[124786] in North American Network Operators' Group
Re: what about 48 bits?
daemon@ATHENA.MIT.EDU (Steven Bellovin)
Sun Apr 4 23:41:09 2010
From: Steven Bellovin <smb@cs.columbia.edu>
In-Reply-To: <n2w202705b1004042029pe4b3f6d4m18a56cef7a33db89@mail.gmail.com>
Date: Sun, 4 Apr 2010 23:40:27 -0400
To: Jorge Amodio <jmamodio@gmail.com>
Cc: nanog@nanog.org
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Apr 4, 2010, at 11:29 47PM, Jorge Amodio wrote:
>> The N connectors were easier to deal with than the vampire taps. To =
add
>> a node, you just "spliced" a new xceiver box onto the line where you
>> needed it by screwing a new length of cable into the new + existng
>> xceivers, then connecting the AUI drop cable from the box to the =
node.
>=20
> I've to say it, the AUI cables were an absolute pain in the ass to =
deal with.
>=20
> We had also a thick coax with the vampire taps and AUI fan outs from
> Excellan. Dealing with the coax was not that bad since we made an
> inverted U and had a coax run on each of the two vertical raisers this
> particular building had.
>=20
> The biggest challenge was to go from the raiser using the existing
> horizontal conduits that were not that big, and run the AUI from the
> XCVR to the fanout unit and then from that unit to each desk.
>=20
> Before going to 10BaseT we used pre-standard LattisNet from SynOptics,
> getting rid of the AUI was a relief.
>=20
Oh, the thick coax and the AUIs were lots of fun. The 15-ping cables =
from the hosts to the AUIs were always coming loose, and the slide locks =
didn't help much. The vampire taps tended to either not make good =
enough contact or to break the center conductor. The N-connectors were =
easier to handle -- but cutting the cable and crimping on a pair took =
down the whole network. =20
And then there was the time an electrician accidentally cut the coax and =
decided to splice it with black electrical tape...
--Steve Bellovin, http://www.cs.columbia.edu/~smb
