shibboleth-dev - RE: the big question at the end of this week's call.....
Subject: Shibboleth Developers
List archive
- From: Scott Cantor <>
- To: 'Keith Hazelton' <>, 'shib-dev' <>
- Subject: RE: the big question at the end of this week's call.....
- Date: Wed, 03 Dec 2003 21:41:01 -0500
- Importance: Normal
- Organization: The Ohio State University
> Looks like you're saying look outside Shib proper for the tools to roll
> this kind of app, or wait for Shib/Lib/SAML convergence.
Yeah, I guess. But let's say I needed this functionality today and I didn't
want to make code changes to Shib.
The origin side is simple, it amounts to deploying a second HS with a
properties file specifying the higher end auth method, and then protect it
with my higher end authentication.
Now I either set up a special WAYF that treats the two different methods as
a distinct origin site (even if they aren't really different), or I setup my
target to use vhosts that go directly to one HS or the other.
And I would, in keeping with my usual approach, do nothing at the
application level to make it happen. I'd move the scripts that were subject
to this higher requirement to /secure/medapp or a different vhost and let my
web server configuration do the work. Such is the magic of URLs.
I simply check the auth method header Shib provides me and I know which
policy was used.
This will be much easier when it's part of the protocol, but web servers can
do a lot of this for me today.
Lastly, I'd always warn anyone doing this that I would never attack your
authentication system to crack your medical application. I'd attack your
application's session cookie. As long as the client is a browser, there's
not all that much you can do except to use client certs and check the cert
on every request.
I think a really interesting question is why this authentication context
stuff is so superficially appealing to people. I must be missing something.
;-)
-- Scott
- the big question at the end of this week's call....., Steven_Carmody, 12/03/2003
- RE: the big question at the end of this week's call....., Howard Gilbert, 12/03/2003
- Re: the big question at the end of this week's call....., Keith Hazelton, 12/03/2003
- RE: the big question at the end of this week's call....., Scott Cantor, 12/03/2003
- Re: the big question at the end of this week's call....., Keith Hazelton, 12/03/2003
- RE: the big question at the end of this week's call....., Scott Cantor, 12/03/2003
- Re: the big question at the end of this week's call....., Barry R Ribbeck, 12/04/2003
- RE: the big question at the end of this week's call....., Scott Cantor, 12/04/2003
- RE: the big question at the end of this week's call....., Steven_Carmody, 12/04/2003
- Re: the big question at the end of this week's call....., Barry R Ribbeck, 12/04/2003
- RE: the big question at the end of this week's call....., Scott Cantor, 12/03/2003
- Re: the big question at the end of this week's call....., Keith Hazelton, 12/03/2003
- RE: the big question at the end of this week's call....., Howard Gilbert, 12/03/2003
- RE: the big question at the end of this week's call....., Scott Cantor, 12/03/2003
- Re: the big question at the end of this week's call....., Keith Hazelton, 12/03/2003
- RE: the big question at the end of this week's call....., Howard Gilbert, 12/03/2003
Archive powered by MHonArc 2.6.16.