[mdlug] BIOS viruses ? Oh my!

[Rich Clark <rrclark@rrclark.net>]

On Wed, 19 Mar 2008, Ingles, Raymond wrote:

>> From: Raymond McLaughlin
>>>>>>> On Tue, Nov 20, 2007 at 03:18:01PM -0500, john-thomas
>> richards wrote:
>>>>>>> After pouring much time into this, it has been determined that this
>>>>>>> (and the other two) motherboard has a very special and very specific
>>>>>>> failure that keeps it from booting a GUI.
>
>>> Now, my feeling here is that these machine when running Windows came
>>> across a site that was probably compromised and the compromiser had it
>>> in for any IBM hardware. Created a BIOS hack and got it to load via
>>> Active-Hex via Infernal Exploiter or Lookout... voila, unfixable
>>> machine.
>
> And not cleared by a BIOS update? I won't say it's impossible, but I'd guess
> it more likely that there was a bad production run - perhaps a dodgy capacitor
> was used. Switching video modes is a lot harder on a monitor than a video
> processor, but it still does draw some current and would send some surges into
> the lines. Normally this wouldn't be a problem, but if the filters on the lines
> didn't handle the ringing, or allowed some signal to leak onto other lines...
>
> That'd be the kind of thing that'd take a while to show up, but in a batch
> of machines that were all used in similar ways, they might flake out the same
> way at roughly the same time.
>
> One possible thing to try - a video mode switch without a monitor connected?
> That'd change the load significantly... these things might make useful
> headless servers if that were the case.

You have to take into account the age of the hardware, too. Correct me if 
I'm wrong, but PIII processors and associated hardware were brand spanking 
new and fresh to market in 1997-1998 and only stayed on market for about 
4-5 years. We're talking about 10 years age for most of that gear, and 
even with them sitting on a shelf unpowered and boxed/wrapped capacitors 
are going to corrode, leak and dry out.

I think someone's pulling your leg about the BIOS virus, though. I haven't 
heard anything credible about a BIOS virus since about 1992 or 1993.