Mail Archives: cygwin/2007/05/31/16:31:36
On Thu, May 31, 2007 at 04:27:00PM -0400, Larry Hall (Cygwin) wrote:
>Christopher Faylor wrote:
>> There seems to be some confusion about the CYGWIN tty setting.
>>
>> The only time it should have any effect is when there is a console involved.
>> In that scenario, what effectively happens is that CYGWIN opens up a pty
>> and pretends that the pty is a linux-like tty.
>>
>> Setting CYGWIN=tty for servers should be a no-op. Ditto, having it set
>> while operating under a pty or if there is no console associated with the
>> session.
>>
>> So, if setting CYGWIN=tty actually causes a problem in an ssh session
>> then that is probably a Cygwin bug. Until that problem has become evident
>> lets not assume that there is a problem.
>>
>> I don't mean to be picking on you Larry. You are not the only person
>> I've seen opine about CYGWIN=tty . I remain unconvinced that setting it
>> would present a problem other than the standard problem of running pure
>> Windows programs under a Cygwin pty.
>
>The confusion may come from me. ;-)
>
>So are you saying that all Cygwin services effectively use ptys?
>
>Rereading the original post, I note that KenLee only said 'cmd' was started
>once in the OpenSSH session. I also assumed he invoked 'ssh' from a
>console, which may make a difference if the answer to the above question
>is "no". But that assumption is unsubstantiated by the available
>information. ;-)
If he ssh'ed to somewhere interactively then there is a pty involved on
the remote end where the program was being run so that's what would have
caused any problem, IMO, regardless of the CYGWIN tty setting on the
local end. In that scenario, since ssh is a cygwin program it shouldn't
care about the CYGWIN=tty setting one way or the other.
cgf
--
Unsubscribe info: http://cygwin.com/ml/#unsubscribe-simple
Problem reports: http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ: http://cygwin.com/faq/
- Raw text -