httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Marc Slemko <ma...@worldgate.com>
Subject Re: Graceful child death? (fwd)
Date Mon, 19 Oct 1998 01:43:10 GMT
Hey, that is cheating.  Musta missed that.  Will have to be done another
way if we ever have a setup that allows shared config data, but that is
easy.

Thanks.

On Sun, 18 Oct 1998, Dean Gaudet wrote:

> ap_child_terminate()
> 
> Dean
> 
> On Sun, 18 Oct 1998, Marc Slemko wrote:
> 
> > There is no way to do this.
> > 
> > I think this is a useful feature.
> > 
> > There are two ways it could be added: either the ability to set a flag
> > that the main request reading loop checks or the ability to call
> > clean_child_exit.
> > 
> > ---------- Forwarded message ----------
> > Date: Fri, 16 Oct 1998 11:32:23 +1100
> > From: Daniel Austin <dan@austlii.edu.au>
> > To: apache-modules@covalent.net
> > Subject: Graceful child death?
> > Resent-Date: Thu, 15 Oct 1998 20:42:37 -0500 (CDT)
> > Resent-From: apache-modules@covalent.net
> > 
> > Hi,
> > 
> > I'm writing an Apache module that calls a database API directly.  This
> > database API was never really designed to be used in this way. After
> > some problems I've been advised by the developer that if his API ever
> > returns an error code marked as being "fatal" then my Apache process
> > should quit and not make any further calls on the API.
> > 
> > I guess I could just call exit(0) to force a child process to quit, but
> > then other modules exit handlers (cleanup) code would never be called.
> > And there's probably Apache cleanup code (?). So my question is -- is
> > there any way to indicate to Apache that, after this request is done,
> > the child should exit and a new one spawned?
> > 
> > Dan
> > -- 
> > Daniel Austin (dan@austlii.edu.au)
> > Primary Materials and Programming Support, AustLII
> > 
> > 
> 


Mime
View raw message