home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20020314-20021006
/
000424_jfathman@aol.com_Sun Oct 6 11:03:02 EDT 2002.msg
< prev
Wrap
Text File
|
2002-10-06
|
3KB
|
70 lines
Article: 13760 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!panix!newsfeed.mathworks.com!news-hog.berkeley.edu!ucberkeley!newsfeed.stanford.edu!postnews1.google.com!not-for-mail
From: jfathman@aol.com (Jim)
Newsgroups: comp.protocols.kermit.misc
Subject: YModem
Date: 5 Oct 2002 19:12:51 -0700
Organization: http://groups.google.com/
Lines: 52
Message-ID: <6dfb5332.0210051812.463a13dc@posting.google.com>
NNTP-Posting-Host: 198.81.21.23
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Trace: posting.google.com 1033870371 30440 127.0.0.1 (6 Oct 2002 02:12:51 GMT)
X-Complaints-To: groups-abuse@google.com
NNTP-Posting-Date: 6 Oct 2002 02:12:51 GMT
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:13760
Hello,
For many years, I have been using YModem for serial and modem based
file transfers in an embedded product. Now I am adding a TCP/IP
network interface, while preserving the YModem functionality.
I have the network interface working, but I have encountered a very
odd problem. The YModem support in Kermit 95+ v2.0 and three other
commercial telnet/YModem products I have tried all produce a YModem
initial packet header that, based on my reading of the protocol spec,
is non-compliant.
Chuck Forsberg's spec (granted it is old, dated 9-11-86, but that is
the newest I have found) specifies the format of the initial YModem
packet as:
SOH 00 FF foo.c NUL[123] CRC CRC
But when I use Kermit or the other three commercial telnet/YModem
products to send a file, they *all* produce an additional 0xFF before
the file name.
To rule out my embedded software, I wrote a simple socket server under
Linux that accepts connection requests, issues 'C' prompts to make
Kermit YModem send happy, and hex dumps the incoming bytes from the
Kermit YModem send. As with my embedded product, Kermit YModem send
produces:
[0001] 0x01 = . <-- SOH
[0002] 0x00 = . <-- sequence number
[0003] 0xff = . <-- seq num complement
[0004] 0xff = . <-- !! what is this ? !!
[0005] 0x6a = j <-- start of file name
[0006] 0x6d = m
[0007] 0x66 = f
[0008] 0x2e = .
[0009] 0x6a = j
[0010] 0x6d = m
[0011] 0x66 = f
[0012] 0x00 = .
<etc.>
Kermit and three other commercial products can't all be wrong. What
am I missing here? Is there a newer YModem spec that I just can't
find? Are the rules for YModem over TCP different somehow than YModem
over serial/modem?
Any guidance would be most helpful.
Thanks.
Jim