<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
<div class="moz-cite-prefix">On 21/09/15 12:33, Stian Thorgersen
wrote:<br>
</div>
<blockquote
cite="mid:CAJgngAd+J8Q6C=Wi3Kz9JUvRnzfjeqtZUR5+ogbeOZTOsfQ38g@mail.gmail.com"
type="cite">
<div dir="ltr"><br>
<div class="gmail_extra"><br>
<div class="gmail_quote">On 21 September 2015 at 12:06, Marek
Posolda <span dir="ltr"><<a moz-do-not-send="true"
href="mailto:mposolda@redhat.com" target="_blank">mposolda@redhat.com</a>></span>
wrote:<br>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex">I've
sent the PR . Right now it works like this:<br>
<br>
- ClientModel has flag "offlineTokensEnabled" . It's
possible to<br>
retrieve offline tokens just if flag is enabled<br>
<br>
- Offline token is classic refresh token with 2
differences. It has type<br>
"OFFLINE" when normal refresh token has type "REFRESH" .
And for offline<br>
token, the expiration value is 0, so it never expires.<br>
<br>
- Offline token is generated by auth-server when client
sends<br>
"scope=offline_access" . It's supported for classic
browser flow, but<br>
also for Direct Grant flow or Service account flow.<br>
<br>
- I've added OfflineClientSessionModel and
OfflineUserSessionModel with<br>
CRUD methods on UserModel. So when new offline token is
generated by<br>
Keycloak, some info about current UserSession and
ClientSession is<br>
persisted on UserModel. This means that offline token can
be used to<br>
create new access token even if "normal" UserSession and
ClientSession<br>
are already invalid or logged out.<br>
<br>
- When refreshing access token with offline token, the
auth-server won't<br>
send back another refresh token. It will send just
accessToken +<br>
IDToken. This is to avoid writes to user database for each
token refresh.<br>
<br>
- In account management applications tab, there is new
table column<br>
"Additional grants" where is shown if client has offline
token for user.<br>
The click on "Revoke" button will remove offline tokens
and granted<br>
consents as well - no separate actions for revoke consents
and offline<br>
tokens.<br>
<br>
<br>
Still TODO:<br>
- Properly handle consents (see "Questions" below)<br>
<br>
- More tests, example, export/import , docs<br>
<br>
- More things/refactoring based on your feedback<br>
<br>
<br>
Questions:<br>
- The specs mentions that consent should be displayed when
offline token<br>
is requested. See<br>
<a moz-do-not-send="true"
href="http://openid.net/specs/openid-connect-core-1_0.html#OfflineAccess"
rel="noreferrer" target="_blank">http://openid.net/specs/openid-connect-core-1_0.html#OfflineAccess</a>
.<br>
Right now, I am not doing that. So when Client has
"isConsentRequired"<br>
as false, the consent screen is not displayed. Now we also
don't have<br>
support for "prompt=consent" (not sure if we need this) .
Is it ok to<br>
keep it like this?<br>
</blockquote>
<div><br>
</div>
<div>Wording is "<span
style="color:rgb(0,0,0);font-family:verdana,charcoal,helvetica,arial,sans-serif">MUST
explicitly receive or have consent" - as the client does
not require consent (</span>isConsentRequired=false)<span
style="color:rgb(0,0,0);font-family:verdana,charcoal,helvetica,arial,sans-serif"> that
implies the client already has been given consent by the
admin.</span></div>
<div> </div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><br>
- I am thinking about adding new builtin client role
"offline_access",<br>
which will be created for client when admin enables
"offline tokens"<br>
switch. It will be used also as default role. This will
allow that just<br>
some users are allowed to obtain offline-token (those
which have this<br>
role). The role will be also displayed on consent screen
for the<br>
clients, which needs consent.<br>
But that raises another question. IMO it will be good if
role is<br>
requested and displayed on consent screen just if offline
token is<br>
requested, but not when classic refresh token is
requested.</blockquote>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><br>
Hence I was thinking about adding the flag
"scopeParamMode" to<br>
RoleModel. The value true means that role will be
requested and used in<br>
accessToken/refreshToken just if scope parameter contains
it's value.<br>
This will be the setup for "offline_access" role, so it's
used just for<br>
the offline token requests. Another thing is format of
scope parameter<br>
with respect to realm roles and application roles. We can
use "//" as<br>
delimiter, so realm role will have just "my-role" but
client role will<br>
have "my-client//my-role" . The disadvantage is that for
requesting<br>
offline_access you will then need to use scope like:<br>
"scope=customer-portal//offline_access" as it's client
role.<br>
</blockquote>
<div><br>
</div>
<div>Spec says scope should be "offline_access", so if we
use a different name for it won't comply with the spec.</div>
<div><br>
</div>
<div>Shouldn't the offline_access role be a realm role
rather than a role per-client?</div>
</div>
</div>
</div>
</blockquote>
The issue with realm role is, that user needs to confirm
offline_access consent just once per all clients, which doesn't look
correct to me. <br>
<br>
For example there would be 2 clients "foo" and "bar":<br>
- User john wants offline_access for client "foo"<br>
- Consent screen is displayed with "offline_access" role and he
confirms it<br>
- Then user john wants offline_access for client "bar"<br>
- There is no "offline_access" anymore on consent screen because
john already has consent for realm role "offline_access" . <br>
<br>
<br>
IMO it should be "offline_access" displayed again as it's different
client. Also logically offline_access is granted per client, so the
text on the grant screen is:<br>
<br>
"has Offline Access in foo"<br>
<br>
or <br>
<br>
"has Offline Access in bar"<br>
<br>
which will be with client roles.<br>
<br>
<blockquote
cite="mid:CAJgngAd+J8Q6C=Wi3Kz9JUvRnzfjeqtZUR5+ogbeOZTOsfQ38g@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div><br>
</div>
<div>Another thing to consider is that we'll be moving to
role namespaces instead of realm/client roles soon. In
that case we might want a OpenID Connect namespace that
can hold these scopes. So role could be
"openid/offline_access".</div>
</div>
</div>
</div>
</blockquote>
Ok. Maybe for now I won't do anything tricky but just limit the
scope param support to client roles of current client. So scope
"offline_access" is "offline_access" role of current client. We can
improve it later once we add role namespaces support. WDYT? <br>
<br>
Marek<br>
<blockquote
cite="mid:CAJgngAd+J8Q6C=Wi3Kz9JUvRnzfjeqtZUR5+ogbeOZTOsfQ38g@mail.gmail.com"
type="cite">
<div dir="ltr">
<div class="gmail_extra">
<div class="gmail_quote">
<div> </div>
<blockquote class="gmail_quote" style="margin:0px 0px 0px
0.8ex;border-left-width:1px;border-left-color:rgb(204,204,204);border-left-style:solid;padding-left:1ex"><br>
WDYT? Any better idea?<br>
<br>
Marek<br>
_______________________________________________<br>
keycloak-dev mailing list<br>
<a moz-do-not-send="true"
href="mailto:keycloak-dev@lists.jboss.org">keycloak-dev@lists.jboss.org</a><br>
<a moz-do-not-send="true"
href="https://lists.jboss.org/mailman/listinfo/keycloak-dev"
rel="noreferrer" target="_blank">https://lists.jboss.org/mailman/listinfo/keycloak-dev</a><br>
</blockquote>
</div>
<br>
</div>
</div>
</blockquote>
<br>
</body>
</html>