tomcat-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "William A. Rowe, Jr." <>
Subject Re: [ANN/VOTE] JK 1.2.12 Released
Date Sun, 08 May 2005 05:39:16 GMT
At 10:37 PM 5/7/2005, Mladen Turk wrote:
>William A. Rowe, Jr. wrote:
>>It is NOT a release.
>Sure it is not. That's why there is a VOTE inside, and
>that's why it was posted only on the dev list.
>Perhaps I should subject that email as:
>'1.2.12 tarballs available for testing'.

Yes that would have been clearer - let's use that language in the

>We are speaking about that for more then a month.
>There was a stability vote for 1.2.11 and passed, but was
>not released even enough votes collected, because of bug
>in wc_close, manifesting in multi vhost scenarios.

Ack.  Although in theory it was 'released', I don't see any reason
the RM (release manager) has to honor the vote to release - if they 
think there is a fatal flaw.  If someone disagrees with the RM they
are welcome to serve as RM and roll a release, themself (as the RM.)

That said;

sure makes it appear that 1.2.11 was a released version.

>So instead just fixing that and re-releasing that should
>be done, there was a week left for any other findings.
>>Please, do not let this occur again.  If I missed a vote due to
>>connectivity issues; please, pardon the reminder of this policy.
>I really don't see what is your problem with releasing JK.

Not an issue with releasing (future tense), you totally missed 
the point.  The issue is PROCEDURE.  Post released versions.
Don't announce releases before a vote.

Your email *stated* 1.2.12 is released (present tense.)
states that 1.2.12 is released (development version).

So it appears to me, casual user, that 1.2.12 is released.

We need three +1's here - and I'm sure you will get them.  But
I'm objecting to placing carts before horses.  Your update 
to that /dist/ page makes it appear it is done, while it is not.

One thought; httpd uses the
URL to collect candidates; it makes is a little clearer which
tarballs are candidates, and which are actually released.  Perhaps
we adopt something similar in jakarta-tomcat?

>You can or can not file your vote for this release.

Ack - I look forward to testing - and an informed decision.
(I'm finding CHANGES is rather limited, so trawling through 
the commit logs has taken longer than I expected ;-)

I asked for a few weeks that would allow users to adopt 1.2.11,
test it, report back from the real-world, and have a really solid
release.  The last couple releases were scuttled quick due to new
bugs, I'd love to see that cycle end.  So would you, I know that.

>Further more you can even be a RM for the next one. I'm really sick
>an tired of doing that for the last year.

A thankless task at that.  If I can help back you up with Win32
binaries, I'm happy to offer as my schedule allows.  This last
three months I've had 9 day weeks of work, which don't fit well 
into 7 day calendars :)  I hope this changes in the next month.

To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message