home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
kermit.columbia.edu
/
kermit.columbia.edu.tar
/
kermit.columbia.edu
/
newsgroups
/
misc.20010306-20010921
/
000109_fdc@watsun.cc.columbia.edu_Thu May 10 12:59:35 EDT 2001.msg
< prev
next >
Wrap
Text File
|
2001-09-20
|
4KB
|
98 lines
Article: 12399 of comp.protocols.kermit.misc
Path: newsmaster.cc.columbia.edu!not-for-mail
From: fdc@watsun.cc.columbia.edu (Frank da Cruz)
Newsgroups: comp.protocols.kermit.misc,comp.unix.programmer
Subject: Re: Need help with Mac OS X 10.0.2
Date: 8 May 2001 11:27:33 -0400
Organization: Columbia University
Lines: 81
Message-ID: <9d9395$rt3@watsun.cc.columbia.edu>
References: <9d6t71$lp0$1@watsol.cc.columbia.edu> <9d90vh$ot8$1@jake.esu.edu>
NNTP-Posting-Host: watsun.cc.columbia.edu
X-Trace: newsmaster.cc.columbia.edu 989335657 15358 128.59.39.2 (8 May 2001 15:27:37 GMT)
X-Complaints-To: postmaster@columbia.edu
NNTP-Posting-Date: 8 May 2001 15:27:37 GMT
Xref: newsmaster.cc.columbia.edu comp.protocols.kermit.misc:12399 comp.unix.programmer:128388
In article <9d90vh$ot8$1@jake.esu.edu>,
David H. Hutchens <hutchens@cs.millersville.edu> wrote:
: Thanks for your response, Frank.
:
: Yes, it was working in Mac OS X public beta.
:
: I have not tried recompiling. I'll can give it a shot.
:
Yes, please do.
: I did try setting flow control rts/cts. ( I noticed that oddity after
: I sent the previous message). Maybe I had turned it off in public beta
: because:
:
: Tuning on RTS/CTS resulted in a message that says:
: Warning -SET FLOW RTS/CTS is in effect but modems CTS signal is off.
: Disabling flow control temporarily during dialing.
:
Did RTS/CTS work in the public Beta?
: But then.... I was able to log in on the remote machine. Oddly, I
: got a system beep along with the echo of the first 4 or 5 characters
: of my login name. Then things seemed to work OK.
:
: Is it OK to ignore the above warning? Does it still enable RTS/CTS
: after the connection? How can I tell?
:
The message is worrisome. It probably means that the API for hardware
flow control and/or modem-signal testing changed or broke. Of course I
have no way of knowing that, nor of knowing what they changed to.
: But when I tried to download a file, I got errors.
:
Because flow control was not working.
: I set the buffer size down until I got to 512 and that allowed
: files to download. Even a 1 MB binary file downloaded OK, with no
: reported errors. The download speed was over 2700 CPS. The modem on
: the other end is a 28K modem, so that seemed quite reasonable.
:
A good thing about Kermit protocol is that it can be made to work even
when important parts of the connection are broken. But it would be
preferable to make flow control work right.
: My wife notes that our old 14400 zyxel modem connected to a
: NeXTStation has been having some phone line problems in the last two
: weeks so perhaps we have a more noisy phone line than when I tried
: with the public beta.
:
No, the problem in your case is that Kermit can't sense the CTS line
from the modem. Assuming that you had been using RTS/CTS before the
OS upgrade, it indicates a software change, bug, or incompatibility.
: Finally, I notice that when I quit kermit, it says the modem may
: still be active and I tell it to quit anyway.
:
Another indication that the API for accessing modem signals is broken.
: But it apparently did not hang up. How can I force it to hang
: up the phone?
:
If you exit from Kermit or give it the HANGUP command, it hangs up by
dropping DTR. But since the modem-signal API isn't working on Mac OS X,
probably DTR isn't really being dropped. But Kermit always gives you
another way to work around broken components:
set modem hangup-method modem-command
hangup
This makes Kermit hang up by sending +++ and then ATH0.
So how do we get the real problem fixed? First try recompiling and see
if that clears things up. If not, it would be best if I could get
telnet/ftp access to your Mac, or another one that has the same OS
level so I can poke around the header files, man pages, etc, and see
if I can get at the modem signals. (Of course even then I can't actually
see/hear what's happening, but it's better than nothing.) Failing that,
somebody who has Mac OS X 10.2 and a modem will have to make the code
work (I can point out the relevant modules and functions).
- Frank