A lot of the PL APIs are going to go away. We want tight and focused
SPIs that leverage the rest of the Keycloak platform and vision. Not
broad generic APIs for rolling your own SSO solution. We want something
out of the box that for most cases a user doesn't have to be a Java
programmer to use and deploy Keycloak. You just have to have a focused
and specific security model to provide an out of the box solution and
focused vision.
That being said, what we've learned is that developers have broad and
different requirements when it comes to authorization and permissions.
This is where an API, backed by the authz server, would be very useful.
Pedro was involved with PLs permission APIs and I'm confident he has
some really good ideas for evolving it.
On 10/19/2015 8:45 AM, Arthur Gregório wrote:
Then KC will not have a model in the style of the PL?
This means that those who used the PL hoping that everything in it was
in KC turned out to have a framework discontinued and with no more updates?
Or the PL as it exists today will continue to be developed and there
will be a plus integration with KC?
I am very confused by the merge of the projects, it appears that the
security of my system already died before were born.
at.,
*Arthur P. Gregório*
/+55 45 9958-0302/
@gregorioarthur
www.arthurgregorio.eti.br <
http://www.arthurgregorio.eti.br>
2015-10-19 10:24 GMT-02:00 Pedro Igor Silva <psilva(a)redhat.com
<mailto:psilva@redhat.com>>:
Hey Crhisthian,
As Bill said, we are working on an Authz Server for Keycloak in
order to provide fine-grained permissions. It is still a working in
progress, although we already have a baseline for a first release
which will happen very soon.
From a migration perspective, while PL provides a rich
Permission Java API, Keycloak will provide a distributable
authorization server based on a RESTful API to manage resources,
policies, evaluate policies, obtain entitlements and plus other
goodies. In other words, Keycloak will become a PAP (Policy
Administration Point), a PDP (Policy Decision Point) and a
Entitlements Server. Everything based on OpenID Connect (and of
course, oAuth2).
As you know, Keycloak is a feature rich, OOTB and easy to use
security as a service solution. We are considering these same
premises for the authz server, so you can protect web apps, RESTful
APIs or any other resources very easily. For instance, you'll be
able to write policies using JBoss Drools, EL and easily extend your
existing oAuth2 clients in order ask for permissions or enforce them
(in case your client acts as a resource server).
I'm afraid there will be no "migragration path" between PL and
KC, at this sense. But we can work together to make this migration
easier. For instance, we are going to provide a Protection API which
can be used to manage resources and policies remotely.
Regards.
Pedro Igor
----- Original Message -----
From: "Cristhian Camilo Lopez" <calovi86(a)gmail.com
<mailto:calovi86@gmail.com>>
To: keycloak-dev(a)lists.jboss.org <mailto:keycloak-dev@lists.jboss.org>
Sent: Sunday, October 18, 2015 3:16:30 PM
Subject: Re: [keycloak-dev] Authz Model Implementation
Hi Pedro,
I'm migrating from Picketlink, but I haven't found the way to use
fine-grained permissions, Could u give me some advice on this ?
Thanks,
Cristhian.
_______________________________________________
keycloak-dev mailing list
keycloak-dev(a)lists.jboss.org <mailto:keycloak-dev@lists.jboss.org>
https://lists.jboss.org/mailman/listinfo/keycloak-dev
_______________________________________________
keycloak-dev mailing list
keycloak-dev(a)lists.jboss.org <mailto:keycloak-dev@lists.jboss.org>
https://lists.jboss.org/mailman/listinfo/keycloak-dev
_______________________________________________
keycloak-dev mailing list
keycloak-dev(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/keycloak-dev