[keycloak-user] Brut force attack questions

[Stian Thorgersen]

----- Original Message -----
> From: "Alexander Chriztopher" <alexander.chriztopher at gmail.com>
> To: "Stian Thorgersen" <stian at redhat.com>
> Cc: "Bill Burke" <bburke at redhat.com>, keycloak-user at lists.jboss.org
> Sent: Wednesday, 26 November, 2014 11:46:21 AM
> Subject: Re: [keycloak-user] Brut force attack questions
> 
> i think we will go for Infinispan or jpa for roadmap reasons !
> 
> how would we access the infinispan data ?

You'll need to use 1.1.0.Beta1 as it's not available in 1.0.x. Have a look at http://docs.jboss.org/keycloak/docs/1.1.0.Beta1/userguide/html/clustering.html#d4e1662

You can inject the cache into your own application with:

  @Resource(lookup="java:jboss/infinispan/Keycloak")
  private CacheContainer container;   

You'll want to remove/change entries in the loginFailures cache. Look at the source for model/sessions-infinispan for more info on how the Infinispan userSession provider works.

Have a look at www.mastertheboss.com/jboss-frameworks/infinispan/develop-a-clustered-application-with-infinispan-data-grid for more info on using the Infinispan subsystem from your app.

> 
> are there any documents on this integration aspect within keycloak ?

What integration aspect are you referring to? The work around proposed with accessing JPA or Infinispan directly is not recommended so won't be documented.

> 
> On Wed, Nov 26, 2014 at 11:02 AM, Stian Thorgersen <stian at redhat.com> wrote:
> 
> >
> >
> > ----- Original Message -----
> > > From: "Alexander Chriztopher" <alexander.chriztopher at gmail.com>
> > > To: "Stian Thorgersen" <stian at redhat.com>
> > > Cc: "Bill Burke" <bburke at redhat.com>, keycloak-user at lists.jboss.org
> > > Sent: Wednesday, 26 November, 2014 10:51:11 AM
> > > Subject: Re: [keycloak-user] Brut force attack questions
> > >
> > > nice !
> > >
> > > is there a way to play with this via rest -regardless to wether am jpa,
> > > infinispan or in-memory ?
> >
> > Afraid not, that's what we'll need to add. Not sure when we get time
> > though. If you're up for it we would be more than happy to accept a
> > contribution, I can give you some pointers on how to do it. Alternatively,
> > you can create your own app that uses JAX-RS and either JPA or the
> > Infinispan subsystem to remove login-failure entries.
> >
> > >
> > > On Wed, Nov 26, 2014 at 10:39 AM, Stian Thorgersen <stian at redhat.com>
> > wrote:
> > >
> > > > By default user sessions (and login failures) are stored in-memory not
> > in
> > > > the database. Unless you specify JPA for the userSession provider those
> > > > tables will stay empty.
> > > >
> > > > You could either do what you're trying to do, which should work if you
> > use
> > > > the jpa userSession provider. The other if you're worried about the
> > > > performance of storing user sessions in the db is to use the Infinispan
> > > > provider, then you can manually delete login failures from the
> > userSession
> > > > cache from another application.
> > > >
> > > > We should add a mechanism to both view and remove login-failure
> > entries to
> > > > the admin console though.
> > > >
> > > > ----- Original Message -----
> > > > > From: "Alexander Chriztopher" <alexander.chriztopher at gmail.com>
> > > > > To: "Bill Burke" <bburke at redhat.com>, keycloak-user at lists.jboss.org
> > > > > Sent: Wednesday, 26 November, 2014 9:45:42 AM
> > > > > Subject: Re: [keycloak-user] Brut force attack questions
> > > > >
> > > > > Am to find a workaround in order to be able to unlock a user
> > account. So
> > > > far
> > > > > i have tried to disable then enable the user account but this does
> > not do
> > > > > the trick apparently.
> > > > >
> > > > > I have also tried to tweek the database but it looks like the lock
> > > > > information is not stored in the db even though there is the table :
> > > > > USERNAME_LOGIN_FAILURE. Is it normal that this table stays empty
> > even on
> > > > > login failures ?
> > > > >
> > > > > Do you think of any other good workaround ?
> > > > >
> > > > >
> > > > > On Tue, Nov 25, 2014 at 11:03 PM, Alexander Chriztopher <
> > > > > alexander.chriztopher at gmail.com > wrote:
> > > > >
> > > > >
> > > > > Nice ! Again, thank you.
> > > > >
> > > > >
> > > > >
> > > > > > On 25 Nov 2014, at 21:39, Bill Burke < bburke at redhat.com > wrote:
> > > > > >
> > > > > >
> > > > > >
> > > > > >> On 11/25/2014 3:27 PM, Alexander Chriztopher wrote:
> > > > > >> Hi Bill and thanks.
> > > > > >>
> > > > > >> Do you think we will be able to have this within a short period of
> > > > time
> > > > > >> (4-6 weeks) or is it going to be planned for the long run ?
> > > > > >
> > > > > > Not sure on the priority of this. We have face to face meetings in
> > a
> > > > couple
> > > > > > of weeks to discuss priority, then of course, its christmas
> > vacation.
> > > > > >
> > > > > >> When is the value of max wait used as there is already a wait
> > > > increment
> > > > > >> out there ?
> > > > > >
> > > > > > Correct. It will increase the wait after each failure until the
> > max is
> > > > hit.
> > > > > >
> > > > > >>
> > > > > >>
> > > > > >>> On 25 Nov 2014, at 20:05, Bill Burke < bburke at redhat.com >
> > wrote:
> > > > > >>>
> > > > > >>>
> > > > > >>>
> > > > > >>>> On 11/25/2014 12:32 PM, Alexander Chriztopher wrote:
> > > > > >>>> Hi,
> > > > > >>>>
> > > > > >>>> I have a some question with regards to Brut Force Attack
> > Protection
> > > > :
> > > > > >>>>
> > > > > >>>> # 1 / When brut force attack protection is enabled is there a
> > way to
> > > > > >>>> know when a user account is locked ? I am thinking about the
> > admin
> > > > > >>>> console.
> > > > > >>>>
> > > > > >>>> # 2 / When a user account is locked is there a way to unlock it
> > > > from the
> > > > > >>>> admin console ?
> > > > > >>>
> > > > > >>> Unfortunately no for the above. I'll log a jira.
> > > > > >>>
> > > > > >>>> # 3 / What is the difference between wait increment (When
> > failure
> > > > > >>>> threshold has been met, how much time should the user be locked
> > > > out?)
> > > > > >>>> and max wait (Max time a user will be locked out.).
> > > > > >>>
> > > > > >>> correct on both.
> > > > > >>>
> > > > > >>>
> > > > > >>> --
> > > > > >>> Bill Burke
> > > > > >>> JBoss, a division of Red Hat
> > > > > >>> http://bill.burkecentral.com
> > > > > >>> _______________________________________________
> > > > > >>> keycloak-user mailing list
> > > > > >>> keycloak-user at lists.jboss.org
> > > > > >>> https://lists.jboss.org/mailman/listinfo/keycloak-user
> > > > > >
> > > > > > --
> > > > > > Bill Burke
> > > > > > JBoss, a division of Red Hat
> > > > > > http://bill.burkecentral.com
> > > > >
> > > > >
> > > > > _______________________________________________
> > > > > keycloak-user mailing list
> > > > > keycloak-user at lists.jboss.org
> > > > > https://lists.jboss.org/mailman/listinfo/keycloak-user
> > > >
> > >
> >
>