Thanks for the pointers Werner, I will have a look.
Marek
> On 18 Aug 2016, at 16:03, Werner Keil <werner.keil_at_gmail.com> wrote:
>
> Hi,
>
> I am not aware of Oracle internal policies or requirements, but from the EC Member who also asked Anil what should happen with existing (or future) JSRs after java.net <http://java.net/> is gone (https://jcp.org/aboutJava/communityprocess/ec-public/materials/2016-08-09/August-2016-Public-Minutes.html <https://jcp.org/aboutJava/communityprocess/ec-public/materials/2016-08-09/August-2016-Public-Minutes.html>) and from experience of helping several JSRs and other projects with the biggest public CI servers, I can confirm:
>
> Travis CI works pretty well. It did so for JSR 354 (here Moneta: https://travis-ci.org/JavaMoney/jsr354-ri <https://travis-ci.org/JavaMoney/jsr354-ri>, since it's running Java 8 it may be a good example for Java EE 8 JSRs) and based on our experience with it, the JSR 375 RI, Soteria also set up a Travis instance: https://travis-ci.org/javaee-security-spec/soteria <https://travis-ci.org/javaee-security-spec/soteria>
>
> JSR 363, just filed FAB last night uses CircleCI (the TCK: https://circleci.com/gh/unitsofmeasurement/unit-tck/ <https://circleci.com/gh/unitsofmeasurement/unit-tck/> among other builds, including fully transparent results of coverage reports or https://circleci.com/gh/unitsofmeasurement/unit-tck-usage <https://circleci.com/gh/unitsofmeasurement/unit-tck-usage> showing TCK test results for 3 implementations)
> I tested CircleCI against others and it worked ever since.
>
> Biggest caveat with Jenkins/Cloudbees is, that it puts a 2 week deadline into free or Open Source instances.
> I don't have to elaborate how that would have reacted to the overly inactive and slow Java EE JSRs (including this one)
> While manual reactivation is usually possible, unless you got a commercial company like Oracle running Cloudbees/Jenkins in a way that's appropriate to JSRs (especially once they're final and no work may be done for 2 years in many cases) I would refrain from the public Jenkins unless they change their policy on that.
>
> Regards,
>
> Werner Keil | JCP Executive Committee Member, JSR 363 Co Spec Lead | Eclipse UOMo Lead, Babel Language Champion | Apache Committer
> Twitter @wernerkeil | @UnitAPI | @JSR354 | @AgoravaProj | @DeviceMap | #DevOps | #EclipseUOMo
> Skype werner.keil | Google+ gplus.to/wernerkeil <http://gplus.to/wernerkeil>
>
>
>
> On Thu, Aug 18, 2016 at 11:15 AM, <users-request_at_jax-rs-spec.java.net <mailto:users-request_at_jax-rs-spec.java.net>> wrote:
> Table of contents:
>
> 1. [jax-rs-spec users] Re: Early draft 2.1 API for Central - Marek Potociar <marek.potociar_at_oracle.com <mailto:marek.potociar_at_oracle.com>>
> 2. [jax-rs-spec users] Re: Early draft 2.1 API for Central - Marek Potociar <marek.potociar_at_oracle.com <mailto:marek.potociar_at_oracle.com>>
>
>
>
> ---------- Forwarded message ----------
> From: Marek Potociar <marek.potociar_at_oracle.com <mailto:marek.potociar_at_oracle.com>>
> To: jsr370-experts_at_jax-rs-spec.java.net <mailto:jsr370-experts_at_jax-rs-spec.java.net>
> Cc: Santiago Pericasgeertsen <santiago.pericasgeertsen_at_oracle.com <mailto:santiago.pericasgeertsen_at_oracle.com>>
> Date: Thu, 18 Aug 2016 10:29:29 +0200
> Subject: [jax-rs-spec users] Re: Early draft 2.1 API for Central
> I will have a look. I am not a travis expert though so this option may take some time.
>
> Marek
>
>> On 16 Aug 2016, at 19:56, Markus KARG <markus_at_headcrashing.eu <mailto:markus_at_headcrashing.eu>> wrote:
>>
>> Why not simply using the Travis CI service powered by Github, as there is already a Github mirror of Jersey anyways?
>> -Markus
>> <>
>> From: Marek Potociar [mailto:marek.potociar_at_oracle.com <mailto:marek.potociar_at_oracle.com>]
>> Sent: Dienstag, 16. August 2016 13:38
>> To: jsr370-experts_at_jax-rs-spec.java.net <mailto:jsr370-experts_at_jax-rs-spec.java.net>
>> Cc: Santiago Pericasgeertsen
>> Subject: Re: [jax-rs-spec users] Re: Early draft 2.1 API for Central
>>
>> If you can get us a stable and free of charge public Jenkins compute where we could setup and run automated JAX-RS build jobs, that would be great.
>> Otherwise I will continue using our Oracle internal infrastructure for these builds and there is no way you could help there unless you switch sides :)
>>
>> Marek
>>
>>> On 16 Aug 2016, at 12:15, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>
>>> Hi Marek
>>>
>>> Sounds good, thanks, let me know please if you'd like me to help somehow
>>>
>>> Cheers, Sergey
>>> On 16/08/16 09:31, Marek Potociar wrote:
>>>> Hi Sergey,
>>>>
>>>> I understand your concern, I will try to spend some time on setting up automated release job sometime next week.
>>>>
>>>> Cheers,
>>>> Marek
>>>>
>>>>> On 15 Aug 2016, at 17:50, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>>>
>>>>> Hi Marek
>>>>>
>>>>> Yes, but it was a 2.1-SNAPSHOT in
>>>>> https://maven.java.net/content/repositories/snapshots/ <https://maven.java.net/content/repositories/snapshots/>
>>>>>
>>>>> (https://maven.java.net/#nexus-search;quick~javax.ws.rs <https://maven.java.net/#nexus-search;quick%7Ejavax.ws.rs>)
>>>>>
>>>>> CXF master-jaxrs21 branch depends on it, it helped, but we can not start merging master-jaxrs21 branch
>>>>> into CXF master until a non-SNAPSHOT API release is available. The concern is that if we do it
>>>>> then we can block the master release due to a SNAPSHOT dependency.
>>>>>
>>>>> We understand that if such a release were made then it would not represent the final API.
>>>>>
>>>>> Cheers, Sergey
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On 15/08/16 16:18, Marek Potociar wrote:
>>>>>> Hi Sergey,
>>>>>>
>>>>>> We did release some preliminary bits earlier based on your request, didn’t we? I would prefer waiting with more releases until we have some new content to justify the effort.
>>>>>>
>>>>>> Marek
>>>>>>
>>>>>>> On 15 Aug 2016, at 14:19, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>>>>>
>>>>>>> Hi Marek
>>>>>>>
>>>>>>> Seeing you replying in other threads makes me feel JAX-RS 2.1 has a very bright future :-)
>>>>>>>
>>>>>>> So what do you think about cutting out the current early draft release for it to make it to Central ?
>>>>>>> As I said, at this stage it would not be about meeting certain spec deadlines but rather supporting
>>>>>>> the implementation projects to do the early adoption and releases. Not only for CXF.
>>>>>>>
>>>>>>> By the way I should have a write access to a JAX-RS git repo. So if it can help I can do the release
>>>>>>> (I'd only need some guidance on how to reflect the early draft status in the artifact name, ex, 2.1-early-draft-m1, etc).
>>>>>>> Think about it please if it can work.
>>>>>>>
>>>>>>> Thanks, Sergey
>>>>>>>
>>>>>>> On 26/07/16 13:33, Sergey Beryozkin wrote:
>>>>>>>> Hi Marek, Santiago
>>>>>>>>
>>>>>>>> Hope you are doing well.
>>>>>>>>
>>>>>>>> We'd like in CXF to start merging our JAX-RS 2.1 branch to the main CXF 3.2.0-SNAPSHOT.
>>>>>>>>
>>>>>>>> My colleague had implemented SSE (server only so far) but it is a start and interesting, today, we've had a user
>>>>>>>> asking for this SSE code.
>>>>>>>>
>>>>>>>> For us to merge the branch code to the CXF master an early draft non-SNAPSHOT 2.1 API would need to be available
>>>>>>>> in Central. This is so that we can release CXF 3.2.0 eventually, knowing that we do not have SNAPSHOT deps.
>>>>>>>> It will also help us to go ahead and implement the NIO experiment and CompletableFuture client code directly on the master.
>>>>>>>>
>>>>>>>> Marek, can you please support it and do the very first API release ? (add some qualifier like early_draft, etc, to make it obvious that
>>>>>>>> it is far from being complete) ?
>>>>>>>>
>>>>>>>> a demand for JAX-RS is still very strong. If CXF manages to ship an early draft implementation, alongside with RI, it would be easier for
>>>>>>>> the community at large to continue working toward 2.1 being eventually completed. Depending on how things go I might consider submitting
>>>>>>>> a talk to JaveOne to raise the awareness of JAX-RS being implemented in ASF.
>>>>>>>>
>>>>>>>> Many thanks, Sergey
>>>>>>>>
>
>
>
> ---------- Forwarded message ----------
> From: Marek Potociar <marek.potociar_at_oracle.com <mailto:marek.potociar_at_oracle.com>>
> To: jsr370-experts_at_jax-rs-spec.java.net <mailto:jsr370-experts_at_jax-rs-spec.java.net>
> Cc: Santiago Pericasgeertsen <santiago.pericasgeertsen_at_oracle.com <mailto:santiago.pericasgeertsen_at_oracle.com>>
> Date: Thu, 18 Aug 2016 10:30:27 +0200
> Subject: [jax-rs-spec users] Re: Early draft 2.1 API for Central
> We have been using OSS CloudBees account for Jersey already but I am not too impressed.
>
> Marek
>
>> On 16 Aug 2016, at 15:48, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>
>> Hi Marek
>>
>> Not sure if it helps (or new to you), but the recommendation that I also hear from others is to consider
>> engaging with Cloudbees, perhaps starting with a Free Trial at
>>
>> https://www.cloudbees.com/get-started <https://www.cloudbees.com/get-started>
>>
>> and continue from there - who knows what may happen at the end of the free trial :-).
>>
>> Sorry can't suggest anything more specific at this stage, so
>> you might want continue with the existing infra for a time being :-).
>>
>> Cheers, Sergey
>>
>>
>> On 16/08/16 13:28, Sergey Beryozkin wrote:
>>> Hi Marek
>>>
>>> let me ask around :-)
>>>
>>> Cheers, Sergey
>>> On 16/08/16 12:37, Marek Potociar wrote:
>>>> If you can get us a stable and free of charge public Jenkins compute where we could setup and run automated JAX-RS build jobs, that would be great.
>>>> Otherwise I will continue using our Oracle internal infrastructure for these builds and there is no way you could help there unless you switch sides :)
>>>>
>>>> Marek
>>>>
>>>>> On 16 Aug 2016, at 12:15, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>>>
>>>>> Hi Marek
>>>>>
>>>>> Sounds good, thanks, let me know please if you'd like me to help somehow
>>>>>
>>>>> Cheers, Sergey
>>>>> On 16/08/16 09:31, Marek Potociar wrote:
>>>>>> Hi Sergey,
>>>>>>
>>>>>> I understand your concern, I will try to spend some time on setting up automated release job sometime next week.
>>>>>>
>>>>>> Cheers,
>>>>>> Marek
>>>>>>
>>>>>>> On 15 Aug 2016, at 17:50, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>>>>>
>>>>>>> Hi Marek
>>>>>>>
>>>>>>> Yes, but it was a 2.1-SNAPSHOT in
>>>>>>> https://maven.java.net/content/repositories/snapshots/ <https://maven.java.net/content/repositories/snapshots/>
>>>>>>>
>>>>>>> (https://maven.java.net/#nexus-search;quick~javax.ws.rs <https://maven.java.net/#nexus-search;quick%7Ejavax.ws.rs>)
>>>>>>>
>>>>>>> CXF master-jaxrs21 branch depends on it, it helped, but we can not start merging master-jaxrs21 branch
>>>>>>> into CXF master until a non-SNAPSHOT API release is available. The concern is that if we do it
>>>>>>> then we can block the master release due to a SNAPSHOT dependency.
>>>>>>>
>>>>>>> We understand that if such a release were made then it would not represent the final API.
>>>>>>>
>>>>>>> Cheers, Sergey
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> On 15/08/16 16:18, Marek Potociar wrote:
>>>>>>>> Hi Sergey,
>>>>>>>>
>>>>>>>> We did release some preliminary bits earlier based on your request, didn’t we? I would prefer waiting with more releases until we have some new content to justify the effort.
>>>>>>>>
>>>>>>>> Marek
>>>>>>>>
>>>>>>>>> On 15 Aug 2016, at 14:19, Sergey Beryozkin <sberyozkin_at_talend.com <mailto:sberyozkin_at_talend.com>> wrote:
>>>>>>>>>
>>>>>>>>> Hi Marek
>>>>>>>>>
>>>>>>>>> Seeing you replying in other threads makes me feel JAX-RS 2.1 has a very bright future :-)
>>>>>>>>>
>>>>>>>>> So what do you think about cutting out the current early draft release for it to make it to Central ?
>>>>>>>>> As I said, at this stage it would not be about meeting certain spec deadlines but rather supporting
>>>>>>>>> the implementation projects to do the early adoption and releases. Not only for CXF.
>>>>>>>>>
>>>>>>>>> By the way I should have a write access to a JAX-RS git repo. So if it can help I can do the release
>>>>>>>>> (I'd only need some guidance on how to reflect the early draft status in the artifact name, ex, 2.1-early-draft-m1, etc).
>>>>>>>>> Think about it please if it can work.
>>>>>>>>>
>>>>>>>>> Thanks, Sergey
>>>>>>>>>
>>>>>>>>> On 26/07/16 13:33, Sergey Beryozkin wrote:
>>>>>>>>>> Hi Marek, Santiago
>>>>>>>>>>
>>>>>>>>>> Hope you are doing well.
>>>>>>>>>>
>>>>>>>>>> We'd like in CXF to start merging our JAX-RS 2.1 branch to the main CXF 3.2.0-SNAPSHOT.
>>>>>>>>>>
>>>>>>>>>> My colleague had implemented SSE (server only so far) but it is a start and interesting, today, we've had a user
>>>>>>>>>> asking for this SSE code.
>>>>>>>>>>
>>>>>>>>>> For us to merge the branch code to the CXF master an early draft non-SNAPSHOT 2.1 API would need to be available
>>>>>>>>>> in Central. This is so that we can release CXF 3.2.0 eventually, knowing that we do not have SNAPSHOT deps.
>>>>>>>>>> It will also help us to go ahead and implement the NIO experiment and CompletableFuture client code directly on the master.
>>>>>>>>>>
>>>>>>>>>> Marek, can you please support it and do the very first API release ? (add some qualifier like early_draft, etc, to make it obvious that
>>>>>>>>>> it is far from being complete) ?
>>>>>>>>>>
>>>>>>>>>> a demand for JAX-RS is still very strong. If CXF manages to ship an early draft implementation, alongside with RI, it would be easier for
>>>>>>>>>> the community at large to continue working toward 2.1 being eventually completed. Depending on how things go I might consider submitting
>>>>>>>>>> a talk to JaveOne to raise the awareness of JAX-RS being implemented in ASF.
>>>>>>>>>>
>>>>>>>>>> Many thanks, Sergey
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>
>>>>>>>>
>>>>>>>
>>>>>>
>>>>>
>>>>
>>>
>>
>
>
> End of digest for list users_at_jax-rs-spec.java.net <mailto:users_at_jax-rs-spec.java.net> - Thu, 18 Aug 2016
>
>