1:48 p.m.
It's required, so don't remove.
If we don't have a cancel button there's no way for users to go back to the
application if they don't want to login (or can't for some reason). Also, there
are other situations where a login can fail, in which an error query param is returned to
application instead of a code. For example oauth client grant page (a user can accept or
reject giving the client the required permissions), etc.. The adapters needs to be able to
handle these properly. IMO if login is cancelled there's two basic use-cases:
* User clicked on log in link - in this case application should just return to the initial
page
* User clicked on a page that requires login - in this case the application should
probably show a 'unauthorized access' page which needs to be customizable by the
application
----- Original Message -----
From: "Stan Silvert" <ssilvert(a)redhat.com>
To: keycloak-user(a)lists.jboss.org
Sent: Friday, 10 October, 2014 1:40:12 PM
Subject: Re: [keycloak-user] What is the point of the cancel button on the log-in
screen?
Does the cancel button EVER work properly?
I'm starting to side with Alarik. In any situation where we know the
cancel button won't work, we need to either fix it or remove it.
On 10/10/2014 3:09 AM, Stian Thorgersen wrote:
> The back button still submits the form, but the instead of processing the
> login redirects with error set. So it's already not an open redirect.
>
> We should fix the adapter to show a error page though. Another thing is
> that the adapter needs some way of customising error pages.
>
> ----- Original Message -----
>> From: "Bill Burke" <bburke(a)redhat.com>
>> To: keycloak-user(a)lists.jboss.org
>> Sent: Thursday, 9 October, 2014 7:02:18 PM
>> Subject: Re: [keycloak-user] What is the point of the cancel button on the
>> log-in screen?
>>
>> We would have to rememer referrer information somehow via the adapter to
>> know where to redirect to. This cancel redirection URL would be an
>> extension to OIDC I think and would require to be validated so that we
>> don't create an open redirector security vulnerabilities. Maybe we
>> should we just show a Keycloak rendered error page?
>>
>>
>> On 10/9/2014 12:46 PM, Stan Silvert wrote:
>>> I guess I'm stating the obvious, but the cancel button should take you
>>> back to where you were before being challenged by the login screen. To
>>> the extent that is possible, the cancel button should stay. We should
>>> never rely on the back button.
>>>
>>> I just tried it on our demo and recreated the 400 error. We should fix
>>> this if possible.
>>>
>>> On 10/9/2014 12:18 PM, Alarik Myrin wrote:
>>>> At least with the Wildfly adapter, clicking cancel gets you a HTTP 400
>>>> -- Bad Request on your protected resource, and doing something more
>>>> graceful would take some thinking.
>>>>
>>>> It's not clear to me what *should* happen when clicking cancel.
Users
>>>> in a browser have a back button, or a button to close the tab, and
>>>> they can always use that to get out of the login screen.
>>>>
>>>> Maybe the cancel button should just be removed?
>>>>
>>>> Alarik
>>>>
>>>>
>>>>
>>>> _______________________________________________
>>>> keycloak-user mailing list
>>>> keycloak-user(a)lists.jboss.org
>>>> https://lists.jboss.org/mailman/listinfo/keycloak-user
>>>
>>>
>>> _______________________________________________
>>> keycloak-user mailing list
>>> keycloak-user(a)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(a)lists.jboss.org
>> https://lists.jboss.org/mailman/listinfo/keycloak-user
>>
> _______________________________________________
> keycloak-user mailing list
> keycloak-user(a)lists.jboss.org
> https://lists.jboss.org/mailman/listinfo/keycloak-user
_______________________________________________
keycloak-user mailing list
keycloak-user(a)lists.jboss.org
https://lists.jboss.org/mailman/listinfo/keycloak-user