I have some time to work on this project as well so I don't mind
making it a learning experience. It's not something I'd like to do
for a living mind you but I think it will be good knowledge to have.
Thanks all,
Jon
On Jan 27, 2006, at 8:06 PM, Dan Shoop wrote:
At 6:21 PM -0800 1/27/06, Jon Wynacht wrote:
Hi,
Kind of off topic but not really...
I'm looking for recommendations for network testing equipment.
Well the first thing to do is to define the physical network types
you're referring to. I have a boatload of Thickwire and AUI stuff
that helps me squat these days with my Ethernets.
But the answer here is the buy the tester you're already most
familiar with using. There in lies the rub and catch-22.
I have to install an Xserve and a bunch of macs and networking
equipment for a customer. They're moving to a new building and the
previous tenant left their equipment rack and all their wiring in
place and nicely set up.
Don't trust it.
However, not everything is labeled so I'm looking for the best
device(s) to:
1. Test whether the Cat 5 is good.
Consider it bad.
2. Test to see if both ends of the Cat 5 are good, so I can
determine which port in an office maps back to which port on the
patch panel.
You need much more than this to verify the stuff. You'd need a TDR,
impedance testers, and to run these sorts of tests you generally
first have to know which cable is which.
Your fist task is to label the cables. You could do this simply by
plugging all the cables coming into the rack into a bunch of
switches and stepwise going to each end of the wiring and plugging
in an Ethernet Loopback connector. This will light up the port on
the switch (provided its good). If you don't have a loopback
connector make one. If you can't make one use a Powerbook or
notebook with a a known good NI. You'll still however want a
loopback connector for further tests.
Then you need to verify each cable. For this you need a good cable
tester. Some of these may also come with doggles so that you can
plug the doggle into the far end and given the way it shorts out
the pairs can identify with doggle (and therefore which cable)
you've attached. Better cable testers will also allow you to print
the results of the test runs and the graphs necessary for
evaluating the characteristics of the cable, most important of
which on "passed" cables is noise.
None of this is a pleasant task and is very time consuming.
The cost of all the time and equipment to do so is probably around
the same as new cable runs, and with the latter you've better
assurances.
However if you're asking for recommendations then this means you've
not done this before, otherwise you'd already know models you've
used before and probably select one of those. Unfortunately this
isn't one of those cases of where learning on the job pays off well
as misreading or not knowing what tests to run and what the results
indicate is far worse than just presuming everything all works.
If you don't want to just rip it all out and sell it for scrap, by
far the more prudent call, then consider calling in a network
consultant group to test all this for you. It's probably going to
cost less that buying the equipment yourself and all your time,
plus once they certify it they're going to have to stand behind
their work. You may even want them to then maintain it for you.
--
-dhan
----------------------------------------------------------------------
--
Dan Shoop AIM:
iWiring
Systems & Networks Architect http://
www.iwiring.net/
email@hidden http://
www.ustsvs.com/
1-646-217-4725
