httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <wr...@lnd.com>
Subject RE: cvs commit: apache-1.3/src/os/win32 service.c
Date Thu, 15 Jun 2000 23:46:14 GMT
> From: Greg Stein [mailto:gstein@lyra.org]
> Sent: Thursday, June 15, 2000 5:48 PM
> 
> On Thu, Jun 15, 2000 at 05:18:53PM -0500, William A. Rowe, Jr. wrote:
> > > From: Keith Wannamaker [mailto:krw@raleigh.ibm.com]
> > > Sent: Thursday, June 15, 2000 5:14 PM
> > > 
> > > |yet.  I don't know that I will.  But if you are about to object 
> > > |on the face of it please do so now, before I waste more hours.
> > > 
> > > I am neither a voting member nor do I have commit access, I am
> > > just offering an opinion.
> > 
> > You are a developer who's opinion I hold in high regard.  I would
> > prefer to see broad comments on ideas early in the process, though!!!
> 
> I'm with Keith on this one.
> 
> If you want to test the change, then do it in 2.0. NOT 1.3.
> 
> If the change works in 2.0, then backport. But not until.
> 
> You mentioned two items that this is meant to fix. One item 
> said that you're
> experimenting for a 2.0 concept. The other was for a new 
> feature. As Keith
> said: don't experiment in 1.3, and don't introduce change to 
> enable a new,
> questionable feature.
> 
> -1
> 
> I'll gladly lift that if you could explain what it fixes. As 
> an enhancement,
> then I worry about the change. I'm +1 on testing it in 2.0; go for it.

You jumped tracks on me :-)

This post, or plea, was for Keith's reaction to the request by Ken
that we not close the console window if there is an error.  It's a
pending work that I won't waste my time with if folks are going to
nix it.  The fix for 2.0 will look quite different (since we can,
and already have started fixing the exit() balony.)

The commit in the subject is to fix the fact that we don't need to
be wasting Win32 resources creating a friggin console and all it's
extra garbage when we launch the service, if we stick to just testing
what MS already tells us that this is a forced, no prompt, no wait
window.  It has the beautiful side effect of allowing developers to
launch Apache any way they like, as long as they don't set that very
same flag.  But they can launch it hidden or whatever they choose.

I'm assuming you object to the latter.  I'm looking for testers, not
gripers, if you aren't testing, then you aren't part of the solution :-)
Win9x folks can ignore the patch, it doesn't affect that code path.  
I can test only NT 4, so Win2000 testers are -greatly- desired.  
Andrew is my most committed tester, and he will be waking up in 
Austrialia sometime soon with my request.  If he fails, it's out.  
If anyone else testing can cause it to fail, it's out.  But from all 
I read this should be quite reliable.

What suprizes me about the tone of comments in the past few weeks
isn't that radical changes can be bad... but that the Win32 port is
percieved to be 'lumbering along' well enough.  I strongly disagree
with that sentiment.

Bill

Mime
View raw message