----- Original Message -----
From: "Boleslaw Dawidowicz" <bdawidow(a)redhat.com>
To: "Bill Burke" <bburke(a)redhat.com>
Cc: "Stian Thorgersen" <stian(a)redhat.com>, keycloak-dev(a)lists.jboss.org
Sent: Thursday, April 30, 2015 9:16:35 AM
Subject: Re: [keycloak-dev] Supported environments
> On 28 Apr 2015, at 17:14, Bill Burke <bburke(a)redhat.com> wrote:
>
>
>
> On 4/28/2015 10:02 AM, Stian Thorgersen wrote:
>>
>>
>> ----- Original Message -----
>>> From: "Bill Burke" <bburke(a)redhat.com>
>>> To: keycloak-dev(a)lists.jboss.org
>>> Sent: Tuesday, April 28, 2015 4:00:46 PM
>>> Subject: Re: [keycloak-dev] Supported environments
>>>
>>>
>>>
>>> On 4/28/2015 3:33 AM, Boleslaw Dawidowicz wrote:
>>>>
>>>>
>>>> W dniu 2015-04-27 o 08:37, Stian Thorgersen pisze:
>>>>> For 1.2.0.Final the server should be deployable to:
>>>>>
>>>>> * WildFly 8.2.0.Final
>>>>> * EAP 6.4.0.GA
>>>>>
>>>>> After 1.2.0.Final we should move on to WildFly 9.0.0.Final (once
it's
>>>>> released). As APIs has changed, and there's also new features in
WF 9
>>>>> we
>>>>> can leverage, the cleanest way to do this is to drop support for
>>>>> deploying to EAP 6.4.0.GA.
>>>>>
>>>>> With regards to adapters what versions should we support:
>>>>>
>>>>> * AS7 - can we drop this?
>>>>> * WildFly - only last (9.0.0.Final) or two last (8.2.0.Final and
>>>>> 9.0.0.Final)?
>>>>> * EAP - only last (6.4), two last (6.3, 6.4) or three last (6.2,
6.3,
>>>>> 6.4)?
>>>>
>>>> Would go with only latest stable WF and only last minor of EAP 6.
>>>>
>>>
>>> We should never drop adapter support for any platform. The larger our
>>> adapter pool, the larger our user base will be.
>>
>> So we're maintaining AS7 support for ever?!?
>>
>
> I don't think it will be much of an issue. These old platforms are
> pretty static and the adapters are generic enough to evolve.
>
IMO it is all about maintenance cost of covering many versions of same
container. If it is low enough then it shouldn’t be a problem - especially
for adapters.
We need to ability to test adapters with full containers, not just embedded containers.
Once we have that for an adapter the maintenance cost goes down significantly. Currently
even though we have tests for the adapters, we still don't know if they work until
we've manually tested it. For example if there's a problems in modules or the
adapter subsystem that can only be properly tested on a full WildFly server.
Same goes for the server and examples - we need to run tests against a full server build,
not just an embedded Undertow.
>
> --
> Bill Burke
> JBoss, a division of Red Hat
>
http://bill.burkecentral.com
> _______________________________________________
> keycloak-dev mailing list
> keycloak-dev(a)lists.jboss.org
>
https://lists.jboss.org/mailman/listinfo/keycloak-dev