home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.19970929-19971216
/
000107_news@newsmaster….columbia.edu _Wed Oct 15 18:24:04 1997.msg
< prev
next >
Wrap
Internet Message Format
|
1997-12-15
|
4KB
Return-Path: <news@newsmaster.cc.columbia.edu>
Received: from newsmaster.cc.columbia.edu (newsmaster.cc.columbia.edu [128.59.35.30])
by watsun.cc.columbia.edu (8.8.5/8.8.5) with ESMTP id SAA23212
for <kermit.misc@watsun.cc.columbia.edu>; Wed, 15 Oct 1997 18:24:03 -0400 (EDT)
Received: (from news@localhost)
by newsmaster.cc.columbia.edu (8.8.5/8.8.5) id SAA27825
for kermit.misc@watsun; Wed, 15 Oct 1997 18:24:02 -0400 (EDT)
Path: news.columbia.edu!panix!howland.erols.net!news.maxwell.syr.edu!scanner.worldgate.com!rover.ucs.ualberta.ca!alberta!not-for-mail
From: Vladimir Alexiev <vladimir@cs.ualberta.ca>
Newsgroups: comp.protocols.kermit.misc
Subject: Re: Kermit via PPP under DOS? (SOLVED!)
Date: 15 Oct 1997 16:06:41 -0600
Organization: University of Alberta, Computing Science
Lines: 63
Message-ID: <omen5m8yvj.fsf@tees.cs.ualberta.ca>
References: <k1c7kBQEU5Wv@cc.usu.edu> <omvhza9x7g.fsf@tees.cs.ualberta.ca>
<5jzp7SiZjuUm@cc.usu.edu> <omn2kcqegc.fsf@tees.cs.ualberta.ca>
NNTP-Posting-Host: tees.cs.ualberta.ca
In-reply-to: Vladimir Alexiev's message of 14 Oct 1997 14:25:55 -0600
to: jrd@cc.usu.edu (Joe Doupnik)
X-Newsreader: Gnus v5.0.15
Xref: news.columbia.edu comp.protocols.kermit.misc:7886
Alright!
Joe, you were right on these points:
- It's EPPPD's fault, not Kermit's fault
- I should have talked earlier to EPPPD's author
I was right on these points:
- It's easy to fix.
- Other apps work because they don't check something that should be checked.
Kermit fails because it's right, but "too righteous". Worse is better :-)
Something that threw me off is that MeritPPP also failed. Two drivers against
one kermit, it's "natural" to assume it's kermit's fault :-) I wonder if
MeritPPP has the same bug as EPPPD, and if we can get someone at Merit to fix
it.
I'll give it a try and inform you.
Regards, Vlad
Date: Wed, 15 Oct 1997 23:41:36 +0100 (GMT+0100)
From: Toni <tonilop@redestb.es>
Subject: Re: Class 1 PPP drivers under DOS vs true Ethernet
Hello Vladimir,
I already knew of the Dospppd + Kermit problem, as some users reported the
failure to me some time ago. The problem is corrected now, there is a pre-beta
of Dospppd v0.6 package at the following URL:
http://www.redestb.es/personal/tonilop/dosppp06.zip
Please don't redistribute this package, as it is incomplete and subject to
changes before I release the publicly available one. However, I'm interested in
knowing how well it performs, or if it has bugs, so let me know how it goes in
case you have the time for testing it. You may want to inform Joe Dupnik also.
A summary of what I've discovered and the solution follows:
Real Ethernet ARP reply packets include the source address (the address for the
machine which is responding to the ARP request) in two places, one is in the
Ethernet header source address field, and the other is in the ARP packet source
hardware address field.
It seems that most applications look at the source hardware field of the ARP
packet, so the ARP emulation in Dospppd was filling only this one with the fake
Ethernet address.
On the other hand, Kermit seems to look at the Ethernet header source address
field, or maybe it is doing a consistency check by ensuring that both fields
contain the same value (I don't know for sure, maybe Mr. Dupnik would give some
insight on this issue).
The solution was to put the same fake Ethernet address in both fields, Kermit
started to work fine after that. It is not a Kermit problem though, an Ethernet
emulation should fill both fields in order to be acurate. The fact that most
other WATTCP applications worked doesn't imply that Kermit was wrong, maybe
WATTCP developers decided to relax the Ethernet address checks after detecting
that the stack didn't work with Ethernet emulation drivers.
Best regards.
Toni