home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.19950329-19950528
/
000316_news@columbia.edu_Fri May 5 05:48:47 1995.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
5KB
Received: from apakabar.cc.columbia.edu by watsun.cc.columbia.edu with SMTP id AA26851
(5.65c+CU/IDA-1.4.4/HLK for <kermit.misc@watsun.cc.columbia.edu>); Fri, 5 May 1995 18:35:29 -0400
Received: by apakabar.cc.columbia.edu id AA15380
(5.65c+CU/IDA-1.4.4/HLK for kermit.misc@watsun); Fri, 5 May 1995 18:35:26 -0400
Path: news.columbia.edu!panix!news.mathworks.com!gatech!swrinde!cs.utexas.edu!news.cs.utah.edu!cc.usu.edu!jrd
From: jrd@cc.usu.edu (Joe Doupnik)
Newsgroups: comp.protocols.kermit.misc
Subject: Re: C-Kermit 5A(190) server mode problem
Message-Id: <1995May5.114847.49869@cc.usu.edu>
Date: 5 May 95 11:48:47 MDT
References: <BAM.95Apr21171236@wcl-l.bham.ac.uk><B.A.MCCAULEY.95May4095228@wcl-l.bham.ac.uk><1995May4.173322.49759@cc.usu.edu> <B.A.MCCAULEY.95May5142149@wcl-l.bham.ac.uk>
Organization: Utah State University
Lines: 94
Apparently-To: kermit.misc@watsun.cc.columbia.edu
In article <B.A.MCCAULEY.95May5142149@wcl-l.bham.ac.uk>, B.A.McCauley@bham.ac.uk writes:
> In article <1995May4.173322.49759@cc.usu.edu> jrd@cc.usu.edu (Joe Doupnik) writes:
>>In article <B.A.MCCAULEY.95May4095228@wcl-l.bham.ac.uk>, B.A.McCauley@bham.ac.uk writes:
>>> In article <BAM.95Apr21171236@wcl-l.bham.ac.uk> B.A.McCauley@bham.ac.uk writes:
>>>>When the terminal is MS-KERMIT 3.13.0 it works fine when the terminal
>>>>is C-Kermit 5A(190) then the generic finish command is ignored (it's
>>>>not even logged in the packet log). My software first sends a "N"
>>>>packet (in case the server had send a "S" and I had missed it)
>>>>followed by an "E" packet and finally sends an "I" packet, after which
>>>>C-Kermit will accept the "GF" packet.
>> Hmmm. I don't think you want to do all that. First, a Kermit
>>server sits around waiting for a client to start a session and make
>>a request. It does NOT send S packets while waiting. It may send NAKs
>>now and then to stimulate old clock-less Kermit clients. E packets are
>>powerful things because they terminate a file transfer operation; don't
>>use unless *really* needed.
>
> I know all this and I agree that my client may not take the ideal
> approach to attempting to recover the situation. This is not the issue
> here, what I want to know is why C-Kermit is ignoring the GF packet in
> the first place.
>
>> C Kermit did see the GF packet, ACK'd it, and logged both, as
>>your C Kermit packet log below clearly shows.
>
> No it does not. Look again.
>
>> Your E packet also shows.
>
> And like I said this was sent to force a reset *after* the server failed
> to respond to the GF packet. After this reset the next GF packet worked.
>
>> Now then, I've lost track of what was the real question.
>
> What happened to the first GF packet? Not the one I sent after
> resetting the server.
>
>>>>Here is C-Kermit's packet.log (long data packet truncated).
>>>>
>>>>r-00-00-0 I~\ @-#Y2 J 5S#
>>>>s-00-00-5 Y~* @-#Y2~^!5$0___CX
>>>>r-00-01-% GD S
>>>>s-00-01-5 S~* @-#Y2~^!5$0___AP
>>>>r-00-01-0 Y~\ @-#N2 J 5S(
>>>>s-01-01-*!Xls -l );
>>>>r-01-00-$!Y">
>>>>s-02-00-2"A."U1"#AMJ*!A@ -T
>>>>r-02-00-$"Y"?
>>>>s-03-04- #D08Rtotal 5349#M#Jdrwxr-xr-x 2 bam staff 1024 Apr 5 12:54 backup.home#M#Jdrwxr-xr-x 2 bam staf
>>>>r-03-08-$#Y"@
>>>>s-04-08-$$Z"B
>>>>r-04-09-$$Y"A
>>>>s-05-09-$%B"+
>>>>r-05-09-$%Y"B
>
> This is where I sent the GF packet. What happened to it? (BTW this is
> reproducable, it's not a comms problem).
>
>>>>r-00-19-# N3
>>>>#-00-19-<echo:ignored>
>
> What does this mean? This looks nothing like the last packet!
Well, it means two things. First, the obvious one is a NAK
was received and C Kermit declared it to be an echo of what it sent.
This makes more sense as we look at the receive packet sequence numbers,
the second pair of digits in the debug log. There is a gap of about
ten of them above and that suggests C Kermit went back to being a quiet
server and sent NAKs out now and then. The log picks up one echo and
CK ignores it. Could your program be sending the NAK by any chance?
Taken together with your earlier comments on flow control
problems and extra padding etc (see the MS-DOS Kermit log for details)
it does appear that the wire part of the circuit is not behaving well.
Given that, CK may never have heard your first GF packet or if it did
then maybe the packet was mangled to be a non-packet and hence not
logged by CK. MSK logs everything, rational or not, as we have noticed.
Beyond this I think Frank da Cruz needs to plug in his crystal
ball and beam it in this direction.
Joe D.
> OK my client is desparate now so it sends an E and I packet to reset
> the server.
>
>>>>r-00-29-E EToo many retries (expecting SXBFY)U
>>>>r-00-30-0 I~\ @-#Y2 J 5S#
>>>>s-00-30-5 Y~* @-#Y2~^!5$0___CX
>
> OK the server is listening again so let's try the GF command.
>
>>>>r-00-31-$ GF4
>>>>s-00-31-# Y>
>
> This time the GF packet worked.