Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84 exptools; site ihuxn.UUCP
Path: utzoo!watmath!clyde!burl!ulysses!mhuxr!mhuxt!houxm!ihnp4!ihuxn!res
From: res@ihuxn.UUCP (Rich Strebendt)
Newsgroups: net.bugs.uucp
Subject: Re: uucp hang, all U's
Message-ID: <999@ihuxn.UUCP>
Date: Wed, 20-Mar-85 23:02:53 EST
Article-I.D.: ihuxn.999
Posted: Wed Mar 20 23:02:53 1985
Date-Received: Fri, 22-Mar-85 02:00:36 EST
References: <229@tellab2.UUCP>
Organization: AT&T Bell Laboratories
Lines: 37

In reply to:

> I am having problems with out uucp link to ihnp4.  Every now and then, for
> no reason that I know of, ihnp4 will start sending out 'UUUUUUUUUUUUUUU...'.
> The connection will not time out, and both uucps are happy to sit there for
> hours on end like this.  We call three other sites and haven't ever experienced
> problems with them, so I think that our uucp is okay, but I can't imagine that
> ihnp4 has bad uucp problems!
> 
> Has anyone else seen this problem before?  If so, HELP!! (please).

AHHHHHH, the infamous digital loopback problem strikes again !!

The problem is that one or both of the modems involved in the
connection have the option for the digital loop-back test enabled (as
it is set up at the factory).  This is option YK on my Penril 300/1200
modem (internal dip switch S2-7).  Set your modem up to option YL
(throw the dip switch the other way) and yell and scream and holler at
the comp center administrators to fix ALL of their modems properly.
Also expect as new modems are installed and old ones come back from 
repair that the situation will gradually degenerate until you find it
necessary to yell and scream and holler once more.

Why, you may ask, does this show up as an eternal stream of U's?
It just so happens that the ASCII code for 'U' is 0x55 or, in binary,
01010101 -- the square wave used for the digital loop test.  And how,
you ask again, does the link get into this state?  Just a bit of noise
on the line that the modem decides looks like a request to enter this
state.

Perhaps this question and answer should be added to the new-netter
set of questions.  It is one of those things likely to bite each
of us someday and is very mysterious until the explanation is
obtained.

					Rich Strebendt
					...!ihnp4!ihuxn!res