home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20031118-20041115
/
000422_fdc@columbia.edu_Thu Oct 21 11:14:50 2004.msg
< prev
next >
Wrap
Internet Message Format
|
2004-11-14
|
3KB
Path: newsmaster.cc.columbia.edu!not-for-mail
From: Frank da Cruz <fdc@columbia.edu>
Newsgroups: comp.protocols.kermit.misc
Subject: Re: Kermit Script Output Out of Order
Date: 21 Oct 2004 15:13:51 GMT
Organization: Columbia University
Lines: 40
Message-ID: <slrncnfkhf.lng.fdc@sesame.cc.columbia.edu>
References: <cl29ql$dm7$1@zcars0v6.ca.nortel.com> <slrncnan01.3s1.fdc@sesame.cc.columbia.edu> <cl3tc5$5n0$1@zcars0v6.ca.nortel.com> <_mfdd.83561$Ot3.22349@twister.nyc.rr.com> <cl4k1s$97u$1@zcars0v6.ca.nortel.com> <slrncnddme.qmr.fdc@sesame.cc.columbia.edu> <cl6rqi$mo0$1@zcars0v6.ca.nortel.com>
Reply-To: fdc@columbia.edu
NNTP-Posting-Host: sesame.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 1098371631 11768 128.59.59.56 (21 Oct 2004 15:13:51 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 21 Oct 2004 15:13:51 GMT
User-Agent: slrn/0.9.8.0 (SunOS)
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:15206
On 2004-10-20, Robert Simmons <robertls@nortelnetworks.com> wrote:
: I tried the fork with the stdout and stdin unbuffered. But this had no
: effect. Probably because as you wrote the echos and help text output is
: buffered internally in kermit. Is there any way to flush the buffer for
: echo and help text? Or set it so that these are not buffered?
:
There is no universal setting to make all output the same (buffered or
unbuffered). This could be added, but it would be a very big job.
Well, there IS one thing you might try. If you can build from source,
you can recompile as follows:
make clean
make linux (or whatever) "KFLAGS=-DNONOSETBUF"
This is supposed to make all output unbuffered. You probably won't like
the result, but at least it should be synchronized. (I can't guarantee this
will work, as some of the setbuf() calls are pretty deeply embedded in
#ifdefs, but failing that, just add "setbuf(stdout,NULL);" to some part of
sysinit() in ckutio.c, that is not conditionally compiled.)
Let us know the results; if favorable, I suppose this could be converted
from a compile-time option to a runtime setting.
: Thanks for the advise on the script. I'll use it. The book I have is
: outdated. I'll use the website for more up to date info.
:
: As for the looping send: We are sending the same file as a soak test of the
: connection so we don't care about the content, just that the connection
: stays up and the transfers complete.
:
Here's an undocumented feature you might want to use for testing and
benchmarking:
SEND /CALIBRATE:xxx
This creates a random-content uncompressable dummy file on the fly for
sending, of length xxx K bytes.
- Frank