From: "J R FOX" <os2-wireless_users@2rosenthals.com>
DomainKey-Signature:a=rsa-sha1; q=dns; c=nofws;
  s=s1024; d=pacbell.net;
  h=Message-ID:X-YMail-OSG:Received:X-Mailer:Date:From:Subject:To:In-Reply-To:MIME-Version:Content-Type;
  b=lNLngeWhFSdL4iVu2lcyTbgBk3UtFQM3Cxldx2Q4HJ6ePyO9JBSjOisQMCqcwnF+G+3qhd7yfF+8ZUXxYMFzK9XTH2iH/oNpKNnA/Bn+S9rBFFfYGPzOx6VvlL5Hm80N98LSW5R/O3RJiP10AQOJasKUQXgWUXgHnd7uk0ZII54=;
Message-ID: <901206.37542.qm@web81404.mail.mud.yahoo.com>
Date: Mon, 15 Feb 2010 20:41:29 -0800 (PST)
Subject: Re: Troubleshooting a non-booting computer  [was: Driver Problem]
To: os2-wireless_users@2rosenthals.com
In-Reply-To: <4B7708F1.2000104@2rosenthals.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii

First of all, thanks so much Lewis for reconstituting all of that 
List material that fell into the abyss over here !

In regard to the ailing Shuttle, it is now looking as though you and 
Ed got it right.  

Ed wrote:
"If you can get into the BIOS setup set if you can clear the settings 
back to factory settings, if not, power off and see if there is a reset 
CMOS jumper on the motherboard and if not, remove the CMOS battery and 
leave it out for a few minutes before returning it."

"This sounds like a CMOS data corruption problem to me."

There is actually a tiny, recessed button on the back panel that can 
reset the CMOS.  But, I did not use it.  I recalled that the clock 
on this computer had been unreliable for many months, and that was 
probably a clue.

Lewis wrote:
"It sounds like a bad CMOS battery."

" . . .  I could let this box sit a while), and brought it in and 
put it on the bench. Sure enough, it came up this time, but the 
clock was reset (a telltale sign)."

"It's likely that the CMOS battery is a CR2032, which is a 3v battery." 
If you have a digital voltmeter, you can test it. Otherwise, it's a relatively inexpensive test to simply replace it and see what happens." 

I have a battery tester that I think can test this type of battery, 
but it is one of innumerable things I still can't find, after the 
move.  I have to go to Radio Shack anyway, and may have them test it, 
just to confirm what is almost certain.

"Another suggestion would be to disconnect *all* peripherals from the
power supply and board, and see if you get any further. Sometimes, 
the short may lie in a connected component."

To date, the only other installed component is the optical drive. 

"Check that battery & let us know how you make out."

I did happen to have a couple of spare CR-2032s, and those I could 
find.  The space inside this type of computer is extremely tight.  
To get best access to the battery, it could be necessary to do some 
partial disassembly, then reassembly, and a simple job becomes a half-
hour pain in the rear.  Having the old or the replacement battery drop 
into the works would have been worse.  So, I used a strip of masking 
tape to tether each battery while doing this, which worked out quite 
well.

The system has now come back up a couple times.  I've had to redo the 
CMOS settings, and later will compare them to a file I saved somewhere 
detailing what they were.  The one I'm least sure about right now is 
the CPU frequency, which can range from 200 to 204.  200 seems to be 
the default, so I've left it there for now.  The fans still seem 
noticeably quieter than they were, even after I've put their CMOS 
setting back to 'Medium', which is where I thought they had been.  
Whether this will be sufficient, I don't know, as I haven't run 
anything yet that stresses the system.

Do systems of the last few years automatically shut themselves down, 
if they get too hot . . . or will they let everything fry ?  I thought 
I read something about a failsafe shutdown becoming standard ?

I've now installed Speedfan, which may prove helpful here, once I learn 
how to use it.  Do we have anything like this on the OS/2 side ?

Thanks again for your suggestions.


 Jordan