Well that's interesting to note. I reboot our server every week aswell. What
I also noticed about this problem was that, if I stop Apache altogether, the
wa.exe processes remain visible in task manager. That is, they are not
terminated by Apache shutting down. I suppose that Apache has little to do
with wa.exe other than handling the requests to call it and dealing with the
html wa.exe feeds back to it (pls correct me if I'm wrong). If that is the
case, then I don't think Apache configuration is going to help me, which
means it's back to pointing the finger at wa.exe. LSoft suggested I post
here originally.
Is there a setting that controls the timeout on wa.exe processes somewhere
in Listserv?

--
Spencer Warhurst
Listserv Service Manager


> -----Original Message-----
> From: Paul Karagianis [mailto:[log in to unmask]]
> Sent: 15 March 2002 21:11
> To: [log in to unmask]
> Subject: Re: Apache & wa.exe problems
>
>
> On 12 Mar 2002 08:12:13 Warhurst, SI (Spencer) wrote:
>
> > We run Apache 1.3.20 on a Windows NT4 box (dual Xeon 550 w
> 512Mb RAM), and
> > have a problem where our web server stalls. That is, it
> will cease to
> > respond to any requests.
>
> FWIW we also see this under OpenVMS.  We had this happening
> intermittently
> back in the Fall of 2000 and it went away, then it came back
> this month with
> a vengence.  We ran from Nov 2000 to Dec 2001 without a
> re-boot and have been
> rebooting every 2 weeks or less since, for unrelated reasons.
>  Earlier this
> month the Web stalls started... the application (Purveyor)
> keeps creating
> new tasks which do nothing until it runs out of the 20
> available slots.
> From then on, it's dead until it's restarted.  I was also
> guessing this was
> a resource problem but have no idea what.  I finally noticed
> that the stalls
> all seem to occur right after a single one of our 640 lists is being
> accessed, always from the same domain: stonesoft.com.  The
> list is public
> and I don't see any stonesoft.com subscibers, but the early
> morning stalls
> during extended sessions suggest they are caused by something routine.
> It's an unremarkable list, maybe even smaller than most in
> terms of archive
> size and traffic.  I'm still looking into it.
>
>                                                 -Kary
>