keycloak-scim/topics/overview/terminology.adoc
2016-05-31 17:36:14 -03:00

106 lines
No EOL
5.5 KiB
Text
Executable file

== Terminology
Before going further, it is important to understand some terms and concepts introduced with {{book.project.name}} {{book.project.module}}.
==== Resource Server
If you are familiar with OAuth2, a Resource Server is the server hosting the protected resources, capable of accepting
and responding to protected resource requests.
Resource servers usually rely on some kind of information to decide whether access to a protected resource should be granted or not. For RESTful-based resource servers,
that information is usually obtained from a security token, usually sent as a bearer token on every single request to the server. For web applications that rely on a session to
authenticate their users, that information is usually stored into user's session and retrieved from there on every single request.
In Keycloak, any *confidential* client application may act as a resource server. Whose resources and their respective scopes are
protected and ruled by a set of authorization policies.
==== Resource Server
A resource is part of the assets of an application and the organization. It may be a single API
endpoint, a set of API endpoints, a classic web resource such as an HTML page, and so on.
In authorization policy terminology, a resource is the _object_ being protected.
Every single resource has a unique identifier, where a resource may also be used to represent a single
or a set of resources. For instance, you may want to manage a _Banking Account Resource_ that represents and defines a set of authorization policies for all banking accounts.
But you may also have a _Alice Banking Account_, which represents a single resource owned by a single customer, which may have its own set of authorization policies.
In {{book.project.name}}, a resource defines a small set of information that is common to different types of resources, such as:
* *Name*
+
A human-readable and unique string describing a set of one or more resources.
+
* *Type*
+
A string uniquely identifying the type of a set of one or more resources. Usually, the type is a URN that can be used to
group different resource instances.
+
* *URI*
+
A URI that provides the location/address for the resource. For HTTP resources, the URI
is usually the relative path used to serve these resources.
+
* *Scopes*
+
A bounded extent of access that is possible to perform on a resource. In authorization
policy terminology, a scope is one of the potentially many <literal>verbs</literal> that can logically
apply to a resource.
+
* *Owner*
+
An entity that owns the resource. It can be the resource server itself or even a
specific user.
==== Scope
A bounded extent of access that is possible to perform on a resource. In authorization policy
terminology, a scope is one of the potentially many _verbs_ that can logically apply to a resource.
Usually, a scope is defined as a URN that indicates what can be done with a given resource. Example of scopes are _urn:domain:resource:scope:view_,
_urn:domain:scopes:admin:manage_, etc.
Scopes have a small set of information as follows:
* *Name*
+
A human-readable and unique string describing the scope.
A single scope may be associated with zero or more resources.
==== Permission
A permission associates the object being protected and the policies that must be evaluated in order to decide whether access should be granted or not.
In {{book.project.name}}, a permission can be defined for a:
* *Resource*
+
In this case, the permission is associated with a set of one or more resources. Here you can define that only a specific
resource with a specific _name_ or _identifier_ is protected or even use a _type_ to protect any resource with a given type.
* *Scope*
+
In this case, the permission is associated with a set of one or more scopes. Where you may want to protect scopes associated with a
specific resource or any scope regardless the resources they are associated.
When associating policies to permissions, you can also define the _decision strategy_ that will be used during the evaluation of these
policies in order to decide whether a permission is granted or not depending on the outcome of each associated policy.
==== Policy
A policy defines the conditions that must be satisfied to grant access to an object. Different than permissions, you don't really specify the object being protected
but the conditions that must be satisfied to get access to a given object (eg.: resource, scope or both).
Policies are strongly related with the different _access control mechanism_ that you can use to actually protect your resources.
Within a policy you can use ABAC, RBAC, Context-based Access Control or any combination of these.
Keycloak leverages the concept of policies and how you define them by providing the concept of *Aggregated Policies*, where you can build a "policy of policies" and still control the behavior of the evaluation.
Instead of writing a single and huge policy with all conditions that must be satisfied to get access to a given resource, policies in Keycloak follows the *divide-and-conquer* technique,
so you can create individual policies, reuse them on different permissions and build more complex policies by combining them into a single one.
==== Policy Provider
Policy providers are responsible to support a specific policy type. Although {{book.project.name}} provides some built-in policies, backed by their corresponding
policy providers, nothing stops you to create your own policy types in order to better support your requirements or a specific use case.
{{book.project.name}} provides a *SPI* (Service Provider Interface) that you can use to plug your own policy providers.