44 lines
1.7 KiB
Text
44 lines
1.7 KiB
Text
|
= Highlights
|
||
|
|
|
||
|
|
== Identity Brokering Sync Mode
|
||
|
|
|
||
|
|
With Identity Brokering Sync Mode it is now possible to control if user profiles are updated on first login, or
|
||
|
|
every login from an external Identity Provider. It is also possible to override this behaviour on individual mappers.
|
||
|
|
|
||
|
|
Thanks to https://github.com/Martin-Idel-SI[Martin Idel]
|
||
|
|
|
||
|
|
|
||
|
|
== Client Session Timeout for OpenID Connect / OAuth 2.0
|
||
|
|
|
||
|
|
Typically, an SSO session last for days if not months, while individual client sessions should ideally be a lot shorter.
|
||
|
|
With the introduction of client session timeout it is now possible to configure a separate timeout for individual clients,
|
||
|
|
as well as a default for all clients within a realm.
|
||
|
|
|
||
|
|
Thanks to https://github.com/y-tabata[Yoshiyuki Tabata]
|
||
|
|
|
||
|
|
|
||
|
|
== OAuth 2.0 Token Revocation (RFC 7009)
|
||
|
|
|
||
|
|
For applications that use Keycloak as an OAuth 2.0 Authorization Server there is now support to revoke refresh tokens
|
||
|
|
through the token revocation endpoint.
|
||
|
|
|
||
|
|
Thanks to https://github.com/y-tabata[Yoshiyuki Tabata]
|
||
|
|
|
||
|
|
|
||
|
|
== Security Headers SPI and Response Filter
|
||
|
|
|
||
|
|
A new SPI was introduced to allow better flexibility when setting security related headers on responses. This provides
|
||
|
|
a cleaner implementation within Keycloak, but also allows full customisation if needed. Security headers are now set
|
||
|
|
by a response filter instead of within the code itself, which makes it less error prone, removing the chance that
|
||
|
|
some response are missing headers.
|
||
|
|
|
||
|
|
|
||
|
|
== Upgrade to WildFly 19
|
||
|
|
|
||
|
|
Keycloak server was upgraded to use WildFly 19 under the covers.
|
||
|
|
|
||
|
|
|
||
|
|
== Other improvements
|
||
|
|
|
||
|
|
* Support for invoking Application Initiated Actions added to Keycloak JavaScript adapter
|
||
|
|
* Performance improvements to fetching resources and policies during evaluation
|