From: Jim Jagielski <jim@jaguNET.com>
Message-Id: <199809211223.IAA01707@devsys.jaguNET.com>
Subject: Re: More on broken mod_status output
To: new-httpd@apache.org
Date: Mon, 21 Sep 1998 08:23:28 -0400 (EDT)
In-Reply-To: <199809211157.HAA01561@devsys.jaguNET.com> from "Jim Jagielski"
 at Sep 21, 98 07:57:57 am
Content-Type: text
Sender: new-httpd-owner@apache.org
Precedence: bulk
Reply-To: new-httpd@apache.org

Jim Jagielski wrote:
> 
> Rasmus Lerdorf wrote:
> > 
> > So, the question is, could I be doing something in the PHP module which
> > would cause this?  For example, what if the module caused a core dump in
> > the middle of a request?  Would the scoreboard get stuck in write mode for
> > that process?
> > 
> 
> But if the process dies for any reason, then the main httpd process
> should get the signal that it's died and take care of thing. The fact that
> there aren't any zombies means that the main httpd process has reaped
> the core-dumpers, at which case it should reset the correct scoreboard
> entries :/ All this implies a problem with Apache itself, not PHP...
> 
> What system are they running? (I'll also look at the original bug report
> as well).
> 

Rasmus,

Have them look at error.log for any entries with 'child' in them. It looks
like the above could be happening iff we lose track of a child's slot,
but if this happens this will be logged. No idea why or how we could have
lost the slot info though...

-- 
===========================================================================
   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"