home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Columbia Kermit
/
kermit.zip
/
newsgroups
/
misc.19990725-20000114
/
000366_news@columbia.edu _Fri Dec 17 19:05:43 1999.msg
< prev
next >
Wrap
Internet Message Format
|
2020-01-01
|
2KB
Return-Path: <news@columbia.edu>
Received: from newsmaster.cc.columbia.edu (newsmaster.cc.columbia.edu [128.59.59.30])
by watsun.cc.columbia.edu (8.8.5/8.8.5) with ESMTP id TAA15839
for <kermit.misc@watsun.cc.columbia.edu>; Fri, 17 Dec 1999 19:05:43 -0500 (EST)
Received: (from news@localhost)
by newsmaster.cc.columbia.edu (8.8.5/8.8.5) id SAA02697
for kermit.misc@watsun.cc.columbia.edu; Fri, 17 Dec 1999 18:37:01 -0500 (EST)
X-Authentication-Warning: newsmaster.cc.columbia.edu: news set sender to <news> using -f
From: jrd@cc.usu.edu (Joe Doupnik)
Subject: Re: Timeout / retry on send
Message-ID: <qxHGNBSm21fQ@cc.usu.edu>
Date: 17 Dec 99 14:29:22 MDT
Organization: Utah State University
To: kermit.misc@columbia.edu
In article <s5l4gmh1qrs137@corp.supernews.com>, "Steve Baus" <steve@baus-systems.com> writes:
> I am using DOS ver 3.15 to automate a send and receive over a modem.
> Computer A is in server mode waiting for computer B to send a file. Once it
> gets the file, computer A does some processing, creates a file, which may be
> a couple of hundred K in size, and tries to send that file. If there is a
> problem with this send, I would like computer A to abort within a maximum
> time or retries.
>
> My thought is since the file transfer may take multiple minutes, I should
> have it abort on too many retries like 20. If I could set the time between
> retries to be say 2 seconds then in theory if I get a dropped connection it
> should abort in 40 seconds. Looking through my Using MS-DOS Kermit book and
> C-Kermit I would think in my script I would simply have to set retries 20
> and set send timeout 2 and that should do it. When I test this concept,
> every retry seems to take longer and longer, the first few zip right along
> then it takes a couple of seconds between and each one is longer than the
> last and when I set retries to 20, it abouts on retry 61, when I set retry
> to 8 it aborts on 25.
>
> Any thoughts?
> Steve
----------
Retry intervals are purposefully made exponentially longer because
the range of comms channel characteristics is very wide indeed. Usually
accepting only two or maybe three retries is fairly safe because one packet
may be dropped but not likely two or three. That will shorten the window
of detecting no-client. Also your modem will have hung up the connection
if the client side does so first.
Joe D.