On 04/19/2014 07:46 PM, Nordgren, Bryce L -FS wrote:
I've run out of time for today, but the external collaboration pages
are slowly evolving.
Dimitri observed that my RFE page was too long. I observe it also has
too much stuff unrelated to the actual meat of the RFE. So I factored
out most of the Kerberos stuff into a different page. I also tried to
focus the RFE to just creating entries in LDAP for external users so
they can: a] participate in POSIX groups; and b] have locally-defined
This is where all the Kerberos stuff went. I also added in "Option A"
from Petr's email. Option B will come along later, when I pick this up
again. Mechanism three has more to do with Ipsilon than IPA, and basic
functions required of the Ipsilon gateway server are articulated there
(regardless of the particular authentication method.)
Send comments to the list. I really appreciate Option A! Send more
stuff I didn't think of.
I finally read the pages, sorry for the delay. great writeup!
Here are some comments.
1) You are right that we need to have a record in IPA to be able to have
a DN and take over some of the posix attributes. We already have this
use case and this is a high priority. We call it views:
Once this is implemented we will have mechanism to have a local entry
without credential for the external user.
2) The second issue is provisioning as automatic as possible. And this
is where there will be some issues.
If we want to leverage Kerberos trusts then two things should happen:
a) the trust should first be established
b) the home realm should be accessible for the KDC in the collaboration
This rises practical operational questions about what is the home
domain. If the home domain is another collaboration domain then user is
natively have been created in that domain and has his credential in that
domain. Hm but that violates the idea that the collaboration domains
have external "auto-provisioned users". If the home domain is the
internal domain than most likely the cross forest trust can't be
established because admin of the internal domain would not want to
expose his domain to somebody's external domain on the internet.
So IMO the kerberos based auto-provisioning falls apart.
However if we use a gateway that would allow a person to self register
and use technologies similar to OpenID then we would be able to create
his own account. The gateway would check that the user is from some
trusted source that is configured for that domain. We would have to
figure that part out. But IMO this component is external to IPA. It is a
similar gateway to Ipsilon. I suspect that as we move forward Ipsilon
will transform from an IdP server to being a collection of "gateway
services". One would be able to deploy IdP instances, Kerberos -> cert
service, account registration service etc.
This would rely on some of the functionality in IPA but can evolve
IMO if we go this path and you are interested in contributing to this
effort we can start prototyping such service.
We can start simple: create a service that allows one to authenticate
using google or facebook and once user authenticated agains one of them
call an ipa user-add against IPA.
That would be a good first step towards what you want to accomplish.
Then it can be enhanced to redirect to an external IdP (Ipsilon). Then
the setup will be:
* User connects to the self registration portal.
* Portal reditrects him to the IdP that is configured for the portal
* IdP performas an authentication against user home domain and creates
* Assertion is presented to the registration portal
* The portal gets user infor from the assertion and adds a user
It also seems that OpenID connect might be quite relevant here.
So exploring how it can be used in in conjunction with registration
portal would be another path.
3) The problem of the credential yet stays open. If the user can be
created in different ways it might not be quite easy for the user to
know or remember that he must use his kerberos/Google/facebook or other
credential wit ha specific domain. May be we should consider creating a
full user also with a password or OTP token to access the collaboration
domain. Then user would always know that he needs to use his token. I
wonder if actually just OTP would be a good option in this case. It can
be provisioned to the freeOTP app at the moment of the user registration.
Bottom line: let us do practical steps in the right direction. The whole
project seems to have too many weak points to try to solve it as a
single use case.
I would rather focus on building technologies that would gradually make
life of collaboration domains easier and get there over time.
This electronic message contains information generated by the USDA
solely for the intended recipients. Any unauthorized interception of
this message or the use or disclosure of the information it contains
may violate the law and subject the violator to civil or criminal
penalties. If you believe you have received this message in error,
please notify the sender and delete the email immediately.
Freeipa-users mailing list
Sr. Engineering Manager IdM portfolio
Red Hat, Inc.
Freeipa-users mailing list