[GRLUG] BIOS problems - Round 2

[Bob Kline]

The machine has been back up and running for a week or so.
I wish I had a warm fuzzy,  educational,  inspirational  story to
go along with it,  but in fact,  after replacing the battery,  and
removing a number of peripherals,  the BIOS started functioning
again.  I loaded up the default setting, saved them,  and then
let the machine boot a ways a few times to make sure things were
still functioning that far.  I then went for broke and replaced all the
cables to the peripherals again.


Up it all came,  and has been doing so since.  I'll entertain the
notion that one of the cables had a corroded pin or two,  but really
can offer no useful insights.  Magic?   Good as anything.


Anyway,  thanks to Tim for a nice list of shakedown items.  In 10
years I've never had a problem like this. Going forward I'll be in a
better position to proceed if something similar happens again.


I still don't know what a "BIOS ROM checksum error"  points to,
but will be just as happy if I don't ever see it again.


   -Bob





On 2/10/06, Tim Schmidt <timschmidt at gmail.com> wrote:
>
> Well, I've had a little time to think about it, and here are some more
> thoughts...
>
> Without more information, I cannot say what caused the board to stop
> working.  I can say that (short of bad capacitors:
> http://en.wikipedia.org/wiki/Capacitor_plague and dirty power)
> motherboards generally do not stop working when left alone.
>
> I suspect that after you removed the battery for some time, the BIOS
> was reset, and the board is now trying to post from a PCI graphics
> card (often the default).  However, here's a blow-by-blow of common
> issues.
>
> RAM can stop a machine from posting if the timing values in the BIOS
> are too agressive or, more rarely, if the RAM is of a configuration
> incompatible with the motherboard chipset (think 256Mb high-density
> dimm in an i440 BX motherboard).
>
> CPUs can keep a machine from posting if they are not supported by the
> current BIOS of the motherboard (reguardless of the board's
> theoretical ability to support such a CPU).
>
> PCI, AGP, ISA cards can keep a machine from posting with buggy BIOSs,
> interrupt conflicts, poor seating, dusty contacts, excess power draw
> (especially in the case of newer graphics cards), slightly out-of-spec
> clock timings (especially in the case of overclocked machines -- some
> motherboards slightly overclock automatically), or cards and
> motherboards explicitly not supporting each other (as in a PCI 2.1
> card in a PCI 1.0 board or an AGP 8x card in an AGP 2x board (this
> generally works, as most cards are capable of backing down, but not
> always -- and further BIOS interactions can complicate things)).
>
> External peripherals like keyboards, mice, monitors, printers, USB
> gadgets etc. can keep a machine from posting.  Circumstances are
> generally odd, unfortunate, and rare, however.
>
> Drives can easily keep a machine from posting, by being configured
> improperly, cabled improperly, or paired with other drives that don't
> like them.  If you'd like more detail on this I can flesh it out
> sometime soon...  PATA drives are a dying breed however.
>
> And the whole damn lot is subject to grounding problems, possible
> shorts, power supply issues, etc.
>
> --tim
> _______________________________________________
> grlug mailing list
> grlug at grandrapids-lug.org
> http://grlug.org/mailman/listinfo/grlug
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://grlug.org/pipermail/grlug/attachments/20060221/3078e292/attachment-0001.htm