river-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Greg Trasuk <tras...@stratuscom.com>
Subject Re: River 2.2.2
Date Mon, 11 Nov 2013 05:05:19 GMT

As a matter of fact, right now, it’s setup to pull down tags/2.2.2.  There’s a build scheduled,
but it has to wait until the other River builds are done, so probably will run sometime tomorrow.

Cheers,

Greg.

On Nov 10, 2013, at 6:00 PM, Jonathan Costers <jonathan.costers@gmail.com> wrote:

> Hi Greg
> That was exactly where I was going. 
> I hadn't checked in Hudson, which I should have done.
> Thanks
> Jonathan
> 
> Op 10-nov.-2013, om 23:38 heeft Greg Trasuk <trasukg@stratuscom.com> het volgende
geschreven:
> 
>> 
>> We already have this, which has been there since the last release.  
>> 
>> https://builds.apache.org/job/river-2.2-qa-jdk7/
>> 
>> It pulls from the 2.2 branch, which the 2.2.2 tag will be copied from.  Currently
the build passes.
>> 
>> My understanding has always been that release artifacts need to be generated and
signed on a machine that is under local control of the code signer (i.e. release manager,
i.e. me for the 2.2.2 release).  As such, I don’t think we should use the artifacts generated
under Hudson, although the jars are certainly there if anyone wants to grab a snapshot.
>> 
>> Cheers,
>> 
>> Greg Trasuk.
>> 
>> https://builds.apache.org/job/river-2.2-qa-jdk7/
>> 
>> 
>> On Nov 10, 2013, at 3:16 PM, Jonathan Costers <jonathan.costers@gmail.com>
wrote:
>> 
>>> Should we setup a Hudson build that pulls the 2.2.2 tag from SVN, tests it, generates
reports and creates release artifacts?
>>> 
>>> Op 5-okt.-2013, om 05:30 heeft Peter <jini@zeus.net.au> het volgende geschreven:
>>> 
>>>> Sim also did some work based on Gregg's contribution, it's included in trunk,
it was during this time that unrelated synchronization issues caused progress on this work
to stall.  The design is quite elegant and flexible.
>>>> 
>>>> Due to synchronization bugs causing test failures, I branched from an earlier
trunk version that appeared stable.  I don't have access to hardware suitable for generating
the failure conditions, so have been unable to continue working on these test failures.
>>>> 
>>>> Despite being quite impressed by its elegance, there are some fundamental
design flaws with Reggie's implementation regarding mutation, these only come to light are
spending hours working through the code. 
>>>> 
>>>> I have considered rewriting Reggie, after an unsuccessful refactoring attempt.
>>>> 
>>>> Regards,
>>>> 
>>>> Peter.
>>>> 
>>>> ----- Original message -----
>>>>> https://issues.apache.org/jira/browse/RIVER-336?page=com.atlassian.jira
>>>>> 
>>>>> On Oct 3, 2013, at 1039AM, Greg Trasuk wrote:
>>>>> 
>>>>>> 
>>>>>> I'm having trouble finding a reference to that.   Do you happen to
have
>>>>>> a link to email archives or a Jira issue?
>>>>>> 
>>>>>> Thanks,
>>>>>> 
>>>>>> Greg.
>>>>>> 
>>>>>> On 2013-10-02, at 8:15 PM, Dennis Reedy <dennis.reedy@gmail.com>
wrote:
>>>>>> 
>>>>>>> Hey Greg,
>>>>>>> 
>>>>>>> The work that Gregg Wonderly championed with the RMIClassLoaderSpi
>>>>>>> would be one for me.
>>>>>>> 
>>>>>>> Regards
>>>>>>> 
>>>>>>> Dennis
>>>>>>> 
>>>>>>> On Oct 2, 2013, at 546PM, Greg Trasuk wrote:
>>>>>>> 
>>>>>>>> Hi all:
>>>>>>>> 
>>>>>>>> I'm planning to propose a release for River 2.2.2 later this
week,
>>>>>>>> based on the current state of the 2.2. branch.   The only
change
>>>>>>>> from 2.2.1 is the addition of the JMX Entry classes, plus
addition
>>>>>>>> of one more jar file to be added to Maven Central (jsk-policy.jar
>>>>>>>> if I remember correctly).
>>>>>>>> 
>>>>>>>> Any objections or anything else that anyone wants to include
in
>>>>>>>> the release?
>>>>>>>> 
>>>>>>>> 
>>>>>>>> Cheers,
>>>>>>>> 
>>>>>>>> Greg Trasuk.
>>>>>>>> 
>>>>>>> 
>>>>>> 
>>>>> 
>>>> 
>>> 
>> 
> 


Mime
View raw message