

Excellently ventilated P4 enclosure:
Idle: CPU 30'C, System 27'C
Load: CPU 40'C, System 29'C (CPUBurn4)

Adequately ventilated P4 enclosure:
Idle: CPU 35'C, System 30'C
Load: CPU 55'C, System 35'C (CPUBurn4)

I used to get the "adequate" measurements on my P4 tower, which had a
2-fan PS, 2 fans in the bottom front blowing in, and one fan at the top
back just below the PS. The problem with this is, even though there are 3
case fans, they have to suck/blow through tiny holes drilled in the case
material, which significantly reduces CFM flow.

With my new case design, I get the "excellent" measurements. This design
uses the same 2-fan power supply, but only 2 case fans. An 80mm hole is
drilled into the tower side panel, centered over the CPU, DIMMs and AGP
video card. Another 80mm hole is drilled into the top of the tower,
between the PS and the CDROM area. A fan (and grill) are mounted on the
side panel blowing in, with another mounted on the top blowing out. Very
efficient cooling design; however, it gets the internals very dusty/dirty
(after ~1 year there is an ~1mm thin layer of dust on everything),
requiring some clean-up...

NOTE: All fans described above are 80mm.


The problem could be from a number of sources, software or hardware.
Hardware connectors, such as the CPU socket, DIMM slots, or AGP slots
could be dirty or making poor contact. It'd be worth a quick inspection
of the motherboard capacitors (as well as the video card), looking for
"bulging" caps or leaky brown ooze underneith them. To rule out
hardware faults, try the following:

First - Get memtest86, and run it to ensure the memory is stable. Let
it loop for over an hour as well. If you have failures, there is some
sort of fault in the memory system. Try testing individual sticks after
a failure to identify which is the problem.

Next - Get CPUBurn4, and let it run for over an hour. If it fails (or
reboots), the problem may be with the motherboard/CPU/memory/etc. It
will not be so easy to isolate the source of this fault. Here you might
substitute Prime95 or Seti, or run them in series/tandem.

Run some sort of video intensive test/loop. 3dmark provides some good
tests, and I like to run timedemo loops from Quake3 (but any similar
game like Unreal/etc will do). Failures will implicate a number of
sources including motherboard/CPU/memory, and particularly the video
card.

If all the above pass, then I'd be suspicious of a software/driver
issue. I've had these before, but not so much with win2000...
Click to expand...

--
We HAVE been at war with Iraq for 13 years now, bombing their
country on at least a weekly basis.
"U.S.-led sanctions have killed over a million Iraqi citizens,
according to UN studies" - James Jennings
3,000+ innocent Iraqi civilian casualties can't be "wrong"...
 
Thanks for your advice, I've run memtest86 for an hour and so far no
reported problems.

I'll be doing some of your suggestions and post results.

Thanks!

Victor
 
Had the same thought of a virus. I'm running the current engine and
dat files for mcafee.

Unfortunately, Twice when it has rebooted by itself, it rebooted
before the os started loading...so I'm sure it is a hardware problem

Thanks for the advice.

Victor
 
Had the same thought of a virus. I'm running the current engine and
dat files for mcafee.

Unfortunately, Twice when it has rebooted by itself, it rebooted
before the os started loading...so I'm sure it is a hardware problem

Thanks for the advice.

Victor
Click to expand...

Do you have a spare power supply handy ? Give it a try.
Power supply failures are a common cause of flaky behavior.

Paul
 
I've disconnected a new drive and took out a new board I had installed
and the machine so far is stable for the last two days...

Paul, a very good idea. I'll be buying a new power supply...believe
this may be the culprit. The machine runs too good for a serious
problem with the motherboard to exist.

Thanks!

Victor
 
For those interested:

I've replaced the power supply and the computer has been stable and
working well and even shutsdown properly now.

I can only imagine the addition of usb cam, hard drive, usb printer
will rob a lot of power from the power supply.

Thanks for the suggestions and help.

Victor
 
Back
Top