home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20021006-20030409
/
000177_fdc@columbia.edu_Wed Dec 11 15:11:14 EST 2002.msg
< prev
next >
Wrap
Text File
|
2003-04-08
|
3KB
|
79 lines
Article: 13958 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!news.columbia.edu!news-not-for-mail
From: fdc@columbia.edu (Frank da Cruz)
Newsgroups: comp.protocols.kermit.misc
Subject: Re: remote dir
Date: 11 Dec 2002 15:10:49 -0500
Organization: Columbia University
Lines: 62
Message-ID: <at8649$2qj$1@watsol.cc.columbia.edu>
References: <suMJ9.34$CC2.77960@news.uswest.net>
NNTP-Posting-Host: watsol.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1039637450 7420 128.59.39.139 (11 Dec 2002 20:10:50 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 11 Dec 2002 20:10:50 GMT
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:13958
In article <suMJ9.34$CC2.77960@news.uswest.net>,
Frank Huang <huang@monair.com> wrote:
: Hi Everyone:
: I am using TurboPower Async ActiveX.
: On the client, I sent a 'G' packet with 'D' as the Data field.
: Then on 3.15 Server, it tries to send $KERMIT$.TMP over, but when I
: tries to receive the file, it logs
: the following messages.
: On the server and client, I have the Block-check-type set to be CRC
: If I sent a 'G' packet with 'FINISH' or 'LOGOUT' as the Data field, then
: it works find.
:
We don't provide a debugging service for competitors who sell unlicensed
implementations of our protocol. MS-DOS Kermit 3.15 includes a correct
implementation that can be used on DOS and on Windows 3.x. MS-DOS Kermit
is not supported on 32-bit Windows, which has its own Kermit software,
Kermit 95:
http://www.columbia.edu/kermit/k95.html
But that's not the problem in this case.
Your log shows:
: Rpack: ^A, GD READ.ME6^M
: Spack: ^A5 S~( @-#Y3~^?5% ___E#^M
: Rpack: ^A0 Yp% @-#Y3~ ! Z^M
: Spack: ^A)!Xdir !R>^M
: Rpack: ^A# N&0
: <Bad checksum>
:
MS-DOS Kermit sent a correct X packet with the three-byte block check (CRC)
that was successfully negotiated. The other Kermit could not read it and
sent a NAK. Furthermore the NAK itself is malformed (it claims to have a
1-byte checksum, but in fact has a two byte one; even if you ignore the
spurious second byte, the first one is wrong):
: Spack: ^A'!Xdir "^M
: Rpack: 6^M^A# N&0
: <Bad checksum>
:
Since it keeps happening, it's not because of transmission errors; it's
because the other Kermit program is totally broken. Maybe it will work
better with 1-byte checksums.
If you want to embed Kermit protocol in a Windows application you are
developing, perhaps you should come to us for it since we know the protocol
and actually go to some lengths to implement it correctly and efficiently.
See:
http://www.columbia.edu/kermit/k95faq.html#embedding
and:
http://www.columbia.edu/kermit/ek.html
Better still, companies that make products like the one you are trying to
use should come to us and license a supported implementation, instead of
torturing their customers as well as end-users (not to mention us and the
readers of this newsgroup) with broken and/or substandard implementations.
- Frank