"Pred" said:
Got a batch of these all identical and they seem to fail on reboot every now
and again. Some take longer than others but all 20 of them do it even with
the latest BIOS.
3gighz Prescott
2x256 DUAL channel DIMMS Corsair performance
2xSeagate SATA as a raid0 array
Other than that fairly bog standard hardware although I must say I don't
care for the PSU's as they seem a bit generic for my tastes. Anyone seen
anything like this?
The first few motherboards of one of the P4Pxxx family
had a problem like this. It only happened with Antec Truepower
power supplies. A component value change on the motherboard
fixed the problem. (My theory was, it was related to the
power supply ramping up too slowly, and causing the
Vcore protection circuit to trip. Any time a Vcore circuit
has a "latch off" feature in some overvoltage or overcurrent
circuit, it means powering off the board, before the Vcore
circuit will start again. Vcore circuits with "hickup" mode,
will grunt through a condition like that, and continue to
boot.)
I would start by selecting the unit with the highest failure
frequency, and get your hands on a PCI POST card (the one with
the two digit hex display). See whether the display is advancing
or not, when the unit fails to boot. If the display doesn't advance,
then it could mean something like a Vcore failure. You could also
take a voltmeter, and check Vcore when a board fails, and see if
the voltage is a reasonable value for the processor being used,
or the voltage is close to zero (meaning the Vcore circuit is
latched off).
There are many failure paths at bootup. For example, I had problems
at startup caused by an IDE cable being partially inserted. Any
circuit upset near the Southbridge could have an impact on the
startup sequence.
If you are into swapping stuff, take another unit with a high
boot failure rate, and change out the PSU. Buy something like
a PCpowerandcooling supply, where you know you've got quality.
See if that makes any difference.
In any case, you need to collect some more data, in order to
get a better idea of what is letting you down. There can be
some pretty obscure failure conditions. I was in a room on
a customer site once, where there was a metal radiator at
floor height. There were chairs on casters you could roll
around the floor. If the metal undercarriage of a chair
was allowed to bang against the metal radiator, several of
the computers in the room would reset themselves. Try
fixing that
With that many boards involved, you should also phone
Asus tech support, and see if they have any reports of
the problem. They might already be aware of the problem.
HTH,
Paul
