W3C home > Mailing lists > Public > www-jigsaw@w3.org > November to December 1996

Is STDOUT flushed?

From: Anselm Baird_Smith <abaird@www43.inria.fr>
Date: Mon, 25 Nov 1996 13:37:02 +0100 (MET)
Message-Id: <199611251237.NAA21906@www43.inria.fr>
To: bpm@techapp.com
Cc: www-jigsaw@w3.org
bpm@terraweb.com writes:
 > 
 > I am in the process of setting up Fast CGI applications to run under
 > Jigsaw.  (This is an interim step until I can get the apps
 > re-written!)  I am using the fast-cgi development kit from the
 > www.fastcgi.com (Open Market) site.  They have a cgi-fcgi application
 > that is a cgi to fcgi gateway.  You can set up a CgiResource that calls
 > a script that invokes the cgi-fcgi gateway and run the fastCGI
 > application.  It works really great, except for one item that has me
 > baffled.  On the initial connection to the fastCGI application, the
 > browser waits and spins.  The browser is waiting for an EOF from the
 > server.  If I click on the "Stop" menu button, I get a:
 > "
 > > 
 > 
 > 
 > Transfer interrupted!
 > "
 > on the browser screen.  But I can go ahead and fill out the form,
 > submit and it works fine.  ANY OTHER connections to the fastCGI
 > application works just fine.  It is just the initial connection.
 > 
 > NCSA 1.5 used to have this same condition until Brandon Long noticed
 > that he was closing the file handle without flushing.
 > 
 > I looked at the w3c.jigsaw.http.Client.java emitReply method, but it
 > looks like output is being flushed.

I think it is flushed, yes. The problem might be elsewhere, though,
and I have work a lot on the Client/ClientPool classes...

Providing a thread dump might help us,

Anselm.
BTW: Do you plan to make your resource available for free ? I would me
more then happy to include it in next release ;-)
Received on Monday, 25 November 1996 07:37:33 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 9 April 2012 12:13:25 GMT