Security question (sort of)

From: jpero_at_cgocable.net <(jpero_at_cgocable.net)>
Date: Sat Mar 20 23:09:01 1999

Date sent: Sat, 20 Mar 1999 20:34:02 -0500 (EST)
Send reply to: classiccmp_at_u.washington.edu
From: Max Eskin <max82_at_surfree.com>
To: "Discussion re-collecting of classic computers" <classiccmp_at_u.washington.edu>
Subject: Re: Security question (sort of)
Originally to: Discussion re-collecting of classic computers <classiccmp_at_u.washington.edu>

> On Sun, 21 Mar 1999, Tony Duell wrote:
>
> >The problem is, I can't think of a single way to damage any UART chip by
> >programming it. I've read the data sheets for quite a lot of standard
> >(and obscure) serial chips and there is no way to damage them...

Uh, uhuh... can anyone still their assummpations and endless
debates about which OSes is good for a second? Some have good
points but let's face this, some chipsets and CPU's can be blown
by "overrevving" and been done before in rare cases. And that
sounds like Jason might had it happen to him via backorifice
infected system or similar methods. Too bad Jason's case is less
certain because he erased the HD without a careful fine tooth thru
it to find out how that came about.

PnP on all the cards are modifiable plus the bios (PnP muddles
with it by PnP bios and again by Win9x if configured improperly.
Write protected jumper is really a bad joke only protects the boot
area on bios, what worse, M$ decreed that no jumpers of certain
items be used on current boards of last 2 years and future so it's
wide open now. :-(
That is true of boards out there currently without any hardware
jumpers to protect the bios.

>
> For one thing, maybe Jason had a coincidental power surge? Certainly not
> spark leaping from my finger to the mouse).

That is another possible way to hose that chipset.

Happened to me once on a beloved Asus 'TP4XEG board but I was
able to swap the SMD I/O chip for another off a 5 dolllar junkyard
board special.

>
> --Max Eskin (max82_at_surfree.com)
>

To Jason, try to avoid recontacting the "viruii" on your data stuff,
remove everything, take your bios ic out and do the hot swap
reflash on a known working system. Take the CMOS coin out,
short the battery contacts and short all the chips with grounded
blunt metallic object few times around and finally check that stupid
PSU for -5V and -12V. Then put the coin back in, and a good bios,
HD's and cards left out outside of computer system, power up and
see if you can re-enable and test the serial/parallel ports.

All Dallas style and clones based on this with internal battery
backed IC requires PSU in place and CMOS clear jumper in place
and blip on the power for 5 seconds then power off, pull the that
jumper. to clear the CMOS contents.

cih viruii is known to mess up the bios as well.

BTW, what model of that system or what brand or model of that
board that that has conked out i/o ports? Details please! Intel is
infamous for using nonstandard NVRAM on many of their boards.
:-(

If you have any sick PnP cards, see if you can clear it out by
resetting the settings or exhange them where you got those parts
from.

Wizard
Received on Sat Mar 20 1999 - 23:09:01 GMT