Your problem is most likely an intermittent short in one of your
lengths of cable.  Are your cables home made?  I usually use two bent
paperclips, an AA battery and a flashlight bulb to test the segments and
make sure they're not shorted or disconnected.  Take the first paperclip,
push it into the end of a T connector so that it touches the dialectric and
and the outer shell of the T connector.  Then, plug one end of your cable
segment into the other side of the T connecter.  Next, connect the other
end to another T connector.  Now, connect the battery, bulb and paperclip
to the other side of your second T connector such that the cable is part of
your circuit.  Now, if you disconnect the first paperclip and all is well,
the light wil go out.  If your light still shines, then you have a short.
If your light doesn't work at all, then you either have a bad cable  or a
bad T connector.
	If your cable segments are pre-connectorized, then the most likely
cause of your trouble is a bad T connector.  I have seen some T connectors,
as well as cables, stop working after the stress of connecting and
disconecting.
	The device timeout message means that the driver isn't seeing any
activity from the ethernet card when it sends out a packet.  This could be
either that you have the card on the wrong interrupt, an impossibility
since it worked before the network work began and since you didn't say you
took out the card in your working machine, or that the medium (the cable)
is bad somewhere.  Since you've been working on your network and since a
break in the wire at any point will take out the entire net, you'll have to
make an educated guess as to where the trouble is, or test each segment 
individually.

Hope this helps.
-Brian

On Feb 13,  9:17am, Mason Loring Bliss wrote:
} Subject: ThinNet Madness...!
} Last night I was about to install an i386 on my ThinNet. Prior to this, I
} had my Mac Quadra 610 running MacOS, an old Mac IIcx running MacOS, and an
} i386 running NetBSD-current. All worked pretty well.
} 
} I unplugged the i386, which acts as a router, and removed the terminator.
} Then I added a length of cable, and plugged the terminator into the end of
} that. (Well... Actually, I put another T connector on the end, for the new
} machine, and plugged the terminator into *that*.)
} 
} Now, nothing works. The IIcx on one end of the chain doesn't report errors,
} but the Quadra says, helpfully, "there's an error" and refuses to see its
} net connection. The i386 gives me "we0: device timeout". The new i386, when
} booted from an install floppy, gives me the same thing. I didn't have all
} the machines powered down when I removed the terminator, but I can't believe
} that the things are that fragile. If they were, organizations would have to
} buy new network cards for all their machines every time a wire accidentally
} got chopped, and the medium would have died out long ago. (Well, okay, maybe
} it *has* already died out, but it's what I've got!)
} 
} I've tried isolating the i386 and my Quadra with a T connector and two
} terminators, but both give me the same errors, regardless of the presence
} or lack thereof of others machines on the wire.
} 
} This is baffling. I've tried connections with two arbitrary machines on the
} wire with no luck.
} 
} This acts like all the machines torched their EtherNet controllers all at
} once, but I find this to be an implausible explanation. One interesting
} fact is that I got the same message on the Quadra when first attempting to
} wire the machinery together, back when it was just the Quadra and the IIcx.
} I gave up in disgust after a while, and left the whole mess sitting there,
} connected but unused. A few days after that, I tried using the EtherNet so
} that I could show the error to a friend, and it worked. It's worked reliably
} for a month or two now, and I was able to add in an i386 to use as a router
} with no problems.
} 
} What can cause these "device timeout" messages? I'm not getting any other
} errors, and "device timeout" isn't quite detailed enough for me to guess what
} might be going on with any measure of assurance.
} 
} Thank you for help and/or advice. None of this seems even slightly logical
} to me, and I've run out of possible solutions.
} 
} -- 
} Mason Loring Bliss...mason@acheron.middleboro.ma.us...www.webtrek.com/mason
} "In the drowsy dark cave of the mind dreams build their nest with fragments
}  dropped from day's caravan."--Rabindranath Tagore...awake ? sleep : dream;
>-- End of excerpt from Mason Loring Bliss