Xref: utzoo comp.sys.ibm.pc.misc:6629 comp.dcom.lans:7292 comp.unix.sysv386:5145
Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!swrinde!elroy.jpl.nasa.gov!usc!wuarchive!hsdndev!husc6!ukma!s.ms.uky.edu!kherron
From: kherron@ms.uky.edu (Kenneth Herron)
Newsgroups: comp.sys.ibm.pc.misc,comp.dcom.lans,comp.unix.sysv386
Subject: Starlan and RN:  together at last
Message-ID: <kherron.666920178@s.ms.uky.edu>
Date: 18 Feb 91 23:36:18 GMT
Distribution: na
Organization: University Of Kentucky, Dept. of Math Sciences
Lines: 21


Thanks to all of you who replied.  The problem, just as Les Mikesell
stated, was that read() from a stream device could return -1 where a
similar read from a real tty would return 0.  Rn was blindly using the
returned value, resulting in an apparent read() of 2^31-1 (or so)
characters which it tried to process as commands.  A fixed version
of rn has been sent to the site; they report that it works great.
Along the way I also got them a later version of kermit, which they
say also works great (but didn't fix the problem, of course).

Posting a patch to rn would be rather pointless since I wasn't even
working with the latest version, but if anyone needs more detail,
feel free to ask.

And a special thanks to Les Mikesell who, I gather, figured this out
the hard way some time back.
-- 
Kenneth Herron                                            kherron@ms.uky.edu
University of Kentucky                                        (606) 257-2975
Department of Mathematics 
                                "Never trust gimmicky gadgets" -- the Doctor