Updates to the status module

[Jim Jagielski <ji...@jaguNET.com>]

I'm looking into increasing the amount of valuable data available
from the status module but when STATUS is not used. I've noticed that
there are 2 places where ap_update_child_status is only called if
STATUS is defined but the reasons are unclear: one is during DNS
lookups and the other during logging of the request. I think that
knowing when processes are in this state is worthwhile even if
STATUS is not defined, as well as the fact that the overhead in doing
ap_update_child_status (again, with STATUS not defined) should be
minimal when one considers that DNS and logging are expensive anyway.

Unless anyone complains, or sees something I don't, I'll allow
ap_update_child_status even without STATUS (I won't much around with
other things like the ap_time_process_request, which doesn'tt make
sense w/o STATUS)
-- 
===========================================================================
   Jim Jagielski   |||   jim@jaguNET.com   |||   http://www.jaguNET.com/
            "That's no ordinary rabbit... that's the most foul,
            cruel and bad-tempered rodent you ever laid eyes on"

Re: Updates to the status module

[Dean Gaudet <dg...@arctic.org>]

It would be really nice if you could make the STATUS stuff config-time
selectable... have a global, which is controlled by a directive in
mod_status, default it to off... so that when there's no mod_status loaded
you don't waste the cycles.  And mod_status can default it to on in its
init function.

This would really help the folks doing distributions.

Dean

On Fri, 7 Aug 1998, Jim Jagielski wrote:

> I'm looking into increasing the amount of valuable data available
> from the status module but when STATUS is not used. I've noticed that
> there are 2 places where ap_update_child_status is only called if
> STATUS is defined but the reasons are unclear: one is during DNS
> lookups and the other during logging of the request. I think that
> knowing when processes are in this state is worthwhile even if
> STATUS is not defined, as well as the fact that the overhead in doing
> ap_update_child_status (again, with STATUS not defined) should be
> minimal when one considers that DNS and logging are expensive anyway.
> 
> Unless anyone complains, or sees something I don't, I'll allow
> ap_update_child_status even without STATUS (I won't much around with
> other things like the ap_time_process_request, which doesn'tt make
> sense w/o STATUS)
> -- 
> ===========================================================================
>    Jim Jagielski   |||   jim@jaguNET.com   |||   http://www.jaguNET.com/
>             "That's no ordinary rabbit... that's the most foul,
>             cruel and bad-tempered rodent you ever laid eyes on"
>