httpd-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Stein <>
Subject Re: Tagging Apache 2.0 for beta candidate
Date Sat, 19 May 2001 00:27:42 GMT
On Fri, May 18, 2001 at 11:26:50PM +0200, Graham Leggett wrote:
> Greg Stein wrote:
> > Even better, the proxy guys should say "okay. we have verified that our
> > stuff works with the 2.0.18 tarball, so let's release an apache+proxy
> > tarball."
> Really yuck for the end user.

WTF? How is that "yucky"?? The user sees two tarballs: apache and
apache+proxy. What is the big deal?

> There should be just one archive out there. There is no point in
> releasing an archive with A in it, then something with A+B - it's a
> complete waste of time.

Not at all. I don't want proxy. Many other people don't. Why is it a waste
of time for those people?

The waste of time is trying to get the httpd RM to include proxy into the
main tarball. The tarball is snapped at an arbitrary point in time, after
some arbitrary sets of changes. How can the RM know that proxy has been
updated to work with that particular snapshot? Answer: they can't.
Therefore: they shouldn't presume that it is okay for inclusion and release
against that specific httpd snapshot.

The proxy developers are the only people who know when mod_proxy is stable
for a particular snapshot of the httpd repository. They should be
responsible for making the snapshot of mod_proxy for a given httpd snapshot.

> > How was FirstBill to know whether proxy should have been included
> > or not? Was it stable and did it work against 2.0.18? No... he doesn't know
> > that.
> Then lets finalise the decision and they will know.

Finalize what decision? And who is "they"?

Bill made a snapshot and release of the httpd tree. That was done and in the
can with a very low overhead on his part. Heck, it is out the door already
for testing. Bing bam boom.

Low overhead releases are what we need.


Greg Stein,

View raw message