httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Marr <gr...@alum.wpi.edu>
Subject RE: logging APR errors
Date Mon, 17 Jul 2000 20:42:24 GMT
At 01:26 PM 07/17/2000 -0700, rbb@covalent.net wrote:
>On Mon, 17 Jul 2000, Greg Marr wrote:
>
> > At 12:57 PM 07/17/2000 -0700, rbb@covalent.net wrote:
> > > > This all assumes that the callback isn't intelligent enough to make the
> > > > decision on its own whether or not the error should be logged.
> > >
> > >Then show me how to make the callback stack intelligent enough to do
> > >it.  Nobody has been able to show a design with a sufficiently 
> intelligent
> > >callback yet.
> >
> > Nobody has been able to show a need for it yet.
>
>I've shown the need.  Sometimes we log errors and sometimes we don't, all 
>with the same apr function.  Having proven that, we need a logging 
>callback that can determine when to log and when not to log without 
>affecting the main-line code.

I'm assuming that you're talking about ap_stat here.  You have proved that 
sometimes we log an error, and sometimes not.  I agree with that.  However, 
I don't see a need for the extended error information callback to be used 
by ap_stat, and so there's no need to teach the callback when to log the 
error, and when not to log the error.  You have not proved that sometimes 
the same call to the same APR function that has meaningful extended status 
information results in a call to the error callback, and thus the callback 
needs to decide whether or not to log based on some piece of information 
that it doesn't have available.


Mime
View raw message