You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tuscany.apache.org by Simon Nash <na...@apache.org> on 2010/11/04 15:37:34 UTC
[VOTE] Release Tuscany SCA 1.6.1 RC1
Please review and vote on RC1 of the SCA Java 1.6.1 release.
The distribution artifacts, RAT report, Eclipse update site, and
Maven staging repository are available for review at:
http://people.apache.org/~nash/tuscany/1.6.1-RC1/
The release tag is at:
https://svn.apache.org/repos/asf/tuscany/sca-java-1.x/tags/1.6.1-RC1/
Here's my +1.
Simon
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Nash <na...@apache.org>.
Simon Nash wrote:
> Thanks for testing this. See comments inline below.
>
> Simon
>
> ant elder wrote:
>> On Thu, Nov 4, 2010 at 2:37 PM, Simon Nash <na...@apache.org> wrote:
>>> Please review and vote on RC1 of the SCA Java 1.6.1 release.
>>>
>>> The distribution artifacts, RAT report, Eclipse update site, and
>>> Maven staging repository are available for review at:
>>>
>>> http://people.apache.org/~nash/tuscany/1.6.1-RC1/
>>>
>>> The release tag is at:
>>>
>>> https://svn.apache.org/repos/asf/tuscany/sca-java-1.x/tags/1.6.1-RC1/
>>>
>>> Here's my +1.
>>>
>>> Simon
>>>
>>>
>>
>> I've had a look and run some samples, the only issues I've found are:
>>
>> - itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
>> build and the reactor build when building from the top.
> >
> This unfortunately crept in as a result of a late fix to the release
> branch that wasn't correctly merged into the tag.
>
> I scanned the whole codebase to check for other SNAPSHOT occurrences
> and I found two more in tools/eclipse/features/core/pom.xml and
> tools/eclipse/plugins/core/pom.xml. Given that this isn't a single
> isolated occurrence, I think it's worth doing a respin to fix it.
>
>> - tools\maven\maven-incremental-build fails with: Error extracting
>> plugin descriptor: 'No mojo definitions were found for plugin:
>> org.apache.tuscany.sca:tuscany-maven-incremental-build.'
>>
> After much puzzlement I think I have figured this out. The Mojo source
> code
> for this module has been incorrectly stripped from the source distribution
> because of an error in distribution/src/main/assembly/src.xml. This was
> wrong in 1.6 as well. I thought I had compared the tag with the source
> distribution but somehow I overlooked this difference.
>
>> - using Tomcat 7 i see a warning about a memory leak:
>> SEVERE: The web application [/sample-calculator-webapp] created a
>> ThreadLocal with key of type [null] (value
>> [org.apache.tuscany.sca.core.invocation.ThreadMessageContext$1@1b071c0])
>> and a value of type
>> [org.apache.tuscany.sca.core.invocation.MessageImpl] (value
>> [org.apache.tuscany.sca.core.invocation.MessageImpl@fbf51d]) but
>> failed to remove it when the web application was stopped. This is very
>> likely to create a memory leak.
>>
> I understand what's going on here and I'm looking to see if there's a
> simple fix.
>
I've got to the bottom of this and there does appear to be a simple fix
(see TUSCANY-3770). I have verified that the fix works for the sample
calculator webapp and I'm testing it on a full build now.
Simon
> Simon
>
>> Those first two issues should really be fixed but if there's no other
>> reason for re-spinning then +1 from me with the proviso that there
>> would be a errata link on the download page mentioning the build
>> problems and how to bypass them.
>>
>> ...ant
>>
>>
>
>
>
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Luciano Resende <lu...@gmail.com>.
On Mon, Nov 8, 2010 at 6:54 AM, Simon Nash <na...@apache.org> wrote:
> Thanks for testing this. See comments inline below.
>
Simon, from the comments on this thread I'm assuming you are going to
create a new RC, so I'm going to delay my vote and wait for the next
RC.
BTW, you are doing an excellent job as RC.
--
Luciano Resende
http://people.apache.org/~lresende
http://twitter.com/lresende1975
http://lresende.blogspot.com/
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Nash <na...@apache.org>.
Thanks for testing this. See comments inline below.
Simon
ant elder wrote:
> On Thu, Nov 4, 2010 at 2:37 PM, Simon Nash <na...@apache.org> wrote:
>> Please review and vote on RC1 of the SCA Java 1.6.1 release.
>>
>> The distribution artifacts, RAT report, Eclipse update site, and
>> Maven staging repository are available for review at:
>>
>> http://people.apache.org/~nash/tuscany/1.6.1-RC1/
>>
>> The release tag is at:
>>
>> https://svn.apache.org/repos/asf/tuscany/sca-java-1.x/tags/1.6.1-RC1/
>>
>> Here's my +1.
>>
>> Simon
>>
>>
>
> I've had a look and run some samples, the only issues I've found are:
>
> - itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
> build and the reactor build when building from the top.
>
This unfortunately crept in as a result of a late fix to the release
branch that wasn't correctly merged into the tag.
I scanned the whole codebase to check for other SNAPSHOT occurrences
and I found two more in tools/eclipse/features/core/pom.xml and
tools/eclipse/plugins/core/pom.xml. Given that this isn't a single
isolated occurrence, I think it's worth doing a respin to fix it.
> - tools\maven\maven-incremental-build fails with: Error extracting
> plugin descriptor: 'No mojo definitions were found for plugin:
> org.apache.tuscany.sca:tuscany-maven-incremental-build.'
>
After much puzzlement I think I have figured this out. The Mojo source code
for this module has been incorrectly stripped from the source distribution
because of an error in distribution/src/main/assembly/src.xml. This was
wrong in 1.6 as well. I thought I had compared the tag with the source
distribution but somehow I overlooked this difference.
> - using Tomcat 7 i see a warning about a memory leak:
> SEVERE: The web application [/sample-calculator-webapp] created a
> ThreadLocal with key of type [null] (value
> [org.apache.tuscany.sca.core.invocation.ThreadMessageContext$1@1b071c0])
> and a value of type
> [org.apache.tuscany.sca.core.invocation.MessageImpl] (value
> [org.apache.tuscany.sca.core.invocation.MessageImpl@fbf51d]) but
> failed to remove it when the web application was stopped. This is very
> likely to create a memory leak.
>
I understand what's going on here and I'm looking to see if there's a
simple fix.
Simon
> Those first two issues should really be fixed but if there's no other
> reason for re-spinning then +1 from me with the proviso that there
> would be a errata link on the download page mentioning the build
> problems and how to bypass them.
>
> ...ant
>
>
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Laws <si...@googlemail.com>.
>> In the src RAT there are a couple of files in
>> itest\wsdlgen\src\test\resources\wsdl that look like there need
>> headers. Even though they are originally generated they're copied into
>> svn. I don't believe they're used for comparison.
>>
> I presume you're referring to these two files:
> 1)
> tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService_schema1.xsd
> 2)
> tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService_schema2.xsd
>
> And possibly this file as well, though it does have a comment saying that
> it's generated:
> 3)
> tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService.wsdl
Yes, they're the ones.
>
> There's a README in the same directory explaining how these files were
> produced.
> I looked at this and I thought it best to leave the contents of this
> directory as is,
> as I feel a bit uncomfortable about adding Apache license headers to code
> that was
> generated by a tool that isn't Apache licensed. Also, the files in this
> directory
> have presumably been released in this form in previous 1.x releases.
Maybe I thought the same at the time as I note that I added them originally.
>
> Instead of adding full Apache headers, how about adding comments to these 3
> files
> pointing people to the README for further information? Proposed wording:
> <!-- This file contains code generated by JAX-WS: see the README file in
> this directory for further information -->
>
> Does this seem OK?
I think if we were proposing editing them then we should just add the
header. No on else has raised concerns and, as I said before , this is
not a blocker for me as they are originally generated files so if
there is no appetite lets leave as is.
>
> Simon
>
>> Excellent effort for an RC1! I be happy once the build issue is fixed.
>>
>> Simon
>>
>
>
--
Apache Tuscany committer: tuscany.apache.org
Co-author of a book about Tuscany and SCA: tuscanyinaction.com
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Nash <na...@apache.org>.
Simon Laws wrote:
> On Tue, Nov 9, 2010 at 10:15 AM, Simon Laws <si...@googlemail.com> wrote:
>>> - itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
>>> build and the reactor build when building from the top.
>> I just tried on my linux box and predictably get the same result. I'll
>> fix and let it run and see if anything else crops up.
>>
>> Simon
>>
>> --
>> Apache Tuscany committer: tuscany.apache.org
>> Co-author of a book about Tuscany and SCA: tuscanyinaction.com
>>
>
> OK, on my linux box I just see the two issues that Ant's already mentioned.
>
Thanks: I'll fix these for RC2.
> I tried a variety of samples and they all worked as expected.
> The license looks ok
> In the src RAT there are a couple of files in
> itest\wsdlgen\src\test\resources\wsdl that look like there need
> headers. Even though they are originally generated they're copied into
> svn. I don't believe they're used for comparison.
>
I presume you're referring to these two files:
1) tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService_schema1.xsd
2) tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService_schema2.xsd
And possibly this file as well, though it does have a comment saying that it's generated:
3) tuscany-sca-1.6.1-src/itest/wsdlgen/src/test/resources/wsdl/HelloWorldImplService.wsdl
There's a README in the same directory explaining how these files were produced.
I looked at this and I thought it best to leave the contents of this directory as is,
as I feel a bit uncomfortable about adding Apache license headers to code that was
generated by a tool that isn't Apache licensed. Also, the files in this directory
have presumably been released in this form in previous 1.x releases.
Instead of adding full Apache headers, how about adding comments to these 3 files
pointing people to the README for further information? Proposed wording:
<!-- This file contains code generated by JAX-WS: see the README file in this directory for further information -->
Does this seem OK?
Simon
> Excellent effort for an RC1! I be happy once the build issue is fixed.
>
> Simon
>
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Laws <si...@googlemail.com>.
On Tue, Nov 9, 2010 at 10:15 AM, Simon Laws <si...@googlemail.com> wrote:
>>
>> - itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
>> build and the reactor build when building from the top.
>
> I just tried on my linux box and predictably get the same result. I'll
> fix and let it run and see if anything else crops up.
>
> Simon
>
> --
> Apache Tuscany committer: tuscany.apache.org
> Co-author of a book about Tuscany and SCA: tuscanyinaction.com
>
OK, on my linux box I just see the two issues that Ant's already mentioned.
I tried a variety of samples and they all worked as expected.
The license looks ok
In the src RAT there are a couple of files in
itest\wsdlgen\src\test\resources\wsdl that look like there need
headers. Even though they are originally generated they're copied into
svn. I don't believe they're used for comparison.
Excellent effort for an RC1! I be happy once the build issue is fixed.
Simon
--
Apache Tuscany committer: tuscany.apache.org
Co-author of a book about Tuscany and SCA: tuscanyinaction.com
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by Simon Laws <si...@googlemail.com>.
>
> - itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
> build and the reactor build when building from the top.
I just tried on my linux box and predictably get the same result. I'll
fix and let it run and see if anything else crops up.
Simon
--
Apache Tuscany committer: tuscany.apache.org
Co-author of a book about Tuscany and SCA: tuscanyinaction.com
Re: [VOTE] Release Tuscany SCA 1.6.1 RC1
Posted by ant elder <an...@gmail.com>.
On Thu, Nov 4, 2010 at 2:37 PM, Simon Nash <na...@apache.org> wrote:
> Please review and vote on RC1 of the SCA Java 1.6.1 release.
>
> The distribution artifacts, RAT report, Eclipse update site, and
> Maven staging repository are available for review at:
>
> http://people.apache.org/~nash/tuscany/1.6.1-RC1/
>
> The release tag is at:
>
> https://svn.apache.org/repos/asf/tuscany/sca-java-1.x/tags/1.6.1-RC1/
>
> Here's my +1.
>
> Simon
>
>
I've had a look and run some samples, the only issues I've found are:
- itest/wsdlgen/pom.xml is using -SNAPSHOT dependencies so fails to
build and the reactor build when building from the top.
- tools\maven\maven-incremental-build fails with: Error extracting
plugin descriptor: 'No mojo definitions were found for plugin:
org.apache.tuscany.sca:tuscany-maven-incremental-build.'
- using Tomcat 7 i see a warning about a memory leak:
SEVERE: The web application [/sample-calculator-webapp] created a
ThreadLocal with key of type [null] (value
[org.apache.tuscany.sca.core.invocation.ThreadMessageContext$1@1b071c0])
and a value of type
[org.apache.tuscany.sca.core.invocation.MessageImpl] (value
[org.apache.tuscany.sca.core.invocation.MessageImpl@fbf51d]) but
failed to remove it when the web application was stopped. This is very
likely to create a memory leak.
Those first two issues should really be fixed but if there's no other
reason for re-spinning then +1 from me with the proviso that there
would be a errata link on the download page mentioning the build
problems and how to bypass them.
...ant