<html xmlns:v="urn:schemas-microsoft-com:vml" xmlns:o="urn:schemas-microsoft-com:office:office" xmlns:w="urn:schemas-microsoft-com:office:word" xmlns:m="http://schemas.microsoft.com/office/2004/12/omml" xmlns="http://www.w3.org/TR/REC-html40">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
<meta name="Generator" content="Microsoft Word 14 (filtered medium)">
<style><!--
/* Font Definitions */
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0in;
margin-bottom:.0001pt;
font-size:11.0pt;
font-family:"Calibri","sans-serif";}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
a:visited, span.MsoHyperlinkFollowed
{mso-style-priority:99;
color:purple;
text-decoration:underline;}
span.EmailStyle17
{mso-style-type:personal-compose;
font-family:"Calibri","sans-serif";
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-family:"Calibri","sans-serif";}
@page WordSection1
{size:8.5in 11.0in;
margin:1.0in 1.0in 1.0in 1.0in;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
</head>
<body lang="EN-US" link="blue" vlink="purple">
<div class="WordSection1">
<p class="MsoNormal">Session Management / Logout Discussion at IIW 6-May-14<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Back channel logout<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Multiple RP sessions might be logged in at once<o:p></o:p></p>
<p class="MsoNormal">OP doesn't have session identifiers<o:p></o:p></p>
<p class="MsoNormal">Torsten - back channel logout has been shown to cause problems<o:p></o:p></p>
<p class="MsoNormal">Chuck - the session-based SAML logout doesn't work well and isn't supported<o:p></o:p></p>
<p class="MsoNormal">Brian - requiring JavaScript on every page a non-starter in enterprise contexts<o:p></o:p></p>
<p class="MsoNormal">Alan Karp - Better UI could help users better understand what's actually going on<o:p></o:p></p>
<p class="MsoNormal">John - one requirement can be extending session lifetimes across sessions<o:p></o:p></p>
<p class="MsoNormal">Chuck - The only session state typically present is the session cookie<o:p></o:p></p>
<p class="MsoNormal">Brian - SAML has fragile redirect chain<o:p></o:p></p>
<p class="MsoNormal"> Another means is for the IdP to do a GET to each RP endpoint<o:p></o:p></p>
<p class="MsoNormal"> Ping is currently implementing something like that<o:p></o:p></p>
<p class="MsoNormal">Naveen - Unless the browser tab is active, the JavaScript logout doesn't work<o:p></o:p></p>
<p class="MsoNormal">Naveen - Yahoo had a variant where they stored state for all sessions in a single cookie<o:p></o:p></p>
<p class="MsoNormal">Brian - Doing GETs to single-pixel images, which trigger logouts<o:p></o:p></p>
<p class="MsoNormal">John - The RP could check the referer if it wants to secure the logout<o:p></o:p></p>
<p class="MsoNormal"> But in general, not protectable against XSRF<o:p></o:p></p>
<p class="MsoNormal">Chuck - Browsers are getting better about preventing cookie state manipulation in iframes<o:p></o:p></p>
<p class="MsoNormal">Torsten - Will check what DT is doing<o:p></o:p></p>
<p class="MsoNormal">John - ID Token "exp" claim doesn't trigger logout in practice<o:p></o:p></p>
<p class="MsoNormal">Naveen - We could just document both front channel mechanisms as optional<o:p></o:p></p>
<p class="MsoNormal"> Enterprises might choose one method, the Web might choose another<o:p></o:p></p>
<p class="MsoNormal"> We should document both as a next step<o:p></o:p></p>
<p class="MsoNormal">Mike - If we support multiple mechanisms, the RPs would get to decide<o:p></o:p></p>
<p class="MsoNormal">John - Back channel notification is yet a third mechanism<o:p></o:p></p>
<p class="MsoNormal">David Pinter - The front channel won't always be available<o:p></o:p></p>
<p class="MsoNormal">Bill Mills - Security policy may require ability to kill all sessions<o:p></o:p></p>
<p class="MsoNormal">John - A compromise back channel mechanism is notifying RPs on the back channel of a state change<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Mechanisms:<o:p></o:p></p>
<p class="MsoNormal"> postMessage:<o:p></o:p></p>
<p class="MsoNormal"> Pro: RPs get notifications, minimal web traffic<o:p></o:p></p>
<p class="MsoNormal"> Con: Requires RP JavaScript<o:p></o:p></p>
<p class="MsoNormal"> Doesn't work when RP tab not active<o:p></o:p></p>
<p class="MsoNormal"> image/iframe GETs:<o:p></o:p></p>
<p class="MsoNormal"> Pro: Doesn't require JavaScript<o:p></o:p></p>
<p class="MsoNormal"> Still uses session cookies<o:p></o:p></p>
<p class="MsoNormal"> Con: IdP needs to track active RP sessions<o:p></o:p></p>
<p class="MsoNormal"> Ugly logout page<o:p></o:p></p>
<p class="MsoNormal"> All RPs might not be notified before the browser is closed<o:p></o:p></p>
<p class="MsoNormal"> backchannel notifications:<o:p></o:p></p>
<p class="MsoNormal"> Pro: Works even when RP tab not active<o:p></o:p></p>
<p class="MsoNormal"> Con: Requires RP logic to identify and communicate with session to logout<o:p></o:p></p>
<p class="MsoNormal"> IdP scaling issues<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Chuck - Current spec doesn't work for enterprise use cases because of JavaScript requirement<o:p></o:p></p>
<p class="MsoNormal"> and because the RP must be active for the logout to work<o:p></o:p></p>
<p class="MsoNormal">Chuck - We didn't put "jti" in the ID Token - we could for enabling logout<o:p></o:p></p>
<p class="MsoNormal"> That would enable correlating back channel notifications received to active sessions<o:p></o:p></p>
<p class="MsoNormal"> Open questions about whether to use the same ID in multiple responses to same RP<o:p></o:p></p>
<p class="MsoNormal"> Probably use a separate session identifier that is not "jti"<o:p></o:p></p>
<p class="MsoNormal">Dale - Back channel notification is effectively ID Token revocation<o:p></o:p></p>
<p class="MsoNormal">Chuck - The OP wants to be authoritative for the session ID<o:p></o:p></p>
<p class="MsoNormal">Chuck - Revoking a session could be done like an OAuth revocation<o:p></o:p></p>
<p class="MsoNormal"> OAuth revocation supports CORS and JSONP<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Actions:<o:p></o:p></p>
<p class="MsoNormal"> Add image/iframe description to Session Management<o:p></o:p></p>
<p class="MsoNormal"> Also describe back channel mechanism<o:p></o:p></p>
<p class="MsoNormal"> Then we decide what to do after that<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Naveen, Breno - Google is planning to build a token caching layer<o:p></o:p></p>
<p class="MsoNormal"> It would get tokens at login time and clear them at logout<o:p></o:p></p>
<p class="MsoNormal"> It would send notifications when things change<o:p></o:p></p>
<p class="MsoNormal"> It would communicate internally with postMessage<o:p></o:p></p>
<p class="MsoNormal"> postMessage requires a security layer<o:p></o:p></p>
<p class="MsoNormal">George - How is this like the Trusted Agent in the Native Applications work?<o:p></o:p></p>
<p class="MsoNormal">Naveen - It's a lot like that<o:p></o:p></p>
</div>
</body>
</html>