Shibboleth

Rationale

ShibbolethThe university's existing home-grown, cookie-based authentication system (GLAuth) has security problems and must be replaced. In addition, the university needs to support federated authentication services for interaction with contracted service providers. Existing authentication systems do not support the most prevalent web servers on campus, Apache and IIS for Windows and Linux. Finally, the university needs to enable its departments and units to consume enterprise attributes for authorization of access in an efficient, scalable and secure manner. Shibboleth has been identified as an appropriate system for addressing these needs. By integrating Shibboleth with existing credential and attribute stores, we will be able to meet the four identified challenges.

Goals

Project Sponsor

Contacts

Mail Listservs

Impact

End users will see a single place to sign on. All existing cookie-based authentication will be replaced including GLAuth and CoSign. This will impact over 100 departments and units using these technologies. Enterprise system work will be needed on several major systems. This work varies in complexity but will be transparent to the user.

Lab work will begin in February. A working development model will be in place in March. Preliminary assertions will be identified in March. Implementation of assertions will begin in May. Production infrastructure will be in place in July. Testing will be completed in August. Production services will be available in September. An enterprise system roadmap will be developed during the planning phase along with a roadmap for sunsetting GLAuth and CoSign services.

Presentations

Presentations of the Shibboleth Planning Team are also available.