HP3000-L Archives

December 1997, Week 3

HP3000-L@RAVEN.UTC.EDU

Options: Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Gavin Scott <[log in to unmask]>
Reply To:
Gavin Scott <[log in to unmask]>
Date:
Thu, 18 Dec 1997 10:42:35 -0800
Content-Type:
text/plain
Parts/Attachments:
text/plain (30 lines)
Bruce write:
> Just connecting and disconnecting the NIC won't make a difference. But
> the flexure of the cable that occurs when you connect the NIC may well
> make a difference.

This may not be entirely true.  There are different limits on the length
of an Ethernet(tm)/802.3 lan.  One of them is derived from the packet
timing (what with the speed of light being so damn slow), but I believe
the limit on thinlan (10-base-2) is more electrical since it is
significantly shorter than the timing limit.  From this I assume that
each Network Interface is required to put out some signal level which
is sufficient to drive the receivers in every other NI on the lan at
the same time.  Also each NI has to be able to detect the signal of
any other NI while it is transmitting (collision detection).  So, while
the NI receivers are provably very high impedance, each foot of cable
and each NI connected will increase the attenuation of the signal
somewhat, and a plugged in NI card has a lot more opportunity to screw
things up than an unplugged one :-)

I have seen dysfunctional 10-base-2 networks start working again upon
unplugging a T from a NI card, where it didn't matter *which* T you
unplugged.  I have seen problems that could not be isolated to a
single failure even after individually removing or replacing every
component of the network.  10-base-2 networks are entirely too analog
for my taste, and any network where any failure of any component can
bring down the entire network without a guaranteed time-to-repair
even with $4000 tools is not acceptable.

G.

ATOM RSS1 RSS2