Tom writes:

> >However, I would suspect the cabling first. Here you often have
non-soldered,
>  >corrosion-prone mechanical connectors that you are trusting to make good
>  >connections.
>
>  It's not the cables.

Of course, that was only a hypothesis, ordered on the basis of experience.

If you're absolutely sure that the cabling has been eliminated as a possible
source (including insuring that the terminators are well seated), then it's
on to the next most likely suspects, either the hub or the HP3000 NIC card.
In my instance of debugging nearly exactly this same problem, I was fortunate
enough to have a retired a six year-old 917 sitting next to the new 918. I
found that they used exactly LAN card in the machines and swapped them out,
but to no avail.

In our case, it was the cabling, but that means nothing in your case.
Changing out the hub is the easier thing to do, but the LAN card is also a
perfectly legitimate suspect.

Wirt Atmar

* To join/leave the list, search archives, change list settings, *
* etc., please visit http://raven.utc.edu/archives/hp3000-l.html *