home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.19941031-19941221
/
000150_news@columbia.edu_Mon Nov 14 15:44:28 1994.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
3KB
Received: from apakabar.cc.columbia.edu by watsun.cc.columbia.edu with SMTP id AA28504
(5.65c+CU/IDA-1.4.4/HLK for <kermit.misc@watsun.cc.columbia.edu>); Tue, 15 Nov 1994 05:40:55 -0500
Received: by apakabar.cc.columbia.edu id AA14178
(5.65c+CU/IDA-1.4.4/HLK for kermit.misc@watsun); Tue, 15 Nov 1994 05:40:53 -0500
Path: news.columbia.edu!sol.ctr.columbia.edu!news.kei.com!hookup!usc!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: The Mac version of Kermit has less features than other OS versions.
Message-Id: <1994Nov14.214428.32803@cc.usu.edu>
Date: 14 Nov 94 21:44:28 MDT
References: <3935bt$111o@fidoii.cc.lehigh.edu> <1994Nov11.183525.77435@kuhub.cc.ukans.edu> <1994Nov14.075525.33863@miavx1>
Organization: Utah State University
Lines: 36
Apparently-To: kermit.misc@watsun.cc.columbia.edu
In article <1994Nov14.075525.33863@miavx1>, kacovert@miavx1.acs.muohio.edu (Kent Covert) writes:
> In article <1994Nov11.183525.77435@kuhub.cc.ukans.edu>, tdsmith@kuhub.cc.ukans.edu writes:
>> Note that I've never seen kermit use more than one window, though.
>
> Actually, I think that the Mac version of Kermit reports this incorrectly.
> I've never seen the Mac version report anything higher than 1 also. But,
> if you issue the STATUS command on the server after a transfer, you'll
> usually find that the server reports using more than 1.
>
-----------
You are both right, but maybe not for the reasons you think.
With Columbia Kermits the receiver shows how many packets have
been recognized as packets and are being processed. If packets arrive
in order with none missing then one at a time is processed as bytes are
read from the comms channel; we see one window slot in use. Other bytes
may still be in the comms channel (includes receiver's port buffer) but
not yet parsed as a packet.
If a packet is damaged or really missing then the out of
order packets are recognized and stashed away while the receiver still
tries to find the desired (but missing) packet. Then the number of
window slots in use grows. Kermit uses "selective repeat" sliding windows
so only the missing packet(s) needs to be retransmitted.
The status command should show the number of window slots
available, rather than just the number used. But I have no Mac to see
what's what on it. MS-DOS Kermit shows the "active out of available"
number on the formatted file transfer display.
The transmitter is given permission to send all window slots
worth of packets. It does so, but between each it takes a very quick
peek at the comms channel to see if enough bytes have arrived to
constitute a possible acknowledgment. If there are enough then it is
read and processed on the fly. Thus the transmitter may show many (two
or more) window slots in use even under ideal conditions as it waits for
ACKs from the other side.
Quiz Friday. Homework: what might a receiver say about a missing
packet? Hint: as little as possible, but...
Joe D.