You are viewing a plain text version of this content. The canonical link for it is here.
Posted to general@incubator.apache.org by Andrew Bayer <ab...@apache.org> on 2013/06/01 01:13:36 UTC
[VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Hello,
This is the second release candidate for Apache jclouds 1.6.1-incubating,
the first jclouds release at Apache.
The PPMC has approved the release at this point, so this is an IPMC vote.
It fixes the following issues:
https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12324412&styleName=Html&projectId=12314430
*** Please download, test and vote by Monday, June 3nd, 5pm PDT.
Note that we are voting upon the source (tag), binaries are provided for
convenience.
Source and binary files:
http://people.apache.org/~abayer/jclouds-1.6.1-incubating-candidate-2
Maven staging repo:
https://repository.apache.org/content/repositories/orgapachejclouds-043
The tags to be voted upon:
- jclouds -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tag;h=5123403cd8e146861be8b94d1b5ec2c4574ed810
- jclouds-labs -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;a=tag;h=5e78e6ed2343e84cc48492edc1147f4c68f2d0e3
- jclouds-chef -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-chef.git;a=tag;h=665a98fa04f313645aa484acd7959c721583f808
- jclouds-karaf -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-karaf.git;a=tag;h=fc357921782db55e01d27ad3cc1a392825978ef2
- jclouds-cli -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-cli.git;a=tag;h=c7f3509bbcb40a5cd35f917fbcf2c05222542de4
jclouds KEYS file containing PGP keys we use to sign the release:
http://www.apache.org/dist/incubator/jclouds/KEYS
[ ] +1
[ ] 0
[ ] -1 (explain why)
A.
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Good point - mea culpa on the lazy consensus. I got overly optimistic there, and had gotten mixed messages in the past on the necessity/significance of the PPMC vote.
Let's turn this thread (on the dev list) into the PPMC vote thread - same deadline of 5pm PDT on Monday, if that's ok with everyone. And definitely do follow David's advice on vetting the release candidate.
A.
On Jun 1, 2013, at 6:41 AM, David Nalley <da...@gnsa.us> wrote:
> On Fri, May 31, 2013 at 7:17 PM, Andrew Bayer <an...@gmail.com> wrote:
>> FYI, you don't need to vote this time - I'm taking it as written that we
>> all approved, since there were no -1s earlier.
>>
>> A.
>>
>
>
> So from a legal standpoint you need explicit PMC votes on explicit
> tarballs. No such thing as lazy consensus for a release. Admittedly
> this is a release of the incubator, and thus you need three IPMC
> votes, but you are supposedly doing this as if this were a TLP
> release, and if the PPMC isn't satisfied, the IPMC will almost
> certainly not be.
>
> Also - for PPMC folks voting - the IPMC generally prefers to know how
> you personally vetted the release artifacts. (admittedly it's a bit
> like showing your work on a math problem) So please include those
> details in your vote. Presumably the software is in good shape or we
> wouldn't be performing a release. Did you verify the checksums, the
> GPG signatures? Are you sure that's really Andrew's key, and if so,
> how? Did you run RAT against the release artifacts? Have you done
> additional IP audit? Did you compare the commit hash of the tag with
> the release tarball? Once you become a PMC, you'll be responsible for
> complying with ASF policy and legal requirements around a release - we
> already assume you are capable of releasing good software from a
> technical perspective - so the concern is around ensuring that you are
> performing due diligence to comply with the obligations that go along
> with releases.
>
> --David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by David Nalley <da...@gnsa.us>.
On Fri, May 31, 2013 at 7:17 PM, Andrew Bayer <an...@gmail.com> wrote:
> FYI, you don't need to vote this time - I'm taking it as written that we
> all approved, since there were no -1s earlier.
>
> A.
>
So from a legal standpoint you need explicit PMC votes on explicit
tarballs. No such thing as lazy consensus for a release. Admittedly
this is a release of the incubator, and thus you need three IPMC
votes, but you are supposedly doing this as if this were a TLP
release, and if the PPMC isn't satisfied, the IPMC will almost
certainly not be.
Also - for PPMC folks voting - the IPMC generally prefers to know how
you personally vetted the release artifacts. (admittedly it's a bit
like showing your work on a math problem) So please include those
details in your vote. Presumably the software is in good shape or we
wouldn't be performing a release. Did you verify the checksums, the
GPG signatures? Are you sure that's really Andrew's key, and if so,
how? Did you run RAT against the release artifacts? Have you done
additional IP audit? Did you compare the commit hash of the tag with
the release tarball? Once you become a PMC, you'll be responsible for
complying with ASF policy and legal requirements around a release - we
already assume you are capable of releasing good software from a
technical perspective - so the concern is around ensuring that you are
performing due diligence to comply with the obligations that go along
with releases.
--David
Fwd: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
FYI, you don't need to vote this time - I'm taking it as written that we
all approved, since there were no -1s earlier.
A.
---------- Forwarded message ----------
From: Andrew Bayer <ab...@apache.org>
Date: Fri, May 31, 2013 at 4:13 PM
Subject: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
To: dev@jclouds.incubator.apache.org, general@incubator.apache.org
Hello,
This is the second release candidate for Apache jclouds 1.6.1-incubating,
the first jclouds release at Apache.
The PPMC has approved the release at this point, so this is an IPMC vote.
It fixes the following issues:
https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12324412&styleName=Html&projectId=12314430
*** Please download, test and vote by Monday, June 3nd, 5pm PDT.
Note that we are voting upon the source (tag), binaries are provided for
convenience.
Source and binary files:
http://people.apache.org/~abayer/jclouds-1.6.1-incubating-candidate-2
Maven staging repo:
https://repository.apache.org/content/repositories/orgapachejclouds-043
The tags to be voted upon:
- jclouds -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tag;h=5123403cd8e146861be8b94d1b5ec2c4574ed810
- jclouds-labs -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;a=tag;h=5e78e6ed2343e84cc48492edc1147f4c68f2d0e3
- jclouds-chef -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-chef.git;a=tag;h=665a98fa04f313645aa484acd7959c721583f808
- jclouds-karaf -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-karaf.git;a=tag;h=fc357921782db55e01d27ad3cc1a392825978ef2
- jclouds-cli -
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-cli.git;a=tag;h=c7f3509bbcb40a5cd35f917fbcf2c05222542de4
jclouds KEYS file containing PGP keys we use to sign the release:
http://www.apache.org/dist/incubator/jclouds/KEYS
[ ] +1
[ ] 0
[ ] -1 (explain why)
A.
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Actually, we are going to pause the IPMC vote while we clarify the PPMC vote. I'll send a new thread when we're ready to restart the IPMC vote, but I would definitely appreciate any comments and concerns on the RC2 artifacts in the mean time.
A.
On Jun 1, 2013, at 8:47 AM, Alan Cabrera <li...@toolazydogs.com> wrote:
> How many binding votes are there from the PPMC vote?
>
>
> Regards,
> Alan
>
> On May 31, 2013, at 4:13 PM, Andrew Bayer <ab...@apache.org> wrote:
>
>> Hello,
>>
>> This is the second release candidate for Apache jclouds 1.6.1-incubating,
>> the first jclouds release at Apache.
>>
>> The PPMC has approved the release at this point, so this is an IPMC vote.
>>
>> It fixes the following issues:
>> https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12324412&styleName=Html&projectId=12314430
>>
>> *** Please download, test and vote by Monday, June 3nd, 5pm PDT.
>>
>> Note that we are voting upon the source (tag), binaries are provided for
>> convenience.
>>
>> Source and binary files:
>> http://people.apache.org/~abayer/jclouds-1.6.1-incubating-candidate-2
>>
>> Maven staging repo:
>> https://repository.apache.org/content/repositories/orgapachejclouds-043
>>
>> The tags to be voted upon:
>> - jclouds -
>> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tag;h=5123403cd8e146861be8b94d1b5ec2c4574ed810
>> - jclouds-labs -
>> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;a=tag;h=5e78e6ed2343e84cc48492edc1147f4c68f2d0e3
>> - jclouds-chef -
>> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-chef.git;a=tag;h=665a98fa04f313645aa484acd7959c721583f808
>> - jclouds-karaf -
>> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-karaf.git;a=tag;h=fc357921782db55e01d27ad3cc1a392825978ef2
>> - jclouds-cli -
>> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-cli.git;a=tag;h=c7f3509bbcb40a5cd35f917fbcf2c05222542de4
>>
>> jclouds KEYS file containing PGP keys we use to sign the release:
>> http://www.apache.org/dist/incubator/jclouds/KEYS
>>
>> [ ] +1
>> [ ] 0
>> [ ] -1 (explain why)
>>
>> A.
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
> For additional commands, e-mail: general-help@incubator.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Alan Cabrera <li...@toolazydogs.com>.
How many binding votes are there from the PPMC vote?
Regards,
Alan
On May 31, 2013, at 4:13 PM, Andrew Bayer <ab...@apache.org> wrote:
> Hello,
>
> This is the second release candidate for Apache jclouds 1.6.1-incubating,
> the first jclouds release at Apache.
>
> The PPMC has approved the release at this point, so this is an IPMC vote.
>
> It fixes the following issues:
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12324412&styleName=Html&projectId=12314430
>
> *** Please download, test and vote by Monday, June 3nd, 5pm PDT.
>
> Note that we are voting upon the source (tag), binaries are provided for
> convenience.
>
> Source and binary files:
> http://people.apache.org/~abayer/jclouds-1.6.1-incubating-candidate-2
>
> Maven staging repo:
> https://repository.apache.org/content/repositories/orgapachejclouds-043
>
> The tags to be voted upon:
> - jclouds -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tag;h=5123403cd8e146861be8b94d1b5ec2c4574ed810
> - jclouds-labs -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;a=tag;h=5e78e6ed2343e84cc48492edc1147f4c68f2d0e3
> - jclouds-chef -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-chef.git;a=tag;h=665a98fa04f313645aa484acd7959c721583f808
> - jclouds-karaf -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-karaf.git;a=tag;h=fc357921782db55e01d27ad3cc1a392825978ef2
> - jclouds-cli -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-cli.git;a=tag;h=c7f3509bbcb40a5cd35f917fbcf2c05222542de4
>
> jclouds KEYS file containing PGP keys we use to sign the release:
> http://www.apache.org/dist/incubator/jclouds/KEYS
>
> [ ] +1
> [ ] 0
> [ ] -1 (explain why)
>
> A.
---------------------------------------------------------------------
To unsubscribe, e-mail: general-unsubscribe@incubator.apache.org
For additional commands, e-mail: general-help@incubator.apache.org
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Alan Cabrera <li...@toolazydogs.com>.
How many binding votes are there from the PPMC vote?
Regards,
Alan
On May 31, 2013, at 4:13 PM, Andrew Bayer <ab...@apache.org> wrote:
> Hello,
>
> This is the second release candidate for Apache jclouds 1.6.1-incubating,
> the first jclouds release at Apache.
>
> The PPMC has approved the release at this point, so this is an IPMC vote.
>
> It fixes the following issues:
> https://issues.apache.org/jira/secure/ReleaseNote.jspa?version=12324412&styleName=Html&projectId=12314430
>
> *** Please download, test and vote by Monday, June 3nd, 5pm PDT.
>
> Note that we are voting upon the source (tag), binaries are provided for
> convenience.
>
> Source and binary files:
> http://people.apache.org/~abayer/jclouds-1.6.1-incubating-candidate-2
>
> Maven staging repo:
> https://repository.apache.org/content/repositories/orgapachejclouds-043
>
> The tags to be voted upon:
> - jclouds -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tag;h=5123403cd8e146861be8b94d1b5ec2c4574ed810
> - jclouds-labs -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-labs.git;a=tag;h=5e78e6ed2343e84cc48492edc1147f4c68f2d0e3
> - jclouds-chef -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-chef.git;a=tag;h=665a98fa04f313645aa484acd7959c721583f808
> - jclouds-karaf -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-karaf.git;a=tag;h=fc357921782db55e01d27ad3cc1a392825978ef2
> - jclouds-cli -
> https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds-cli.git;a=tag;h=c7f3509bbcb40a5cd35f917fbcf2c05222542de4
>
> jclouds KEYS file containing PGP keys we use to sign the release:
> http://www.apache.org/dist/incubator/jclouds/KEYS
>
> [ ] +1
> [ ] 0
> [ ] -1 (explain why)
>
> A.
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
I agree with Ignasi - we do need to get rid of the blanket exclusion for
src/test/resources, but the XML files there are, in my opinion, pretty
clearly not creative.
A.
On Mon, Jun 10, 2013 at 11:26 AM, Ignasi <ig...@gmail.com> wrote:
> Thanks for checking Tom!
>
> Please, apologize me if this has already been answered, but I am
> really confused and I still don't get what should be done with the
> headers in test files (and it is important to have it clear because it
> affects what and how we can perform tests).
>
> I really think that XML files in src/test/resources should not have
> test headers. As said in other threads, those XML files are used just
> to verify the output of the cloud APIs. In clouds we have two kinds of
> tests:
>
> * Expect tests (offline): These tests just mock the cloud apis, and we
> use them to properly check that we are generating the requests as
> expected, and we are properly parsing the responses. The XML/JSON
> files in src/test/resources are just a copy of the real responses the
> cloud apis return, and we use them to build the mock responses as
> close to the reality as possible.
> * Live tests (online): These tests actually execute the requests
> against the cloud apis, and check the behavior against real cloud
> providers.
>
> Having this clear, I think those XML/JSON files fail into the
> "non-creative" category (they are just a copies of real http response
> bodies). Furthermore, since the real cloud apis don't return an HTTP
> response body with license headers, our "response templates" shouldn't
> have them either. That's why we think it is important to keep those
> files without the license headers; adding them will make it impossible
> to properly add live tests that check the response bodies.
>
>
> Does this make sense? If it does, can we keep JSON/XML files in the
> src/test/resources directory without the license headers?
>
>
> Thanks for your patience, mentors!
>
>
> Ignasi
>
>
> On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
> > And there are a good number of tests failing because XML files aren't
> > expected to have headers. I'll see what I can do.
> >
> > A.
> >
> > On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >
> >> Also, if those don't full under the category of "non-creative", I don't
> >> know what does. =)
> >>
> >> A.
> >>
> >>
> >> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >>
> >>> Ok, the META-INF/services stuff will need at least test changes to work
> >>> right with license headers added, if not full blown code changes, so
> that
> >>> won't be happening.
> >>>
> >>> A.
> >>>
> >>>
> >>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >>>
> >>>> Will do.
> >>>>
> >>>> A.
> >>>>
> >>>>
> >>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
> >>>>
> >>>>> I ran through basic checks on the source packages (checksums,
> >>>>> signatures, license headers, included binary files, LICENSE, NOTICE,
> >>>>> DISCLAIMER), and generally they look good. In addition to the points
> >>>>> made on
> >>>>>
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
> >>>>> the only thing I would add is that RAT flags up lots of missing
> >>>>> license headers. Many of them are JSON which can be excluded since
> >>>>> JSON doesn't support comments (as discussed above), or tests, but the
> >>>>> META-INF/services, YAML, XML, and properties files should have
> license
> >>>>> headers added for RC3 if possible.
> >>>>>
> >>>>> Cheers,
> >>>>> Tom
> >>>>>
> >>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
> >>>>> wrote:
> >>>>> > David, have you had a chance to take a look at my comments about
> the
> >>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
> >>>>> >
> >>>>> > I've also noted my doubts on the wiki so the conclusions for each
> one
> >>>>> > can be reflected there.
> >>>>> >
> >>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
> >>>>> >> Dealt with for RC3, and noted on
> >>>>> >>
> >>>>>
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
> >>>>> >>
> >>>>> >> A.
> >>>>> >>
> >>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us>
> wrote:
> >>>>> >>
> >>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
> >>>>> andrew.bayer@gmail.com>
> >>>>> >>> wrote:
> >>>>> >>> > I'm extending the vote another 24 hours for our mentors to
> >>>>> respond.
> >>>>> >>> >
> >>>>> >>> > A.
> >>>>> >>> >
> >>>>> >>> >
> >>>>> >>>
> >>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
> >>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
> >>>>> shape.
> >>>>> >>>
> >>>>> >>> --David
> >>>>> >>>
> >>>>>
> >>>>
> >>>>
> >>>
> >>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
No worries. I'm just gonna start cracking the whip for RC3 tomorrow!
A.
On Wed, Jun 12, 2013 at 3:58 PM, David Nalley <da...@gnsa.us> wrote:
> On Wed, Jun 12, 2013 at 6:55 PM, Andrew Bayer <an...@gmail.com>
> wrote:
> > Actually, I'll do this tomorrow - too many balls in the air here at work
> > right now. =)
> >
> > A.
>
>
> I'll try and spend some time later tonight reviewing things to catch
> anything else that pops up.
>
> Thanks for your patience with us - sorry we gotten through it all
> heretofore.
>
> --David
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by David Nalley <da...@gnsa.us>.
On Wed, Jun 12, 2013 at 6:55 PM, Andrew Bayer <an...@gmail.com> wrote:
> Actually, I'll do this tomorrow - too many balls in the air here at work
> right now. =)
>
> A.
I'll try and spend some time later tonight reviewing things to catch
anything else that pops up.
Thanks for your patience with us - sorry we gotten through it all heretofore.
--David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Actually, I'll do this tomorrow - too many balls in the air here at work
right now. =)
A.
On Wed, Jun 12, 2013 at 12:29 PM, Andrew Bayer <an...@gmail.com>wrote:
> Ok, given that we're now nearly two weeks since I cut RC2, I'm going to
> cut RC3 later today, incorporating the changes we've discussed at
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues.
> I'm then going to call a 72 hour vote that will require 3 +1s from mentors,
> and I'm going to email each mentor personally to make sure they respond. =)
>
> A.
>
>
> On Tue, Jun 11, 2013 at 8:27 AM, Andrew Bayer <an...@gmail.com>wrote:
>
>> Test expectations, at least. A ton of tests started failing when I added
>> headers to those files. I don't know if there are functional problems too.
>> I'd ideally like to get that all straightened out in the next release, if
>> we can survive as is for 1.6.1.
>>
>> A.
>> On Jun 11, 2013 7:51 AM, "Tom White" <to...@cloudera.com> wrote:
>>
>>> That makes a lot of sense - thanks for the explanation, Ignasi.
>>>
>>> There are other files in the test directory that should have headers
>>> added though, like the log4j.xml files. It would be easier to filter
>>> if the test expectation files were in an identifiable directory, e.g.
>>> src/test/resources/expected. That way when other files are added to
>>> the test hierarchy (that are not expectation files), then they won't
>>> be filtered inadvertently. This is probably not absolutely required
>>> for the first release though.
>>>
>>> Regarding the META-INF/services files - there are lots of these files
>>> in main that you should be able to add headers to using # as the
>>> comment character, see
>>>
>>> https://github.com/apache/whirr/blob/trunk/core/src/main/resources/META-INF/services/org.apache.whirr.ClusterController
>>> .
>>> Is there any reason why headers can't be added to the test ones too?
>>>
>>> Cheers,
>>> Tom
>>>
>>> On Mon, Jun 10, 2013 at 7:26 PM, Ignasi <ig...@gmail.com>
>>> wrote:
>>> > Thanks for checking Tom!
>>> >
>>> > Please, apologize me if this has already been answered, but I am
>>> > really confused and I still don't get what should be done with the
>>> > headers in test files (and it is important to have it clear because it
>>> > affects what and how we can perform tests).
>>> >
>>> > I really think that XML files in src/test/resources should not have
>>> > test headers. As said in other threads, those XML files are used just
>>> > to verify the output of the cloud APIs. In clouds we have two kinds of
>>> > tests:
>>> >
>>> > * Expect tests (offline): These tests just mock the cloud apis, and we
>>> > use them to properly check that we are generating the requests as
>>> > expected, and we are properly parsing the responses. The XML/JSON
>>> > files in src/test/resources are just a copy of the real responses the
>>> > cloud apis return, and we use them to build the mock responses as
>>> > close to the reality as possible.
>>> > * Live tests (online): These tests actually execute the requests
>>> > against the cloud apis, and check the behavior against real cloud
>>> > providers.
>>> >
>>> > Having this clear, I think those XML/JSON files fail into the
>>> > "non-creative" category (they are just a copies of real http response
>>> > bodies). Furthermore, since the real cloud apis don't return an HTTP
>>> > response body with license headers, our "response templates" shouldn't
>>> > have them either. That's why we think it is important to keep those
>>> > files without the license headers; adding them will make it impossible
>>> > to properly add live tests that check the response bodies.
>>> >
>>> >
>>> > Does this make sense? If it does, can we keep JSON/XML files in the
>>> > src/test/resources directory without the license headers?
>>> >
>>> >
>>> > Thanks for your patience, mentors!
>>> >
>>> >
>>> > Ignasi
>>> >
>>> >
>>> > On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
>>> >> And there are a good number of tests failing because XML files aren't
>>> >> expected to have headers. I'll see what I can do.
>>> >>
>>> >> A.
>>> >>
>>> >> On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <
>>> andrew.bayer@gmail.com>wrote:
>>> >>
>>> >>> Also, if those don't full under the category of "non-creative", I
>>> don't
>>> >>> know what does. =)
>>> >>>
>>> >>> A.
>>> >>>
>>> >>>
>>> >>> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <
>>> andrew.bayer@gmail.com>wrote:
>>> >>>
>>> >>>> Ok, the META-INF/services stuff will need at least test changes to
>>> work
>>> >>>> right with license headers added, if not full blown code changes,
>>> so that
>>> >>>> won't be happening.
>>> >>>>
>>> >>>> A.
>>> >>>>
>>> >>>>
>>> >>>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <
>>> andrew.bayer@gmail.com>wrote:
>>> >>>>
>>> >>>>> Will do.
>>> >>>>>
>>> >>>>> A.
>>> >>>>>
>>> >>>>>
>>> >>>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com>
>>> wrote:
>>> >>>>>
>>> >>>>>> I ran through basic checks on the source packages (checksums,
>>> >>>>>> signatures, license headers, included binary files, LICENSE,
>>> NOTICE,
>>> >>>>>> DISCLAIMER), and generally they look good. In addition to the
>>> points
>>> >>>>>> made on
>>> >>>>>>
>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>>> >>>>>> the only thing I would add is that RAT flags up lots of missing
>>> >>>>>> license headers. Many of them are JSON which can be excluded since
>>> >>>>>> JSON doesn't support comments (as discussed above), or tests, but
>>> the
>>> >>>>>> META-INF/services, YAML, XML, and properties files should have
>>> license
>>> >>>>>> headers added for RC3 if possible.
>>> >>>>>>
>>> >>>>>> Cheers,
>>> >>>>>> Tom
>>> >>>>>>
>>> >>>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
>>> >>>>>> wrote:
>>> >>>>>> > David, have you had a chance to take a look at my comments
>>> about the
>>> >>>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>>> >>>>>> >
>>> >>>>>> > I've also noted my doubts on the wiki so the conclusions for
>>> each one
>>> >>>>>> > can be reflected there.
>>> >>>>>> >
>>> >>>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com>
>>> wrote:
>>> >>>>>> >> Dealt with for RC3, and noted on
>>> >>>>>> >>
>>> >>>>>>
>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>> >>>>>> >>
>>> >>>>>> >> A.
>>> >>>>>> >>
>>> >>>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us>
>>> wrote:
>>> >>>>>> >>
>>> >>>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
>>> >>>>>> andrew.bayer@gmail.com>
>>> >>>>>> >>> wrote:
>>> >>>>>> >>> > I'm extending the vote another 24 hours for our mentors to
>>> >>>>>> respond.
>>> >>>>>> >>> >
>>> >>>>>> >>> > A.
>>> >>>>>> >>> >
>>> >>>>>> >>> >
>>> >>>>>> >>>
>>> >>>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue
>>> that
>>> >>>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
>>> >>>>>> shape.
>>> >>>>>> >>>
>>> >>>>>> >>> --David
>>> >>>>>> >>>
>>> >>>>>>
>>> >>>>>
>>> >>>>>
>>> >>>>
>>> >>>
>>>
>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Ok, given that we're now nearly two weeks since I cut RC2, I'm going to cut
RC3 later today, incorporating the changes we've discussed at
https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues. I'm
then going to call a 72 hour vote that will require 3 +1s from mentors, and
I'm going to email each mentor personally to make sure they respond. =)
A.
On Tue, Jun 11, 2013 at 8:27 AM, Andrew Bayer <an...@gmail.com>wrote:
> Test expectations, at least. A ton of tests started failing when I added
> headers to those files. I don't know if there are functional problems too.
> I'd ideally like to get that all straightened out in the next release, if
> we can survive as is for 1.6.1.
>
> A.
> On Jun 11, 2013 7:51 AM, "Tom White" <to...@cloudera.com> wrote:
>
>> That makes a lot of sense - thanks for the explanation, Ignasi.
>>
>> There are other files in the test directory that should have headers
>> added though, like the log4j.xml files. It would be easier to filter
>> if the test expectation files were in an identifiable directory, e.g.
>> src/test/resources/expected. That way when other files are added to
>> the test hierarchy (that are not expectation files), then they won't
>> be filtered inadvertently. This is probably not absolutely required
>> for the first release though.
>>
>> Regarding the META-INF/services files - there are lots of these files
>> in main that you should be able to add headers to using # as the
>> comment character, see
>>
>> https://github.com/apache/whirr/blob/trunk/core/src/main/resources/META-INF/services/org.apache.whirr.ClusterController
>> .
>> Is there any reason why headers can't be added to the test ones too?
>>
>> Cheers,
>> Tom
>>
>> On Mon, Jun 10, 2013 at 7:26 PM, Ignasi <ig...@gmail.com> wrote:
>> > Thanks for checking Tom!
>> >
>> > Please, apologize me if this has already been answered, but I am
>> > really confused and I still don't get what should be done with the
>> > headers in test files (and it is important to have it clear because it
>> > affects what and how we can perform tests).
>> >
>> > I really think that XML files in src/test/resources should not have
>> > test headers. As said in other threads, those XML files are used just
>> > to verify the output of the cloud APIs. In clouds we have two kinds of
>> > tests:
>> >
>> > * Expect tests (offline): These tests just mock the cloud apis, and we
>> > use them to properly check that we are generating the requests as
>> > expected, and we are properly parsing the responses. The XML/JSON
>> > files in src/test/resources are just a copy of the real responses the
>> > cloud apis return, and we use them to build the mock responses as
>> > close to the reality as possible.
>> > * Live tests (online): These tests actually execute the requests
>> > against the cloud apis, and check the behavior against real cloud
>> > providers.
>> >
>> > Having this clear, I think those XML/JSON files fail into the
>> > "non-creative" category (they are just a copies of real http response
>> > bodies). Furthermore, since the real cloud apis don't return an HTTP
>> > response body with license headers, our "response templates" shouldn't
>> > have them either. That's why we think it is important to keep those
>> > files without the license headers; adding them will make it impossible
>> > to properly add live tests that check the response bodies.
>> >
>> >
>> > Does this make sense? If it does, can we keep JSON/XML files in the
>> > src/test/resources directory without the license headers?
>> >
>> >
>> > Thanks for your patience, mentors!
>> >
>> >
>> > Ignasi
>> >
>> >
>> > On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
>> >> And there are a good number of tests failing because XML files aren't
>> >> expected to have headers. I'll see what I can do.
>> >>
>> >> A.
>> >>
>> >> On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <andrew.bayer@gmail.com
>> >wrote:
>> >>
>> >>> Also, if those don't full under the category of "non-creative", I
>> don't
>> >>> know what does. =)
>> >>>
>> >>> A.
>> >>>
>> >>>
>> >>> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <
>> andrew.bayer@gmail.com>wrote:
>> >>>
>> >>>> Ok, the META-INF/services stuff will need at least test changes to
>> work
>> >>>> right with license headers added, if not full blown code changes, so
>> that
>> >>>> won't be happening.
>> >>>>
>> >>>> A.
>> >>>>
>> >>>>
>> >>>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <
>> andrew.bayer@gmail.com>wrote:
>> >>>>
>> >>>>> Will do.
>> >>>>>
>> >>>>> A.
>> >>>>>
>> >>>>>
>> >>>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com>
>> wrote:
>> >>>>>
>> >>>>>> I ran through basic checks on the source packages (checksums,
>> >>>>>> signatures, license headers, included binary files, LICENSE,
>> NOTICE,
>> >>>>>> DISCLAIMER), and generally they look good. In addition to the
>> points
>> >>>>>> made on
>> >>>>>>
>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>> >>>>>> the only thing I would add is that RAT flags up lots of missing
>> >>>>>> license headers. Many of them are JSON which can be excluded since
>> >>>>>> JSON doesn't support comments (as discussed above), or tests, but
>> the
>> >>>>>> META-INF/services, YAML, XML, and properties files should have
>> license
>> >>>>>> headers added for RC3 if possible.
>> >>>>>>
>> >>>>>> Cheers,
>> >>>>>> Tom
>> >>>>>>
>> >>>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
>> >>>>>> wrote:
>> >>>>>> > David, have you had a chance to take a look at my comments about
>> the
>> >>>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>> >>>>>> >
>> >>>>>> > I've also noted my doubts on the wiki so the conclusions for
>> each one
>> >>>>>> > can be reflected there.
>> >>>>>> >
>> >>>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com>
>> wrote:
>> >>>>>> >> Dealt with for RC3, and noted on
>> >>>>>> >>
>> >>>>>>
>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>> >>>>>> >>
>> >>>>>> >> A.
>> >>>>>> >>
>> >>>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us>
>> wrote:
>> >>>>>> >>
>> >>>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
>> >>>>>> andrew.bayer@gmail.com>
>> >>>>>> >>> wrote:
>> >>>>>> >>> > I'm extending the vote another 24 hours for our mentors to
>> >>>>>> respond.
>> >>>>>> >>> >
>> >>>>>> >>> > A.
>> >>>>>> >>> >
>> >>>>>> >>> >
>> >>>>>> >>>
>> >>>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue
>> that
>> >>>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
>> >>>>>> shape.
>> >>>>>> >>>
>> >>>>>> >>> --David
>> >>>>>> >>>
>> >>>>>>
>> >>>>>
>> >>>>>
>> >>>>
>> >>>
>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Test expectations, at least. A ton of tests started failing when I added
headers to those files. I don't know if there are functional problems too.
I'd ideally like to get that all straightened out in the next release, if
we can survive as is for 1.6.1.
A.
On Jun 11, 2013 7:51 AM, "Tom White" <to...@cloudera.com> wrote:
> That makes a lot of sense - thanks for the explanation, Ignasi.
>
> There are other files in the test directory that should have headers
> added though, like the log4j.xml files. It would be easier to filter
> if the test expectation files were in an identifiable directory, e.g.
> src/test/resources/expected. That way when other files are added to
> the test hierarchy (that are not expectation files), then they won't
> be filtered inadvertently. This is probably not absolutely required
> for the first release though.
>
> Regarding the META-INF/services files - there are lots of these files
> in main that you should be able to add headers to using # as the
> comment character, see
>
> https://github.com/apache/whirr/blob/trunk/core/src/main/resources/META-INF/services/org.apache.whirr.ClusterController
> .
> Is there any reason why headers can't be added to the test ones too?
>
> Cheers,
> Tom
>
> On Mon, Jun 10, 2013 at 7:26 PM, Ignasi <ig...@gmail.com> wrote:
> > Thanks for checking Tom!
> >
> > Please, apologize me if this has already been answered, but I am
> > really confused and I still don't get what should be done with the
> > headers in test files (and it is important to have it clear because it
> > affects what and how we can perform tests).
> >
> > I really think that XML files in src/test/resources should not have
> > test headers. As said in other threads, those XML files are used just
> > to verify the output of the cloud APIs. In clouds we have two kinds of
> > tests:
> >
> > * Expect tests (offline): These tests just mock the cloud apis, and we
> > use them to properly check that we are generating the requests as
> > expected, and we are properly parsing the responses. The XML/JSON
> > files in src/test/resources are just a copy of the real responses the
> > cloud apis return, and we use them to build the mock responses as
> > close to the reality as possible.
> > * Live tests (online): These tests actually execute the requests
> > against the cloud apis, and check the behavior against real cloud
> > providers.
> >
> > Having this clear, I think those XML/JSON files fail into the
> > "non-creative" category (they are just a copies of real http response
> > bodies). Furthermore, since the real cloud apis don't return an HTTP
> > response body with license headers, our "response templates" shouldn't
> > have them either. That's why we think it is important to keep those
> > files without the license headers; adding them will make it impossible
> > to properly add live tests that check the response bodies.
> >
> >
> > Does this make sense? If it does, can we keep JSON/XML files in the
> > src/test/resources directory without the license headers?
> >
> >
> > Thanks for your patience, mentors!
> >
> >
> > Ignasi
> >
> >
> > On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
> >> And there are a good number of tests failing because XML files aren't
> >> expected to have headers. I'll see what I can do.
> >>
> >> A.
> >>
> >> On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >>
> >>> Also, if those don't full under the category of "non-creative", I don't
> >>> know what does. =)
> >>>
> >>> A.
> >>>
> >>>
> >>> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >>>
> >>>> Ok, the META-INF/services stuff will need at least test changes to
> work
> >>>> right with license headers added, if not full blown code changes, so
> that
> >>>> won't be happening.
> >>>>
> >>>> A.
> >>>>
> >>>>
> >>>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >>>>
> >>>>> Will do.
> >>>>>
> >>>>> A.
> >>>>>
> >>>>>
> >>>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
> >>>>>
> >>>>>> I ran through basic checks on the source packages (checksums,
> >>>>>> signatures, license headers, included binary files, LICENSE, NOTICE,
> >>>>>> DISCLAIMER), and generally they look good. In addition to the points
> >>>>>> made on
> >>>>>>
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
> >>>>>> the only thing I would add is that RAT flags up lots of missing
> >>>>>> license headers. Many of them are JSON which can be excluded since
> >>>>>> JSON doesn't support comments (as discussed above), or tests, but
> the
> >>>>>> META-INF/services, YAML, XML, and properties files should have
> license
> >>>>>> headers added for RC3 if possible.
> >>>>>>
> >>>>>> Cheers,
> >>>>>> Tom
> >>>>>>
> >>>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
> >>>>>> wrote:
> >>>>>> > David, have you had a chance to take a look at my comments about
> the
> >>>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
> >>>>>> >
> >>>>>> > I've also noted my doubts on the wiki so the conclusions for each
> one
> >>>>>> > can be reflected there.
> >>>>>> >
> >>>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com>
> wrote:
> >>>>>> >> Dealt with for RC3, and noted on
> >>>>>> >>
> >>>>>>
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
> >>>>>> >>
> >>>>>> >> A.
> >>>>>> >>
> >>>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us>
> wrote:
> >>>>>> >>
> >>>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
> >>>>>> andrew.bayer@gmail.com>
> >>>>>> >>> wrote:
> >>>>>> >>> > I'm extending the vote another 24 hours for our mentors to
> >>>>>> respond.
> >>>>>> >>> >
> >>>>>> >>> > A.
> >>>>>> >>> >
> >>>>>> >>> >
> >>>>>> >>>
> >>>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue
> that
> >>>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
> >>>>>> shape.
> >>>>>> >>>
> >>>>>> >>> --David
> >>>>>> >>>
> >>>>>>
> >>>>>
> >>>>>
> >>>>
> >>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Tom White <to...@cloudera.com>.
That makes a lot of sense - thanks for the explanation, Ignasi.
There are other files in the test directory that should have headers
added though, like the log4j.xml files. It would be easier to filter
if the test expectation files were in an identifiable directory, e.g.
src/test/resources/expected. That way when other files are added to
the test hierarchy (that are not expectation files), then they won't
be filtered inadvertently. This is probably not absolutely required
for the first release though.
Regarding the META-INF/services files - there are lots of these files
in main that you should be able to add headers to using # as the
comment character, see
https://github.com/apache/whirr/blob/trunk/core/src/main/resources/META-INF/services/org.apache.whirr.ClusterController.
Is there any reason why headers can't be added to the test ones too?
Cheers,
Tom
On Mon, Jun 10, 2013 at 7:26 PM, Ignasi <ig...@gmail.com> wrote:
> Thanks for checking Tom!
>
> Please, apologize me if this has already been answered, but I am
> really confused and I still don't get what should be done with the
> headers in test files (and it is important to have it clear because it
> affects what and how we can perform tests).
>
> I really think that XML files in src/test/resources should not have
> test headers. As said in other threads, those XML files are used just
> to verify the output of the cloud APIs. In clouds we have two kinds of
> tests:
>
> * Expect tests (offline): These tests just mock the cloud apis, and we
> use them to properly check that we are generating the requests as
> expected, and we are properly parsing the responses. The XML/JSON
> files in src/test/resources are just a copy of the real responses the
> cloud apis return, and we use them to build the mock responses as
> close to the reality as possible.
> * Live tests (online): These tests actually execute the requests
> against the cloud apis, and check the behavior against real cloud
> providers.
>
> Having this clear, I think those XML/JSON files fail into the
> "non-creative" category (they are just a copies of real http response
> bodies). Furthermore, since the real cloud apis don't return an HTTP
> response body with license headers, our "response templates" shouldn't
> have them either. That's why we think it is important to keep those
> files without the license headers; adding them will make it impossible
> to properly add live tests that check the response bodies.
>
>
> Does this make sense? If it does, can we keep JSON/XML files in the
> src/test/resources directory without the license headers?
>
>
> Thanks for your patience, mentors!
>
>
> Ignasi
>
>
> On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
>> And there are a good number of tests failing because XML files aren't
>> expected to have headers. I'll see what I can do.
>>
>> A.
>>
>> On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <an...@gmail.com>wrote:
>>
>>> Also, if those don't full under the category of "non-creative", I don't
>>> know what does. =)
>>>
>>> A.
>>>
>>>
>>> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <an...@gmail.com>wrote:
>>>
>>>> Ok, the META-INF/services stuff will need at least test changes to work
>>>> right with license headers added, if not full blown code changes, so that
>>>> won't be happening.
>>>>
>>>> A.
>>>>
>>>>
>>>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <an...@gmail.com>wrote:
>>>>
>>>>> Will do.
>>>>>
>>>>> A.
>>>>>
>>>>>
>>>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
>>>>>
>>>>>> I ran through basic checks on the source packages (checksums,
>>>>>> signatures, license headers, included binary files, LICENSE, NOTICE,
>>>>>> DISCLAIMER), and generally they look good. In addition to the points
>>>>>> made on
>>>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>>>>>> the only thing I would add is that RAT flags up lots of missing
>>>>>> license headers. Many of them are JSON which can be excluded since
>>>>>> JSON doesn't support comments (as discussed above), or tests, but the
>>>>>> META-INF/services, YAML, XML, and properties files should have license
>>>>>> headers added for RC3 if possible.
>>>>>>
>>>>>> Cheers,
>>>>>> Tom
>>>>>>
>>>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
>>>>>> wrote:
>>>>>> > David, have you had a chance to take a look at my comments about the
>>>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>>>>>> >
>>>>>> > I've also noted my doubts on the wiki so the conclusions for each one
>>>>>> > can be reflected there.
>>>>>> >
>>>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>>>>>> >> Dealt with for RC3, and noted on
>>>>>> >>
>>>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>>>>> >>
>>>>>> >> A.
>>>>>> >>
>>>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>>>>>> >>
>>>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
>>>>>> andrew.bayer@gmail.com>
>>>>>> >>> wrote:
>>>>>> >>> > I'm extending the vote another 24 hours for our mentors to
>>>>>> respond.
>>>>>> >>> >
>>>>>> >>> > A.
>>>>>> >>> >
>>>>>> >>> >
>>>>>> >>>
>>>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>>>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
>>>>>> shape.
>>>>>> >>>
>>>>>> >>> --David
>>>>>> >>>
>>>>>>
>>>>>
>>>>>
>>>>
>>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Ignasi <ig...@gmail.com>.
Thanks for checking Tom!
Please, apologize me if this has already been answered, but I am
really confused and I still don't get what should be done with the
headers in test files (and it is important to have it clear because it
affects what and how we can perform tests).
I really think that XML files in src/test/resources should not have
test headers. As said in other threads, those XML files are used just
to verify the output of the cloud APIs. In clouds we have two kinds of
tests:
* Expect tests (offline): These tests just mock the cloud apis, and we
use them to properly check that we are generating the requests as
expected, and we are properly parsing the responses. The XML/JSON
files in src/test/resources are just a copy of the real responses the
cloud apis return, and we use them to build the mock responses as
close to the reality as possible.
* Live tests (online): These tests actually execute the requests
against the cloud apis, and check the behavior against real cloud
providers.
Having this clear, I think those XML/JSON files fail into the
"non-creative" category (they are just a copies of real http response
bodies). Furthermore, since the real cloud apis don't return an HTTP
response body with license headers, our "response templates" shouldn't
have them either. That's why we think it is important to keep those
files without the license headers; adding them will make it impossible
to properly add live tests that check the response bodies.
Does this make sense? If it does, can we keep JSON/XML files in the
src/test/resources directory without the license headers?
Thanks for your patience, mentors!
Ignasi
On 10 June 2013 19:56, Andrew Bayer <an...@gmail.com> wrote:
> And there are a good number of tests failing because XML files aren't
> expected to have headers. I'll see what I can do.
>
> A.
>
> On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <an...@gmail.com>wrote:
>
>> Also, if those don't full under the category of "non-creative", I don't
>> know what does. =)
>>
>> A.
>>
>>
>> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <an...@gmail.com>wrote:
>>
>>> Ok, the META-INF/services stuff will need at least test changes to work
>>> right with license headers added, if not full blown code changes, so that
>>> won't be happening.
>>>
>>> A.
>>>
>>>
>>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <an...@gmail.com>wrote:
>>>
>>>> Will do.
>>>>
>>>> A.
>>>>
>>>>
>>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
>>>>
>>>>> I ran through basic checks on the source packages (checksums,
>>>>> signatures, license headers, included binary files, LICENSE, NOTICE,
>>>>> DISCLAIMER), and generally they look good. In addition to the points
>>>>> made on
>>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>>>>> the only thing I would add is that RAT flags up lots of missing
>>>>> license headers. Many of them are JSON which can be excluded since
>>>>> JSON doesn't support comments (as discussed above), or tests, but the
>>>>> META-INF/services, YAML, XML, and properties files should have license
>>>>> headers added for RC3 if possible.
>>>>>
>>>>> Cheers,
>>>>> Tom
>>>>>
>>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
>>>>> wrote:
>>>>> > David, have you had a chance to take a look at my comments about the
>>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>>>>> >
>>>>> > I've also noted my doubts on the wiki so the conclusions for each one
>>>>> > can be reflected there.
>>>>> >
>>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>>>>> >> Dealt with for RC3, and noted on
>>>>> >>
>>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>>>> >>
>>>>> >> A.
>>>>> >>
>>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>>>>> >>
>>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
>>>>> andrew.bayer@gmail.com>
>>>>> >>> wrote:
>>>>> >>> > I'm extending the vote another 24 hours for our mentors to
>>>>> respond.
>>>>> >>> >
>>>>> >>> > A.
>>>>> >>> >
>>>>> >>> >
>>>>> >>>
>>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
>>>>> shape.
>>>>> >>>
>>>>> >>> --David
>>>>> >>>
>>>>>
>>>>
>>>>
>>>
>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
And there are a good number of tests failing because XML files aren't
expected to have headers. I'll see what I can do.
A.
On Mon, Jun 10, 2013 at 10:54 AM, Andrew Bayer <an...@gmail.com>wrote:
> Also, if those don't full under the category of "non-creative", I don't
> know what does. =)
>
> A.
>
>
> On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <an...@gmail.com>wrote:
>
>> Ok, the META-INF/services stuff will need at least test changes to work
>> right with license headers added, if not full blown code changes, so that
>> won't be happening.
>>
>> A.
>>
>>
>> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <an...@gmail.com>wrote:
>>
>>> Will do.
>>>
>>> A.
>>>
>>>
>>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
>>>
>>>> I ran through basic checks on the source packages (checksums,
>>>> signatures, license headers, included binary files, LICENSE, NOTICE,
>>>> DISCLAIMER), and generally they look good. In addition to the points
>>>> made on
>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>>>> the only thing I would add is that RAT flags up lots of missing
>>>> license headers. Many of them are JSON which can be excluded since
>>>> JSON doesn't support comments (as discussed above), or tests, but the
>>>> META-INF/services, YAML, XML, and properties files should have license
>>>> headers added for RC3 if possible.
>>>>
>>>> Cheers,
>>>> Tom
>>>>
>>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com>
>>>> wrote:
>>>> > David, have you had a chance to take a look at my comments about the
>>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>>>> >
>>>> > I've also noted my doubts on the wiki so the conclusions for each one
>>>> > can be reflected there.
>>>> >
>>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>>>> >> Dealt with for RC3, and noted on
>>>> >>
>>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>>> >>
>>>> >> A.
>>>> >>
>>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>>>> >>
>>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <
>>>> andrew.bayer@gmail.com>
>>>> >>> wrote:
>>>> >>> > I'm extending the vote another 24 hours for our mentors to
>>>> respond.
>>>> >>> >
>>>> >>> > A.
>>>> >>> >
>>>> >>> >
>>>> >>>
>>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>>>> >>> LICENSE.txt should be purged. Aside from that it looks in good
>>>> shape.
>>>> >>>
>>>> >>> --David
>>>> >>>
>>>>
>>>
>>>
>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Also, if those don't full under the category of "non-creative", I don't
know what does. =)
A.
On Mon, Jun 10, 2013 at 10:48 AM, Andrew Bayer <an...@gmail.com>wrote:
> Ok, the META-INF/services stuff will need at least test changes to work
> right with license headers added, if not full blown code changes, so that
> won't be happening.
>
> A.
>
>
> On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <an...@gmail.com>wrote:
>
>> Will do.
>>
>> A.
>>
>>
>> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
>>
>>> I ran through basic checks on the source packages (checksums,
>>> signatures, license headers, included binary files, LICENSE, NOTICE,
>>> DISCLAIMER), and generally they look good. In addition to the points
>>> made on
>>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>>> the only thing I would add is that RAT flags up lots of missing
>>> license headers. Many of them are JSON which can be excluded since
>>> JSON doesn't support comments (as discussed above), or tests, but the
>>> META-INF/services, YAML, XML, and properties files should have license
>>> headers added for RC3 if possible.
>>>
>>> Cheers,
>>> Tom
>>>
>>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com> wrote:
>>> > David, have you had a chance to take a look at my comments about the
>>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>>> >
>>> > I've also noted my doubts on the wiki so the conclusions for each one
>>> > can be reflected there.
>>> >
>>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>>> >> Dealt with for RC3, and noted on
>>> >> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>> >>
>>> >> A.
>>> >>
>>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>>> >>
>>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <andrew.bayer@gmail.com
>>> >
>>> >>> wrote:
>>> >>> > I'm extending the vote another 24 hours for our mentors to respond.
>>> >>> >
>>> >>> > A.
>>> >>> >
>>> >>> >
>>> >>>
>>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>>> >>> LICENSE.txt should be purged. Aside from that it looks in good shape.
>>> >>>
>>> >>> --David
>>> >>>
>>>
>>
>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Ok, the META-INF/services stuff will need at least test changes to work
right with license headers added, if not full blown code changes, so that
won't be happening.
A.
On Mon, Jun 10, 2013 at 9:41 AM, Andrew Bayer <an...@gmail.com>wrote:
> Will do.
>
> A.
>
>
> On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
>
>> I ran through basic checks on the source packages (checksums,
>> signatures, license headers, included binary files, LICENSE, NOTICE,
>> DISCLAIMER), and generally they look good. In addition to the points
>> made on
>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
>> the only thing I would add is that RAT flags up lots of missing
>> license headers. Many of them are JSON which can be excluded since
>> JSON doesn't support comments (as discussed above), or tests, but the
>> META-INF/services, YAML, XML, and properties files should have license
>> headers added for RC3 if possible.
>>
>> Cheers,
>> Tom
>>
>> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com> wrote:
>> > David, have you had a chance to take a look at my comments about the
>> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>> >
>> > I've also noted my doubts on the wiki so the conclusions for each one
>> > can be reflected there.
>> >
>> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>> >> Dealt with for RC3, and noted on
>> >> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>> >>
>> >> A.
>> >>
>> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>> >>
>> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com>
>> >>> wrote:
>> >>> > I'm extending the vote another 24 hours for our mentors to respond.
>> >>> >
>> >>> > A.
>> >>> >
>> >>> >
>> >>>
>> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>> >>> LICENSE.txt should be purged. Aside from that it looks in good shape.
>> >>>
>> >>> --David
>> >>>
>>
>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Will do.
A.
On Mon, Jun 10, 2013 at 9:21 AM, Tom White <to...@cloudera.com> wrote:
> I ran through basic checks on the source packages (checksums,
> signatures, license headers, included binary files, LICENSE, NOTICE,
> DISCLAIMER), and generally they look good. In addition to the points
> made on
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
> the only thing I would add is that RAT flags up lots of missing
> license headers. Many of them are JSON which can be excluded since
> JSON doesn't support comments (as discussed above), or tests, but the
> META-INF/services, YAML, XML, and properties files should have license
> headers added for RC3 if possible.
>
> Cheers,
> Tom
>
> On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com> wrote:
> > David, have you had a chance to take a look at my comments about the
> > json stuff? http://markmail.org/message/pk7efsmc6ewmake7
> >
> > I've also noted my doubts on the wiki so the conclusions for each one
> > can be reflected there.
> >
> > On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
> >> Dealt with for RC3, and noted on
> >> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
> >>
> >> A.
> >>
> >> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
> >>
> >>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com>
> >>> wrote:
> >>> > I'm extending the vote another 24 hours for our mentors to respond.
> >>> >
> >>> > A.
> >>> >
> >>> >
> >>>
> >>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
> >>> LICENSE.txt should be purged. Aside from that it looks in good shape.
> >>>
> >>> --David
> >>>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Tom White <to...@cloudera.com>.
I ran through basic checks on the source packages (checksums,
signatures, license headers, included binary files, LICENSE, NOTICE,
DISCLAIMER), and generally they look good. In addition to the points
made on https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues,
the only thing I would add is that RAT flags up lots of missing
license headers. Many of them are JSON which can be excluded since
JSON doesn't support comments (as discussed above), or tests, but the
META-INF/services, YAML, XML, and properties files should have license
headers added for RC3 if possible.
Cheers,
Tom
On Thu, Jun 6, 2013 at 4:03 PM, Ignasi <ig...@gmail.com> wrote:
> David, have you had a chance to take a look at my comments about the
> json stuff? http://markmail.org/message/pk7efsmc6ewmake7
>
> I've also noted my doubts on the wiki so the conclusions for each one
> can be reflected there.
>
> On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
>> Dealt with for RC3, and noted on
>> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>>
>> A.
>>
>> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>>
>>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com>
>>> wrote:
>>> > I'm extending the vote another 24 hours for our mentors to respond.
>>> >
>>> > A.
>>> >
>>> >
>>>
>>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>>> LICENSE.txt should be purged. Aside from that it looks in good shape.
>>>
>>> --David
>>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Ignasi <ig...@gmail.com>.
David, have you had a chance to take a look at my comments about the
json stuff? http://markmail.org/message/pk7efsmc6ewmake7
I've also noted my doubts on the wiki so the conclusions for each one
can be reflected there.
On 5 June 2013 23:07, Andrew Bayer <an...@gmail.com> wrote:
> Dealt with for RC3, and noted on
> https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
>
> A.
>
> On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
>
>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com>
>> wrote:
>> > I'm extending the vote another 24 hours for our mentors to respond.
>> >
>> > A.
>> >
>> >
>>
>> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
>> LICENSE.txt should be purged. Aside from that it looks in good shape.
>>
>> --David
>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Dealt with for RC3, and noted on
https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues
A.
On Wed, Jun 5, 2013 at 1:40 PM, David Nalley <da...@gnsa.us> wrote:
> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com>
> wrote:
> > I'm extending the vote another 24 hours for our mentors to respond.
> >
> > A.
> >
> >
>
> jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
> LICENSE.txt should be purged. Aside from that it looks in good shape.
>
> --David
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by David Nalley <da...@gnsa.us>.
On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com> wrote:
> I'm extending the vote another 24 hours for our mentors to respond.
>
> A.
>
>
jclouds-karaf has both LICENSE and LICENSE.txt and I'd argue that
LICENSE.txt should be purged. Aside from that it looks in good shape.
--David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Ignasi <ig...@gmail.com>.
Thanks for the feedback David,
Those json files do not have the license headers, because they are in
the src/test/resources package and are only intended to be used as
test expectation data. This was already discussed in [1] and the
conclusion was that the licensing creativity clause justified this.
Should we change something here?
Regarding the json files with copyright, it is not the file itself
what is copyrighted. Those files contain the json representation (the
body of a Chef REST API response) of a cookbook. So the cookbook
itself has a license, and if you download the cookbook from the
Opscode Community site, then you have to respect it; but those json
files are just the body of a REST response, which includes the
metadata of the cookbook (and its licensing information too); but it
is just the Api response in json format; nothing we are
re-distributing or similar.
The resources/LICENSE.txt and resources/NOTICE.txt have the same
purpose than the ones in the resource folder in the jclouds main
distribution. Since jclouds-chef is in a different repository, the
resources directory was also included there. Should it be removed?
Thanks for your feedback,
Ignasi
[1] http://markmail.org/message/pg2hxols62sfas3z
On 4 June 2013 15:06, David Nalley <da...@gnsa.us> wrote:
> On Mon, Jun 3, 2013 at 11:03 PM, David Nalley <da...@gnsa.us> wrote:
>> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com> wrote:
>>> I'm extending the vote another 24 hours for our mentors to respond.
>>>
>>> A.
>>>
>>
>
> Looking at jclouds-chef
>
> There are a number of json files that don't have license headers. I
> understand json doesn't really have a concept of comments.
> That said - there are a number of copyright declarations in some of
> those json files.
>
> A side note - those json files are copyrighted by other individuals,
> and I think will require NOTICE attributions.
>
> That said - it's a thorny problem, should we try and emulate the
> copyright statement of some of the json files that have copyright
> statements in them?
>
> I also wonder about resources/LICENSE.txt and resources/NOTICE.txt -
> why are those present?
>
> --David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
FYI, I've created a wiki page at
https://wiki.apache.org/jclouds/1.6.1%20Incubating%20Release%20Issues for
noting issues encountered in RCs and when they're fixed.
A.
On Tue, Jun 4, 2013 at 8:33 AM, Andrew Phillips <ap...@qrmedia.com>wrote:
> Yeah, I'll remove that - I have an unfortunate tendency to leave kruft
>> sitting around after it stops being used. =) I'll nuke the resources dir
>> where appropriate.
>>
>
> We should then also be able to test whether we can remove the exclusion in
> project/pom.xml starting in line 995 [1] or whether it needs to be kept.
>
> ap
>
> [1] https://git-wip-us.apache.org/**repos/asf?p=incubator-jclouds.**
> git;a=blob;f=project/pom.xml;**h=**f93487a5dae0b3b9346b78bfcc230b**
> be5af3d5ef;hb=1.6.x<https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blob;f=project/pom.xml;h=f93487a5dae0b3b9346b78bfcc230bbe5af3d5ef;hb=1.6.x>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Phillips <ap...@qrmedia.com>.
> Yeah, I'll remove that - I have an unfortunate tendency to leave
> kruft sitting around after it stops being used. =) I'll nuke the
> resources dir where appropriate.
We should then also be able to test whether we can remove the
exclusion in project/pom.xml starting in line 995 [1] or whether it
needs to be kept.
ap
[1]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blob;f=project/pom.xml;h=f93487a5dae0b3b9346b78bfcc230bbe5af3d5ef;hb=1.6.x
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Yeah, I'll remove that - I have an unfortunate tendency to leave kruft sitting around after it stops being used. =) I'll nuke the resources dir where appropriate.
A.
On Jun 4, 2013, at 6:36 AM, Andrew Phillips <ap...@qrmedia.com> wrote:
>> I also wonder about resources/LICENSE.txt and resources/NOTICE.txt -
>> why are those present?
>
> If you're referring to the ones in the main jclouds repo [1], these are bundled up as a "shared resource" as part of the build process to be included in all the JARs that are produced.
>
> The *usage* of this shared bundled was removed by Andrew B in a recent commit [3]. If that works as intended we're now getting any required resources included automatically, then we shouldn't need the entire remote resources generation any more.
>
> ap
>
> [1] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tree;f=resources;h=75981cb023e72d98c74fe6a7fd23a9d90d69f1ea;hb=d113b0ba63e1335ca4aa007036e0b0678934841f
> [2] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blob;f=resources/pom.xml;h=b1fd20e5a49b6ec85ade6c44d50dbdc292a273d4;hb=d113b0ba63e1335ca4aa007036e0b0678934841f
> [3] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blobdiff;f=project/pom.xml;h=e1ece9065c36d2a37d396c3d25662ea7ca38fd1e;hp=61f22a42b9650212edc336f4c32668d1b7057e9e;hb=a1c09f8391c1b6e87424c0fce2d4a0eb928bb060;hpb=fdd9f08f40d0881666e30fc9cc2fea98709e07a4
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Phillips <ap...@qrmedia.com>.
> I also wonder about resources/LICENSE.txt and resources/NOTICE.txt -
> why are those present?
If you're referring to the ones in the main jclouds repo [1], these
are bundled up as a "shared resource" as part of the build process to
be included in all the JARs that are produced.
The *usage* of this shared bundled was removed by Andrew B in a recent
commit [3]. If that works as intended we're now getting any required
resources included automatically, then we shouldn't need the entire
remote resources generation any more.
ap
[1]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=tree;f=resources;h=75981cb023e72d98c74fe6a7fd23a9d90d69f1ea;hb=d113b0ba63e1335ca4aa007036e0b0678934841f
[2]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blob;f=resources/pom.xml;h=b1fd20e5a49b6ec85ade6c44d50dbdc292a273d4;hb=d113b0ba63e1335ca4aa007036e0b0678934841f
[3]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=blobdiff;f=project/pom.xml;h=e1ece9065c36d2a37d396c3d25662ea7ca38fd1e;hp=61f22a42b9650212edc336f4c32668d1b7057e9e;hb=a1c09f8391c1b6e87424c0fce2d4a0eb928bb060;hpb=fdd9f08f40d0881666e30fc9cc2fea98709e07a4
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by David Nalley <da...@gnsa.us>.
On Mon, Jun 3, 2013 at 11:03 PM, David Nalley <da...@gnsa.us> wrote:
> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com> wrote:
>> I'm extending the vote another 24 hours for our mentors to respond.
>>
>> A.
>>
>
Looking at jclouds-chef
There are a number of json files that don't have license headers. I
understand json doesn't really have a concept of comments.
That said - there are a number of copyright declarations in some of
those json files.
A side note - those json files are copyrighted by other individuals,
and I think will require NOTICE attributions.
That said - it's a thorny problem, should we try and emulate the
copyright statement of some of the json files that have copyright
statements in them?
I also wonder about resources/LICENSE.txt and resources/NOTICE.txt -
why are those present?
--David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Okiedokie - consider this an "until the mentors have been able to dive through all the release artifacts" period of undetermined length, with RC3 coming when the mentors believe they've found all the problems and we've addressed them for the next build.
Those license headers are...embarrassing. Everything in all the repos is covered by the SGA, so yeah, I'll fix those and get rid of the redundant LICENSE.txt.
A.
On Jun 3, 2013, at 8:03 PM, David Nalley <da...@gnsa.us> wrote:
> On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com> wrote:
>> I'm extending the vote another 24 hours for our mentors to respond.
>>
>> A.
>>
>
> So realistically, and especially with the first one there isn't a
> practical point in an end date. It takes a significant chunk of time
> to audit a release, esp the first release, particularly a
> multi-artifact release like jclouds.
>
> I think we are getting close with -cli, but let me describe issues
> I've found with jclouds-cli
>
> LICENSE and LICENSE.txt exist in the root directory and are not the
> same document.
>
> The following files do not list that they are licensed to the ASF in
> the license headers, yet also don't have a NOTICE entry.
> If they are part of the SGA that Adrian executed, we should change the
> headers accordingly. If they are not - we need to annotate the NOTICE
> file with the copyright holders copyright notice and claim.
>
> ./assembly/pom.xml:~ Copyright (C) 2012, the original authors
> ./assembly/src/main/assembly/win.xml: ~ Copyright (C) 2012, the
> original authors
> ./assembly/src/main/assembly/unix.xml: ~ Copyright (C) 2012, the
> original authors
> ./assembly/src/main/filtered-resources/etc/startup.properties:#
> Copyright (C) 2012, the original authors
> ./assembly/src/main/filtered-resources/etc/custom.properties:#
> Copyright (C) 2012, the original authors
> ./branding/pom.xml:~ Copyright (C) 2012, the original authors
> ./pom.xml:~ Copyright (C) 2012, the original authors
> ./runner/pom.xml:~ Copyright (C) 2012, the original authors
> ./runner/src/main/java/org/jclouds/cli/runner/Main.java: * Copyright
> (C) 2012, the original authors
>
> I don't see any other problems on -cli at present, but other mentors
> may find additional problems.
> I'll try and start working through the others tomorrow.
>
> --David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by David Nalley <da...@gnsa.us>.
On Mon, Jun 3, 2013 at 7:58 PM, Andrew Bayer <an...@gmail.com> wrote:
> I'm extending the vote another 24 hours for our mentors to respond.
>
> A.
>
So realistically, and especially with the first one there isn't a
practical point in an end date. It takes a significant chunk of time
to audit a release, esp the first release, particularly a
multi-artifact release like jclouds.
I think we are getting close with -cli, but let me describe issues
I've found with jclouds-cli
LICENSE and LICENSE.txt exist in the root directory and are not the
same document.
The following files do not list that they are licensed to the ASF in
the license headers, yet also don't have a NOTICE entry.
If they are part of the SGA that Adrian executed, we should change the
headers accordingly. If they are not - we need to annotate the NOTICE
file with the copyright holders copyright notice and claim.
./assembly/pom.xml:~ Copyright (C) 2012, the original authors
./assembly/src/main/assembly/win.xml: ~ Copyright (C) 2012, the
original authors
./assembly/src/main/assembly/unix.xml: ~ Copyright (C) 2012, the
original authors
./assembly/src/main/filtered-resources/etc/startup.properties:#
Copyright (C) 2012, the original authors
./assembly/src/main/filtered-resources/etc/custom.properties:#
Copyright (C) 2012, the original authors
./branding/pom.xml:~ Copyright (C) 2012, the original authors
./pom.xml:~ Copyright (C) 2012, the original authors
./runner/pom.xml:~ Copyright (C) 2012, the original authors
./runner/src/main/java/org/jclouds/cli/runner/Main.java: * Copyright
(C) 2012, the original authors
I don't see any other problems on -cli at present, but other mentors
may find additional problems.
I'll try and start working through the others tomorrow.
--David
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
I'm extending the vote another 24 hours for our mentors to respond.
A.
On Jun 3, 2013, at 3:44 PM, Matt Stephenson <ma...@apache.org> wrote:
> I did some similar tests, took a look at the generated artifacts from a
> build, compared them to the artifacts here, and checked out the sigs.
> Andrew Bayer's key is on the public key servers, so I'm comfortable with
> the signature, I guess we should all sign eachother's keys if that matters
> enough to people.
>
> I'm happy.
>
> +1, PPMC
>
>
> On Mon, Jun 3, 2013 at 2:15 PM, Andrew Phillips <ap...@qrmedia.com>wrote:
>
>> Tests performed (many thanks for the guidance, David and Andrew B!):
>>
>> * checked out the release tag (verified that the commit ID matches the
>> link being voted on) and successfully ran 'mvn clean package' with a clean
>> Maven repo
>>
>> * successfully ran 'mvn apache-rat:check' to verify licenses
>>
>> * built a source release ZIP using 'mvn clean package -Papache-release'
>> and compared with [1] to verify that size, CRC and number of files match
>>
>> * [1] contains one NOTICE and LICENSE file. LICENSE file looks OK but
>> NOTICE file begins "jclouds" and not "Apache jclouds" (see
>> http://apache.org/legal/src-**headers.html#notice<http://apache.org/legal/src-headers.html#notice>
>> )
>>
>> @mentors: is that a blocker?
>>
>> * matched GPG key in KEYS file against http://pgp.mit.edu:11371/pks/**
>> lookup?op=get&search=**0xB12E3E253ADD02D6<http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xB12E3E253ADD02D6>
>>
>> * verified MD5 of ASC against published MD5 [2]
>>
>> * GPG sig of the release ZIP verified:
>>
>> C:\Program Files (x86)\GnuPG>gpg --verify ...\jclouds-1.6.1-incubating-**
>> source-release.zip.asc
>> gpg: Signature made 05/31/13 17:38:15 using RSA key ID 3ADD02D6
>> gpg: Good signature from "Andrew Bayer (CODE SIGNING KEY) <
>> abayer@apache.org>"
>> gpg: WARNING: This key is not certified with a trusted signature!
>> gpg: There is no indication that the signature belongs to the
>> owner.
>> Primary key fingerprint: E2F3 1807 1F65 6A62 F88F 252C B12E 3E25 3ADD 02D6
>>
>> @abayer: trust web?
>>
>> Vote: +1 PPMC.
>>
>> Thanks for all the preparation, Andrew B!
>>
>> ap
>>
>> [1] https://repository.apache.org/**content/repositories/**
>> orgapachejclouds-043/org/**apache/jclouds/jclouds/1.6.1-**
>> incubating/jclouds-1.6.1-**incubating-source-release.zip<https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip>
>> [2] https://repository.apache.org/**content/repositories/**
>> orgapachejclouds-043/org/**apache/jclouds/jclouds/1.6.1-**
>> incubating/jclouds-1.6.1-**incubating-source-release.zip.**asc.md5<https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip.asc.md5>
>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Matt Stephenson <ma...@apache.org>.
I did some similar tests, took a look at the generated artifacts from a
build, compared them to the artifacts here, and checked out the sigs.
Andrew Bayer's key is on the public key servers, so I'm comfortable with
the signature, I guess we should all sign eachother's keys if that matters
enough to people.
I'm happy.
+1, PPMC
On Mon, Jun 3, 2013 at 2:15 PM, Andrew Phillips <ap...@qrmedia.com>wrote:
> Tests performed (many thanks for the guidance, David and Andrew B!):
>
> * checked out the release tag (verified that the commit ID matches the
> link being voted on) and successfully ran 'mvn clean package' with a clean
> Maven repo
>
> * successfully ran 'mvn apache-rat:check' to verify licenses
>
> * built a source release ZIP using 'mvn clean package -Papache-release'
> and compared with [1] to verify that size, CRC and number of files match
>
> * [1] contains one NOTICE and LICENSE file. LICENSE file looks OK but
> NOTICE file begins "jclouds" and not "Apache jclouds" (see
> http://apache.org/legal/src-**headers.html#notice<http://apache.org/legal/src-headers.html#notice>
> )
>
> @mentors: is that a blocker?
>
> * matched GPG key in KEYS file against http://pgp.mit.edu:11371/pks/**
> lookup?op=get&search=**0xB12E3E253ADD02D6<http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xB12E3E253ADD02D6>
>
> * verified MD5 of ASC against published MD5 [2]
>
> * GPG sig of the release ZIP verified:
>
> C:\Program Files (x86)\GnuPG>gpg --verify ...\jclouds-1.6.1-incubating-**
> source-release.zip.asc
> gpg: Signature made 05/31/13 17:38:15 using RSA key ID 3ADD02D6
> gpg: Good signature from "Andrew Bayer (CODE SIGNING KEY) <
> abayer@apache.org>"
> gpg: WARNING: This key is not certified with a trusted signature!
> gpg: There is no indication that the signature belongs to the
> owner.
> Primary key fingerprint: E2F3 1807 1F65 6A62 F88F 252C B12E 3E25 3ADD 02D6
>
> @abayer: trust web?
>
> Vote: +1 PPMC.
>
> Thanks for all the preparation, Andrew B!
>
> ap
>
> [1] https://repository.apache.org/**content/repositories/**
> orgapachejclouds-043/org/**apache/jclouds/jclouds/1.6.1-**
> incubating/jclouds-1.6.1-**incubating-source-release.zip<https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip>
> [2] https://repository.apache.org/**content/repositories/**
> orgapachejclouds-043/org/**apache/jclouds/jclouds/1.6.1-**
> incubating/jclouds-1.6.1-**incubating-source-release.zip.**asc.md5<https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip.asc.md5>
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Phillips <ap...@qrmedia.com>.
Tests performed (many thanks for the guidance, David and Andrew B!):
* checked out the release tag (verified that the commit ID matches the
link being voted on) and successfully ran 'mvn clean package' with a
clean Maven repo
* successfully ran 'mvn apache-rat:check' to verify licenses
* built a source release ZIP using 'mvn clean package
-Papache-release' and compared with [1] to verify that size, CRC and
number of files match
* [1] contains one NOTICE and LICENSE file. LICENSE file looks OK but
NOTICE file begins "jclouds" and not "Apache jclouds" (see
http://apache.org/legal/src-headers.html#notice)
@mentors: is that a blocker?
* matched GPG key in KEYS file against
http://pgp.mit.edu:11371/pks/lookup?op=get&search=0xB12E3E253ADD02D6
* verified MD5 of ASC against published MD5 [2]
* GPG sig of the release ZIP verified:
C:\Program Files (x86)\GnuPG>gpg --verify
...\jclouds-1.6.1-incubating-source-release.zip.asc
gpg: Signature made 05/31/13 17:38:15 using RSA key ID 3ADD02D6
gpg: Good signature from "Andrew Bayer (CODE SIGNING KEY) <ab...@apache.org>"
gpg: WARNING: This key is not certified with a trusted signature!
gpg: There is no indication that the signature belongs to the owner.
Primary key fingerprint: E2F3 1807 1F65 6A62 F88F 252C B12E 3E25 3ADD 02D6
@abayer: trust web?
Vote: +1 PPMC.
Thanks for all the preparation, Andrew B!
ap
[1]
https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip
[2]
https://repository.apache.org/content/repositories/orgapachejclouds-043/org/apache/jclouds/jclouds/1.6.1-incubating/jclouds-1.6.1-incubating-source-release.zip.asc.md5
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Ignasi <ig...@gmail.com>.
Performed the following tests:
* Build jclouds, jclouds-labs and jclouds-chef tags and all builds
succeeded (the build included RAT plugin execution and everything is
ok too)
* Run the live tests for Chef and Hosted Chef. All passing except two
in HostedChef: one due to a timeout, and the other one because of a
known issue in HostedChef, so everything is OK in our side.
* Downloaded the dependencies from the staging repos and tested the
integration with the ComputeService with jclouds-examples, chef-solo,
Abiquo and EC2, and everything is OK. Also verified the signatures of
the chef artifacts in the repo with Andrew's key in the jclouds KEYS
file and they are OK.
+1 binding, PPMC.
On 3 June 2013 17:13, Andrew Bayer <an...@gmail.com> wrote:
> 'cos I didn't know that was an option? =)
>
> people.a.o is back, fwiw.
>
> A.
>
> On Mon, Jun 3, 2013 at 4:46 AM, Olivier Lamy <ol...@apache.org> wrote:
>
>> I will but when people.a.o will be back.
>> Why not using svnpubsub for vote ?
>>
>> 2013/6/3 Andrew Bayer <an...@gmail.com>:
>> > And FYI, I'd be very, very appreciative of any mentors who can review RC2
>> > and vote.
>> >
>> > And I'm +1 binding, PPMC.
>> >
>> > A.
>> >
>> > On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <andrew.bayer@gmail.com
>> >wrote:
>> >
>> >> My tendency in this direction was prompted by David and others pointing
>> >> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
>> >> branch. That does seem like poor form, even though the tag was pushed.
>> I'm
>> >> definitely open to advice and suggestions on this - I've got experience
>> >> with the release plugin, and I've got experience with iterating RCs for
>> ASF
>> >> votes, but I don't have experience meshing the two. =)
>> >>
>> >> A.
>> >>
>> >>
>> >>
>> >> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
>> >> wrote:
>> >>
>> >> >> Re the difference with the rc0 commit and now - experience. The
>> revert
>> >> commits are so I could run the release plugin again for the next RC,
>> and I
>> >> didn't do the reverts until I cut the RC. I'm still figuring out the
>> best
>> >> way to do RCs Apache-style combined with the Maven release plugin...
>> >> >
>> >> > Cool, thanks for the explanation. I'm assuming from that that you feel
>> >> the "prepare-then-revert" style is preferred (vs. the
>> >> "prepare-locally-and-dont-commit").
>> >> >
>> >> > Fine with me, although I'd be interested to know what the thoughts are
>> >> around having multiple commits between the "prepare" and the "revert",
>> all
>> >> of which have a "release" version in their POMs.
>> >> >
>> >> > @mentors: any guidance you can contribute here?
>> >> >
>> >> > Thanks!
>> >> >
>> >> > ap
>> >>
>>
>>
>>
>> --
>> Olivier Lamy
>> Ecetera: http://ecetera.com.au
>> http://twitter.com/olamy | http://linkedin.com/in/olamy
>>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Will do. I didn't make ay changes to the binary assembly, other than its name, so it's not a surprise that there are changes needed there.
A.
On Jun 4, 2013, at 4:15 AM, Olivier Lamy <ol...@apache.org> wrote:
> hehe (have a look at helix release procedure: section 5 of
> http://helix.incubator.apache.org/releasing.html )
>
> I have some issues with jclouds-cli-assembly-1.6.1-incubating.zip/tar.gz
>
> It doesn't include:
> * DISCLAIMER
> * NOTICE (especially this one must contains "I. Included Software
> section" as it contains a lot of jars) (as a sample see this file in
> the Apache Karaf distrib). And BTW the other sections (II. Used
> Software and III. License Summary) I'm pretty maven can do that for
> you ;-) . All included jars have compliant license (mostly Apache
> except bouncycastle but MIT so ok)
> * LICENSE
>
> 2013/6/4 Andrew Bayer <an...@gmail.com>:
>> 'cos I didn't know that was an option? =)
>>
>> people.a.o is back, fwiw.
>>
>> A.
>>
>> On Mon, Jun 3, 2013 at 4:46 AM, Olivier Lamy <ol...@apache.org> wrote:
>>
>>> I will but when people.a.o will be back.
>>> Why not using svnpubsub for vote ?
>>>
>>> 2013/6/3 Andrew Bayer <an...@gmail.com>:
>>>> And FYI, I'd be very, very appreciative of any mentors who can review RC2
>>>> and vote.
>>>>
>>>> And I'm +1 binding, PPMC.
>>>>
>>>> A.
>>>>
>>>> On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <andrew.bayer@gmail.com
>>>> wrote:
>>>>
>>>>> My tendency in this direction was prompted by David and others pointing
>>>>> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
>>>>> branch. That does seem like poor form, even though the tag was pushed.
>>> I'm
>>>>> definitely open to advice and suggestions on this - I've got experience
>>>>> with the release plugin, and I've got experience with iterating RCs for
>>> ASF
>>>>> votes, but I don't have experience meshing the two. =)
>>>>>
>>>>> A.
>>>>>
>>>>>
>>>>>
>>>>> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
>>>>> wrote:
>>>>>
>>>>>>> Re the difference with the rc0 commit and now - experience. The
>>> revert
>>>>> commits are so I could run the release plugin again for the next RC,
>>> and I
>>>>> didn't do the reverts until I cut the RC. I'm still figuring out the
>>> best
>>>>> way to do RCs Apache-style combined with the Maven release plugin...
>>>>>>
>>>>>> Cool, thanks for the explanation. I'm assuming from that that you feel
>>>>> the "prepare-then-revert" style is preferred (vs. the
>>>>> "prepare-locally-and-dont-commit").
>>>>>>
>>>>>> Fine with me, although I'd be interested to know what the thoughts are
>>>>> around having multiple commits between the "prepare" and the "revert",
>>> all
>>>>> of which have a "release" version in their POMs.
>>>>>>
>>>>>> @mentors: any guidance you can contribute here?
>>>>>>
>>>>>> Thanks!
>>>>>>
>>>>>> ap
>>>>>
>>>
>>>
>>>
>>> --
>>> Olivier Lamy
>>> Ecetera: http://ecetera.com.au
>>> http://twitter.com/olamy | http://linkedin.com/in/olamy
>>>
>
>
>
> --
> Olivier Lamy
> Ecetera: http://ecetera.com.au
> http://twitter.com/olamy | http://linkedin.com/in/olamy
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Olivier Lamy <ol...@apache.org>.
hehe (have a look at helix release procedure: section 5 of
http://helix.incubator.apache.org/releasing.html )
I have some issues with jclouds-cli-assembly-1.6.1-incubating.zip/tar.gz
It doesn't include:
* DISCLAIMER
* NOTICE (especially this one must contains "I. Included Software
section" as it contains a lot of jars) (as a sample see this file in
the Apache Karaf distrib). And BTW the other sections (II. Used
Software and III. License Summary) I'm pretty maven can do that for
you ;-) . All included jars have compliant license (mostly Apache
except bouncycastle but MIT so ok)
* LICENSE
2013/6/4 Andrew Bayer <an...@gmail.com>:
> 'cos I didn't know that was an option? =)
>
> people.a.o is back, fwiw.
>
> A.
>
> On Mon, Jun 3, 2013 at 4:46 AM, Olivier Lamy <ol...@apache.org> wrote:
>
>> I will but when people.a.o will be back.
>> Why not using svnpubsub for vote ?
>>
>> 2013/6/3 Andrew Bayer <an...@gmail.com>:
>> > And FYI, I'd be very, very appreciative of any mentors who can review RC2
>> > and vote.
>> >
>> > And I'm +1 binding, PPMC.
>> >
>> > A.
>> >
>> > On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <andrew.bayer@gmail.com
>> >wrote:
>> >
>> >> My tendency in this direction was prompted by David and others pointing
>> >> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
>> >> branch. That does seem like poor form, even though the tag was pushed.
>> I'm
>> >> definitely open to advice and suggestions on this - I've got experience
>> >> with the release plugin, and I've got experience with iterating RCs for
>> ASF
>> >> votes, but I don't have experience meshing the two. =)
>> >>
>> >> A.
>> >>
>> >>
>> >>
>> >> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
>> >> wrote:
>> >>
>> >> >> Re the difference with the rc0 commit and now - experience. The
>> revert
>> >> commits are so I could run the release plugin again for the next RC,
>> and I
>> >> didn't do the reverts until I cut the RC. I'm still figuring out the
>> best
>> >> way to do RCs Apache-style combined with the Maven release plugin...
>> >> >
>> >> > Cool, thanks for the explanation. I'm assuming from that that you feel
>> >> the "prepare-then-revert" style is preferred (vs. the
>> >> "prepare-locally-and-dont-commit").
>> >> >
>> >> > Fine with me, although I'd be interested to know what the thoughts are
>> >> around having multiple commits between the "prepare" and the "revert",
>> all
>> >> of which have a "release" version in their POMs.
>> >> >
>> >> > @mentors: any guidance you can contribute here?
>> >> >
>> >> > Thanks!
>> >> >
>> >> > ap
>> >>
>>
>>
>>
>> --
>> Olivier Lamy
>> Ecetera: http://ecetera.com.au
>> http://twitter.com/olamy | http://linkedin.com/in/olamy
>>
--
Olivier Lamy
Ecetera: http://ecetera.com.au
http://twitter.com/olamy | http://linkedin.com/in/olamy
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
'cos I didn't know that was an option? =)
people.a.o is back, fwiw.
A.
On Mon, Jun 3, 2013 at 4:46 AM, Olivier Lamy <ol...@apache.org> wrote:
> I will but when people.a.o will be back.
> Why not using svnpubsub for vote ?
>
> 2013/6/3 Andrew Bayer <an...@gmail.com>:
> > And FYI, I'd be very, very appreciative of any mentors who can review RC2
> > and vote.
> >
> > And I'm +1 binding, PPMC.
> >
> > A.
> >
> > On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <andrew.bayer@gmail.com
> >wrote:
> >
> >> My tendency in this direction was prompted by David and others pointing
> >> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
> >> branch. That does seem like poor form, even though the tag was pushed.
> I'm
> >> definitely open to advice and suggestions on this - I've got experience
> >> with the release plugin, and I've got experience with iterating RCs for
> ASF
> >> votes, but I don't have experience meshing the two. =)
> >>
> >> A.
> >>
> >>
> >>
> >> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
> >> wrote:
> >>
> >> >> Re the difference with the rc0 commit and now - experience. The
> revert
> >> commits are so I could run the release plugin again for the next RC,
> and I
> >> didn't do the reverts until I cut the RC. I'm still figuring out the
> best
> >> way to do RCs Apache-style combined with the Maven release plugin...
> >> >
> >> > Cool, thanks for the explanation. I'm assuming from that that you feel
> >> the "prepare-then-revert" style is preferred (vs. the
> >> "prepare-locally-and-dont-commit").
> >> >
> >> > Fine with me, although I'd be interested to know what the thoughts are
> >> around having multiple commits between the "prepare" and the "revert",
> all
> >> of which have a "release" version in their POMs.
> >> >
> >> > @mentors: any guidance you can contribute here?
> >> >
> >> > Thanks!
> >> >
> >> > ap
> >>
>
>
>
> --
> Olivier Lamy
> Ecetera: http://ecetera.com.au
> http://twitter.com/olamy | http://linkedin.com/in/olamy
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Olivier Lamy <ol...@apache.org>.
I will but when people.a.o will be back.
Why not using svnpubsub for vote ?
2013/6/3 Andrew Bayer <an...@gmail.com>:
> And FYI, I'd be very, very appreciative of any mentors who can review RC2
> and vote.
>
> And I'm +1 binding, PPMC.
>
> A.
>
> On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <an...@gmail.com>wrote:
>
>> My tendency in this direction was prompted by David and others pointing
>> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
>> branch. That does seem like poor form, even though the tag was pushed. I'm
>> definitely open to advice and suggestions on this - I've got experience
>> with the release plugin, and I've got experience with iterating RCs for ASF
>> votes, but I don't have experience meshing the two. =)
>>
>> A.
>>
>>
>>
>> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
>> wrote:
>>
>> >> Re the difference with the rc0 commit and now - experience. The revert
>> commits are so I could run the release plugin again for the next RC, and I
>> didn't do the reverts until I cut the RC. I'm still figuring out the best
>> way to do RCs Apache-style combined with the Maven release plugin...
>> >
>> > Cool, thanks for the explanation. I'm assuming from that that you feel
>> the "prepare-then-revert" style is preferred (vs. the
>> "prepare-locally-and-dont-commit").
>> >
>> > Fine with me, although I'd be interested to know what the thoughts are
>> around having multiple commits between the "prepare" and the "revert", all
>> of which have a "release" version in their POMs.
>> >
>> > @mentors: any guidance you can contribute here?
>> >
>> > Thanks!
>> >
>> > ap
>>
--
Olivier Lamy
Ecetera: http://ecetera.com.au
http://twitter.com/olamy | http://linkedin.com/in/olamy
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
And FYI, I'd be very, very appreciative of any mentors who can review RC2
and vote.
And I'm +1 binding, PPMC.
A.
On Sun, Jun 2, 2013 at 11:03 AM, Andrew Bayer <an...@gmail.com>wrote:
> My tendency in this direction was prompted by David and others pointing
> out that the release commit for RC0 wasn't actually pushed to the 1.6.x
> branch. That does seem like poor form, even though the tag was pushed. I'm
> definitely open to advice and suggestions on this - I've got experience
> with the release plugin, and I've got experience with iterating RCs for ASF
> votes, but I don't have experience meshing the two. =)
>
> A.
>
>
>
> On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com>
> wrote:
>
> >> Re the difference with the rc0 commit and now - experience. The revert
> commits are so I could run the release plugin again for the next RC, and I
> didn't do the reverts until I cut the RC. I'm still figuring out the best
> way to do RCs Apache-style combined with the Maven release plugin...
> >
> > Cool, thanks for the explanation. I'm assuming from that that you feel
> the "prepare-then-revert" style is preferred (vs. the
> "prepare-locally-and-dont-commit").
> >
> > Fine with me, although I'd be interested to know what the thoughts are
> around having multiple commits between the "prepare" and the "revert", all
> of which have a "release" version in their POMs.
> >
> > @mentors: any guidance you can contribute here?
> >
> > Thanks!
> >
> > ap
>
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
My tendency in this direction was prompted by David and others pointing out that the release commit for RC0 wasn't actually pushed to the 1.6.x branch. That does seem like poor form, even though the tag was pushed. I'm definitely open to advice and suggestions on this - I've got experience with the release plugin, and I've got experience with iterating RCs for ASF votes, but I don't have experience meshing the two. =)
A.
On Jun 2, 2013, at 10:55 AM, Andrew Phillips <ap...@qrmedia.com> wrote:
>> Re the difference with the rc0 commit and now - experience. The revert commits are so I could run the release plugin again for the next RC, and I didn't do the reverts until I cut the RC. I'm still figuring out the best way to do RCs Apache-style combined with the Maven release plugin...
>
> Cool, thanks for the explanation. I'm assuming from that that you feel the "prepare-then-revert" style is preferred (vs. the "prepare-locally-and-dont-commit").
>
> Fine with me, although I'd be interested to know what the thoughts are around having multiple commits between the "prepare" and the "revert", all of which have a "release" version in their POMs.
>
> @mentors: any guidance you can contribute here?
>
> Thanks!
>
> ap
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Phillips <ap...@qrmedia.com>.
> Re the difference with the rc0 commit and now - experience. The
> revert commits are so I could run the release plugin again for the
> next RC, and I didn't do the reverts until I cut the RC. I'm still
> figuring out the best way to do RCs Apache-style combined with the
> Maven release plugin...
Cool, thanks for the explanation. I'm assuming from that that you feel
the "prepare-then-revert" style is preferred (vs. the
"prepare-locally-and-dont-commit").
Fine with me, although I'd be interested to know what the thoughts are
around having multiple commits between the "prepare" and the "revert",
all of which have a "release" version in their POMs.
@mentors: any guidance you can contribute here?
Thanks!
ap
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Bayer <an...@gmail.com>.
Re the difference with the rc0 commit and now - experience. The revert commits are so I could run the release plugin again for the next RC, and I didn't do the reverts until I cut the RC. I'm still figuring out the best way to do RCs Apache-style combined with the Maven release plugin...
A.
On Jun 1, 2013, at 4:27 AM, Andrew Phillips <ap...@qrmedia.com> wrote:
> Any reason why this time we have a "prepare for release" commit [1] and then, a couple of commits later (but not immediately afterwards), a revert of that commit [2]?
>
> I'm comparing it here to the "jclouds-1.6.1-incubating-rc0" tag [3], which was also a "prepare for release" commit, but which isn't in the 1.6.x tree and wasn't reverted.
>
> Also, assuming the prepare-then-revert model is the correct way to go, should the revert come immediately after the "prepare" commit?
>
> Thanks!
>
> ap
>
> [1] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=57718280be4bccea9e7885c5c3c38550f818d0dd
> [2] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=7828eef35957cbbb53b8ffd934c778fbe07dacc2
> [3] https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=d99dae8c1c7b0a311f094bcd8ad5bfb1975fffb1
Re: [VOTE] Release Apache jclouds 1.6.1-incubating, RC2
Posted by Andrew Phillips <ap...@qrmedia.com>.
Any reason why this time we have a "prepare for release" commit [1]
and then, a couple of commits later (but not immediately afterwards),
a revert of that commit [2]?
I'm comparing it here to the "jclouds-1.6.1-incubating-rc0" tag [3],
which was also a "prepare for release" commit, but which isn't in the
1.6.x tree and wasn't reverted.
Also, assuming the prepare-then-revert model is the correct way to go,
should the revert come immediately after the "prepare" commit?
Thanks!
ap
[1]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=57718280be4bccea9e7885c5c3c38550f818d0dd
[2]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=7828eef35957cbbb53b8ffd934c778fbe07dacc2
[3]
https://git-wip-us.apache.org/repos/asf?p=incubator-jclouds.git;a=commit;h=d99dae8c1c7b0a311f094bcd8ad5bfb1975fffb1