Article View: comp.sys.next.sysadmin
Article #1379Re: 10base-T or thinnet?
From: Robin D. Wilson
Date: Thu, 08 Oct 1992 14:39
Date: Thu, 08 Oct 1992 14:39
58 lines
3694 bytes
3694 bytes
In article <root.718320413@merlin> peter@merlin.acadiau.ca (Peter Steele)
writes:
> What's the true story? NeXT says the ports are autosensing, and
> someone has told us they are not. If someone could clarify this
> we'd appreciate it very much. Thanks.
Well, NeXT is right... We had several machines connected to Thinnet, and just
removed them from the thinnet, and plugged in the RJ45 on the 10BASE-T;
rebooted and everything was fine.
I suggest you look at two things:
1) Your twisted pair. We have found that FLAT phone wires work OK for
limited distances, but over long hauls (100+ feet) they seem to have very
flaky problems. We recommend that you use twisted copper wiring that meets
the 10BASE-T spec (x number of twists per foot).
2) Your ethernet port has died. We had one go out on us, but both the
Thinnet and the 10BASE-T died at the same time, so this seems less likely to
be your problem.
NOTE: for wiring problems it is best to test each piece of the wiring back to
your HUB. The easiest way to do this is to test the connection directly at
the HUB. Take the machine to the HUB (or the HUB to the machine), plug in
the wiring that will run from the wall outlet (demarc) to the machine. After
you verify that wire, check the wire that runs from the patch panel to the
HUB. If both of these work, then it is likely that the wiring from the
demarc to the patch panel is the culprit. You could test this by using the
verified wires on another outlet set.
Of course, I have made a broad assumption here that your wiring looks like
this:
Twisted Pair
------ | /-------------------\
|NeXT| |<Wall/ \ Patch Panel
| | | / \ ---------------
------ Demarc\ | / \_____| |x| | | | | |
------__________\+__/ | ||| | | | | |
------ | ---|-----------
|
---|-----------
| ||| | | | | |
| |x| | | | | |
---------------
HUB
If your wiring doesn't look something like this, then you'll need to adjust
your plan accordingly. The basic goal is to identify the smallest piece of
the system that could be broken. So you break off each segment and test it
separately. Good luck.
--
+---------------------------------------------------------------------------+
|They didn't even know I could talk, much less have opinions (yeah right!) |
|USNail: 701 Canyon Bend Dr. Internet: robin@pencom.com |
| Pflugerville, TX 78660 |
| Home: (512)251-1737 Work: (512)343-1111 |
+---------------------------------------------------------------------------+
Technical Details
Message-ID:
<1992Oct8.143937.16320@pencom.com>Path:
arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!darwin.sura.net!zaphod.mps.ohio-state.edu!cs.utexas.edu!natinst.com!news.dell.com!uudell!pensoft!usenetReferences:
<root.718320413@merlin>Show Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!haven.umd.edu!darwin.sura.net!zaphod.mps.ohio-state.edu!cs.utexas.edu!natinst.com!news.dell.com!uudell!pensoft!usenet From: robin@pencom.com (Robin D. Wilson) Newsgroups: comp.sys.next.sysadmin Subject: Re: 10base-T or thinnet? Message-ID: <1992Oct8.143937.16320@pencom.com> Date: 8 Oct 92 14:39:37 GMT References: <root.718320413@merlin> Sender: usenet@pencom.com (Usenet Psuedo User) Reply-To: robin@pencom.com Organization: Pencom Software Lines: 58