Received: (qmail 16478 invoked by uid 65534); 13 Mar 2000 10:37:53 -0000
Message-Id: <20000313103753.16477.qmail@locus.apache.org>
Date: 13 Mar 2000 10:37:53 -0000
From: Frank Schulze <fschulze@mrsfs1.mw.tu-dresden.de>
Reply-To: fschulze@mrsfs1.mw.tu-dresden.de
To: submit@bugz.apache.org
Subject: Cancelation of SSI-request doesn't stop server job
X-Send-Pr-Version: 3.110

>Number:         5867
>Category:       mod_include
>Synopsis:       Cancelation of SSI-request doesn't stop server job
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    apache
>State:          closed
>Quarter:        
>Keywords:       
>Date-Required:
>Class:          sw-bug
>Submitter-Id:   apache
>Arrival-Date:   Mon Mar 13 02:40:00 PST 2000
>Closed-Date:    Mon Mar 13 12:41:03 PST 2000
>Last-Modified:  Mon Mar 13 12:41:03 PST 2000
>Originator:     fschulze@mrsfs1.mw.tu-dresden.de
>Release:        1.3.0
>Organization:
>Environment:

uname -a
SunOS rks6 5.5.1 Generic_103640-32 sun4u sparc SUNW,Ultra-2

>Description:

There's a *.shtml-page that includes the <!--#exec...--> ...ution of
some shell script.  It takes the script a while to do any 
sophisticated things.  The user (client) is impatient an cancels 
the request (e.g. moves towards another webpage):

To me it seems that the execution of the shell script on the server
is completed - and not terminated - although its output will be 
rejected anyway.

>How-To-Repeat:
Easily.  8->
>Fix:

Is it possible to catch any signal about the cancellation of the 
request by the client, and kill the job running on the server?

Is it possible to define for SSI (and for CGI-scripts too) an
upper limit of CPU-time the may consume ... and to kill them
automaticly, if they exceed that limit?

>Release-Note:
>Audit-Trail:

From: Marc Slemko <marcs@znep.com>
To: fschulze@mrsfs1.mw.tu-dresden.de, Apache bugs database <apbugs@apache.org>
Cc:  
Subject: Re: mod_include/5867: Cancelation of SSI-request doesn't stop server
 job
Date: Mon, 13 Mar 2000 08:29:44 -0700 (MST)

 On 13 Mar 2000, Frank Schulze wrote:
 
 > Is it possible to catch any signal about the cancellation of the 
 > request by the client, and kill the job running on the server?
 
 Erm... there is no such signal to catch, unless it outputs "enough" data,
 and then Apache will kill it.
 
 > 
 > Is it possible to define for SSI (and for CGI-scripts too) an
 > upper limit of CPU-time the may consume ... and to kill them
 > automaticly, if they exceed that limit?
 
 Check the docs for the RLimit* directives.
 
State-Changed-From-To: open-closed
State-Changed-By: marc
State-Changed-When: Mon Mar 13 12:41:02 PST 2000
State-Changed-Why:
Current behaviour is correct.
>Unformatted:
 [In order for any reply to be added to the PR database, you need]
 [to include <apbugs@Apache.Org> in the Cc line and make sure the]
 [subject line starts with the report component and number, with ]
 [or without any 'Re:' prefixes (such as "general/1098:" or      ]
 ["Re: general/1098:").  If the subject doesn't match this       ]
 [pattern, your message will be misfiled and ignored.  The       ]
 ["apbugs" address is not added to the Cc line of messages from  ]
 [the database automatically because of the potential for mail   ]
 [loops.  If you do not include this Cc, your reply may be ig-   ]
 [nored unless you are responding to an explicit request from a  ]
 [developer.  Reply only with text; DO NOT SEND ATTACHMENTS!     ]