You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@pivot.apache.org by Martijn Dashorst <ma...@gmail.com> on 2009/04/03 14:33:22 UTC
Voting for a release
I am glad everybody is on board for the release and voting happily.
There is one caveat: you are voting on behalf of the Apache Software
Foundation. Unless you actually looked at the release, confirmed
everything is ok, the tests run, the build is repeatable, you should
not vote +1. As a proposed Committer and PMC member you are expected
to perform due diligence.
A +1 vote carries a lot of weight. When I see +1's immediately
following a vote request on a release, I don't see due diligence, but
an anxious community to get a release out.
Martijn
Re: Voting for a release
Posted by Todd Volkert <tv...@gmail.com>.
I was just thinking about this, and I'm glad you brought this up
Martijn. I second what Martijn's saying. I really did perform the
smoke tests that I said in the email, and I kindly ask that each of
you do the same. Unpack the archives, inspect the contents, see if
they build corectly, run apps against the binaries, etc.
-T
On Fri, Apr 3, 2009 at 8:33 AM, Martijn Dashorst
<ma...@gmail.com> wrote:
> I am glad everybody is on board for the release and voting happily.
> There is one caveat: you are voting on behalf of the Apache Software
> Foundation. Unless you actually looked at the release, confirmed
> everything is ok, the tests run, the build is repeatable, you should
> not vote +1. As a proposed Committer and PMC member you are expected
> to perform due diligence.
>
> A +1 vote carries a lot of weight. When I see +1's immediately
> following a vote request on a release, I don't see due diligence, but
> an anxious community to get a release out.
>
> Martijn
>
Re: Voting for a release
Posted by Todd Volkert <tv...@gmail.com>.
We technically only need three +1 votes before putting this to the
general@ list, but I'd like to see +1 votes from both our mentors
before putting the vote to the incubator, because any reasons the
mentors have for not wanting to release are assumed to be showstoppers
for the larger incubator audience :)
See http://incubator.apache.org/incubation/Incubation_Policy.html#Releases
for more details.
-T
On Fri, Apr 3, 2009 at 11:41 AM, Sandro Martini
<sa...@gmail.com> wrote:
> Hi to all,
> sorry for my premature vote but until next week I'll not have time to
> do other on Pivot ...
>
> Anyway this is what I've done:
> - Unpack zip archives
> - Checked their md5 sums
>
> - reviewed unpacked source directory and BUILD, README, etc. contents
> - run ant tasks: "build", "doc", "dist", "install"
> - looked at the .txt RAT report
>
> Test environment: Java 6 Update 12 on Windows XP
> - loaded dom_demo.html in IE6 and Firefox 3.0.x and confirmed that the
> test application loaded and worked properly
> - loaded im_client.html and tested (but only first panel, without
> trying to login somewhere with it) in IE6 and Firefox 3.0.x
> - copied and adapted the Web Start jnlp sample of "Kitchen Sink" to
> the deploy dir, and tried to run, ok. Note: used a syntax for
> file:///C/ ... codebase , and if possible avoid directory and file
> names with spaces inside, Web Start doesn't like them.
> For example, instead of:
> <jnlp spec="1.5+"
> codebase="http://cwiki.apache.org/confluence/download/attachments/108483"
> href="kitchen_sink.jnlp"
>>
> i modified into (adapted to my test dir):
> <jnlp spec="1.5+"
> codebase="file:///C:/temp/apache-pivot-1.1-incubating-src/deploy"
> href="kitchen_sink.jnlp"
>>
>
>
> Next week I'll find time to do more tests ...
>
>
>
> Ah, one question on Voting for a Release:
> if for many reasons someone does not have time to do the tests and
> maybe for more than 1 day, what is happening ? All is blocked until
> all have voted, or there is a time limit ? Or other cases ? Maybe
> someone has some link/docs for this ...
>
> Thanks in advance for the clarification.
>
>
>
> Bye,
> Sandro
>
Re: Voting for a release
Posted by Sandro Martini <sa...@gmail.com>.
Hi to all,
sorry for my premature vote but until next week I'll not have time to
do other on Pivot ...
Anyway this is what I've done:
- Unpack zip archives
- Checked their md5 sums
- reviewed unpacked source directory and BUILD, README, etc. contents
- run ant tasks: "build", "doc", "dist", "install"
- looked at the .txt RAT report
Test environment: Java 6 Update 12 on Windows XP
- loaded dom_demo.html in IE6 and Firefox 3.0.x and confirmed that the
test application loaded and worked properly
- loaded im_client.html and tested (but only first panel, without
trying to login somewhere with it) in IE6 and Firefox 3.0.x
- copied and adapted the Web Start jnlp sample of "Kitchen Sink" to
the deploy dir, and tried to run, ok. Note: used a syntax for
file:///C/ ... codebase , and if possible avoid directory and file
names with spaces inside, Web Start doesn't like them.
For example, instead of:
<jnlp spec="1.5+"
codebase="http://cwiki.apache.org/confluence/download/attachments/108483"
href="kitchen_sink.jnlp"
>
i modified into (adapted to my test dir):
<jnlp spec="1.5+"
codebase="file:///C:/temp/apache-pivot-1.1-incubating-src/deploy"
href="kitchen_sink.jnlp"
>
Next week I'll find time to do more tests ...
Ah, one question on Voting for a Release:
if for many reasons someone does not have time to do the tests and
maybe for more than 1 day, what is happening ? All is blocked until
all have voted, or there is a time limit ? Or other cases ? Maybe
someone has some link/docs for this ...
Thanks in advance for the clarification.
Bye,
Sandro
Re: Voting for a release
Posted by Todd Volkert <tv...@gmail.com>.
Sorry - had to take my kid to school. My smoke test included the following:
* Unpack all archives (zip and tar.gz, source and binaries) into
distinct folders.
* Verified the md5 and sha1 checksums of all archives
* Verified the pgp signatures of all archives
* Inspected the Javadoc included in the binary release
* Built and ran a suite of proprietary Pivot apps against the binary
JAR files from the command line (DesktopApplicationContext)
* Built the "build", "doc", "dist", and "install" targets from within
the source distribution and verified that the output was correct (by
inspection).
* Looked at the .txt RAT report to make sure that the remaining files
without a license header were acceptable.
* Made sure that my most recent changes are in the new artifacts.
Note that I did not check any applets because I'm running 64 bit
Linux, for which there's no Sun Java plugin yet (and the IcedTea
plug-in isn't up pto snuff yet).
Note also that I built these artifacts from a local checkout of the
1.1 branch (tag).
-T
On Fri, Apr 3, 2009 at 9:18 AM, Martijn Dashorst
<ma...@gmail.com> wrote:
> A couple of the wicket committers lay out precisely what they tested
> (not each individual unit test, but what actions were done) and report
> that back in their vote. This has a couple of advantages: it shows
> what has been done, and provides guidance to others checking the
> release themselves.
>
> So can you elaborate more on what exactly you did to confirm that this
> release is valid? Browsers used, builds ran, etc. Did you use the
> artifacts as provided by the downloadable distribution? Did you check
> out the tag, and build the release from there, and compared it against
> the distribution? Did you use the self build artifacts to check if
> they too work?
>
> Martijn
>
> On Fri, Apr 3, 2009 at 2:52 PM, Greg Brown <gk...@mac.com> wrote:
>> Very good point. I have also performed a smoke test and reaffirm my +1 vote.
>>
>> On Friday, April 03, 2009, at 08:33AM, "Martijn Dashorst" <ma...@gmail.com> wrote:
>>>I am glad everybody is on board for the release and voting happily.
>>>There is one caveat: you are voting on behalf of the Apache Software
>>>Foundation. Unless you actually looked at the release, confirmed
>>>everything is ok, the tests run, the build is repeatable, you should
>>>not vote +1. As a proposed Committer and PMC member you are expected
>>>to perform due diligence.
>>>
>>>A +1 vote carries a lot of weight. When I see +1's immediately
>>>following a vote request on a release, I don't see due diligence, but
>>>an anxious community to get a release out.
>>>
>>>Martijn
>>>
>>>
>>
>
>
>
> --
> Become a Wicket expert, learn from the best: http://wicketinaction.com
> Apache Wicket 1.3.5 is released
> Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.
>
Re: Voting for a release
Posted by Martijn Dashorst <ma...@gmail.com>.
A couple of the wicket committers lay out precisely what they tested
(not each individual unit test, but what actions were done) and report
that back in their vote. This has a couple of advantages: it shows
what has been done, and provides guidance to others checking the
release themselves.
So can you elaborate more on what exactly you did to confirm that this
release is valid? Browsers used, builds ran, etc. Did you use the
artifacts as provided by the downloadable distribution? Did you check
out the tag, and build the release from there, and compared it against
the distribution? Did you use the self build artifacts to check if
they too work?
Martijn
On Fri, Apr 3, 2009 at 2:52 PM, Greg Brown <gk...@mac.com> wrote:
> Very good point. I have also performed a smoke test and reaffirm my +1 vote.
>
> On Friday, April 03, 2009, at 08:33AM, "Martijn Dashorst" <ma...@gmail.com> wrote:
>>I am glad everybody is on board for the release and voting happily.
>>There is one caveat: you are voting on behalf of the Apache Software
>>Foundation. Unless you actually looked at the release, confirmed
>>everything is ok, the tests run, the build is repeatable, you should
>>not vote +1. As a proposed Committer and PMC member you are expected
>>to perform due diligence.
>>
>>A +1 vote carries a lot of weight. When I see +1's immediately
>>following a vote request on a release, I don't see due diligence, but
>>an anxious community to get a release out.
>>
>>Martijn
>>
>>
>
--
Become a Wicket expert, learn from the best: http://wicketinaction.com
Apache Wicket 1.3.5 is released
Get it now: http://www.apache.org/dyn/closer.cgi/wicket/1.3.
Re: Voting for a release
Posted by Greg Brown <gk...@mac.com>.
Very good point. I have also performed a smoke test and reaffirm my +1 vote.
On Friday, April 03, 2009, at 08:33AM, "Martijn Dashorst" <ma...@gmail.com> wrote:
>I am glad everybody is on board for the release and voting happily.
>There is one caveat: you are voting on behalf of the Apache Software
>Foundation. Unless you actually looked at the release, confirmed
>everything is ok, the tests run, the build is repeatable, you should
>not vote +1. As a proposed Committer and PMC member you are expected
>to perform due diligence.
>
>A +1 vote carries a lot of weight. When I see +1's immediately
>following a vote request on a release, I don't see due diligence, but
>an anxious community to get a release out.
>
>Martijn
>
>