Mailing-List: contact bugs-help@httpd.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Apache HTTPD Bugs Notification List" <bugs@httpd.apache.org>
Date: 17 Oct 2002 07:30:31 -0000
Message-ID: <20021017073031.11188.qmail@nagoya.betaversion.org>
From: bugzilla@apache.org
To: bugs@httpd.apache.org
Cc: 
Subject: DO NOT REPLY [Bug 11983]  -
    Connection closed prematurly on CGI requests

DO NOT REPLY TO THIS EMAIL, BUT PLEASE POST YOUR BUG 
RELATED COMMENTS THROUGH THE WEB INTERFACE AVAILABLE AT
<http://nagoya.apache.org/bugzilla/show_bug.cgi?id=11983>.
ANY REPLY MADE TO THIS MESSAGE WILL NOT BE COLLECTED AND 
INSERTED IN THE BUG DATABASE.

http://nagoya.apache.org/bugzilla/show_bug.cgi?id=11983

Connection closed prematurly on CGI requests

hb@websale-ag.de changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
            Version|2.0.40                      |2.0.43


------- Additional Comments From hb@websale-ag.de  2002-10-17 07:30 -------
Version 2.0.43 does not solve this problem, in fact it seems to have become 
worse.

Now Apache doesn't add a "Content-Length" header field to the output of CGI 
programs anymore *and* ignores the "Keep-Alive" in Http/1.0 Requests if 
no "Content-length" is present in the HTTP header.

Now you don't even need a special HTTP client to reproduce the problem. All you 
need is a CGI program that produces a lengthy output and doesn't insert 
a "Content-length" field and an older Netscape version, for example 4.61. 
Create a frameset that calls two instances of the CGI program and you get 
an "connection reset by peer" error almost every time you try open the frameset.

---------------------------------------------------------------------
To unsubscribe, e-mail: bugs-unsubscribe@httpd.apache.org
For additional commands, e-mail: bugs-help@httpd.apache.org