[olug] eth0 again

Dave Burchell burchell at inetnebr.com
Fri Aug 31 06:09:42 UTC 2001

On Thu, Aug 30, 2001 at 07:05:53PM -0500, Tom wrote:
> Ok a couple days back I asked about this and got a bunch of responses I
> have tried everything even moved card but nada.

> Here is what I have and what I get.  It is a EtherEZ SMC-8416T ISA
> card.  The how to says yes it is supported as SMC-ULTRA and they
> recommend disabling the PNP in bios for that IRQ.  So I booted in the
> dos mode and ran the SMC utility to get the I/O and irq and made
> appropriate changes.  I rebooted and it said getting etho delaying init
> of eth0 then fails.

> I added the info to the modules.conf and restarted same thing I went in
> and added it with netconfig now system locks up on boot.  It does not
> delay eth0 but just stops there.  I pulled out the sound card and modem
> on the chance it was a conflict that I was not seeing as when I changed
> it one time the sound card crashed too.  Oh well ANY ideas or where
> would be a good place to start over from.  How can I get by where it
> stops on boot other than reloading linux (which I am getting tired of
> doing).

> Thanks Tom

Have you tested this particular card in another computer and found it
to work?  If not, do so.  If so, read on.

I don't think I have any advice to actually solve your problem, but
perhaps I can make you feel better about failure.

We think of computers as working with ones and zeros, on or off, yes or
no, true or false.  There is no middle ground there; no fuzziness.
Computer science doesn't contain margins of error.  But _computers_
do.  Computers, the computers we build and use today, are about
electrical engineering.  You could become a computer science immortal
without ever touching an electrical computer (Babbage did, for
example).  You could design computer components without ever
understanding the fundamentals of computer science.  The two are not
the same thing.

Unfortunately, sometimes we forget that our desktop boxes are not
Turing machines, running endlessly and flawlessly according to the laws
of pure logic and mathematics.  Our computers are an interconnected
array of capacitors, transistors, resistors, circuit boards, and (of
course) ICs.  These components heat up and cool down.  They sometimes
function at the outer edge of their design specs -- even at the time of
manufacture.  Plus they fail with time or use, sometimes spectacularly,
sometimes subtly.  Your motherboard and your Ethernet card were built
by different people with different design philosophies.  Of course,
they both built to a common interface specification, which is why the
vast majority of PC components interoperate so well.  But sometimes
designers and manufacturers miss the mark.  One is a little low on some
spec; the other a little high.  Almost never are they off spec enough
for there to be a problem.  But once in a while, two small mistakes
multiply into a big one, and two components, which _should_ work just
fine, and do in other environments, just don't want to work together.

When Turing built his theoretical computer in his head he never had a
component failure.  But Babbage had a devil of a time getting all the
brass gears to work together as designed.  And, from time to time, so
do we all.

I'd be happy to mail you a spare ISA card I have that was rumored to
have worked once.

Dave Burchell                                          40.49'N, 96.41'W
Free your mind and your software will follow.              402-467-1619
http://incolor.inetnebr.com/burchell/                  burchell at acm.org     


For help contact olug-help at bstc.net - run by ezmlm
to unsubscribe, send mail to olug-unsubscribe at bstc.net
or `mail olug-unsubscribe at bstc.net < /dev/null`
(c)2001 OLUG http://www.olug.org


More information about the OLUG mailing list