----- Original Message -----
From: "Mandler, George" <[EMAIL PROTECTED]>
To: <[EMAIL PROTECTED]>; "Mandler, George" <[EMAIL PROTECTED]>;
"'Paul Kinnucan'" <[EMAIL PROTECTED]>
Cc: <[EMAIL PROTECTED]>
Sent: Wednesday, May 30, 2001 3:59 PM
Subject: RE: Any solutions for the "Gave up waiting for Emacs to connect to
SIO port: " Win2K woes


> Thanks Javier.
> OK I totally went into brain-fry mode and blew it.  For the record I am
now
> working under SP2.
> My mistake was that I was doing a JDEbug->LaunchApp , but not doing a
> JDE->DebugApp.  Re-reading
> the JDEbugUG for the nth time I see the sentence "At this point, you
should
> set one or more breakpoints in the application and select the Run command
> from the JDEbug menu".
>
> However I'm still convinced I wouldn't have worked under SP1, because I
> never got to the connection even with a jde-bug-sio-connect-delay equal to
> 1.   So for whatever reason installing SP2 either fixed or hid the real
> problem.
> So Paul when you get your dandy Windoze2K machine try it out with SP1 in
> your copious spare time and see what happens.
>

Hi George,

Thanks for persisting. For what it's worth, JDE 2.2.7.1 incorporates a
change contributed by Eugene Gavrilov that may have helped to solve the
problem. The way JDEbug works is that it creates a socket, passes the port
number of the socket back to Emacs, and then waits for Emacs to connect to
the socket at the specified port. Previously, JDEbug created the socket and
started waiting on one thread while another thread passed the socket port
back to Emacs. In the latest version, the same thread that creates the port
passes the port number back to Emacs BEFORE it starts waiting for a connect
(i.e., before it executes socket.accept(). This cleared the hangup on the HP
system that Eugene uses. I'm hoping that it may solve the problem for
others.

- Paul

Reply via email to