Service Description: Service to allow a user to see if his eduGAIN IdP is releasing attributes properly, not too many and not too few. Service URL is http https://release-check.edugain.org/)
Components: Uses a LAMP stack with PHP (using Laravel 5.4 framework) and MySQL.
Code Repository: https://code.geant.net/stash/projects/GN4SA2T2/repos/edugain-attribute-release-check/browse
Infrastructure:
- Test infrastructure: uat test-edugain01.geant.net (change local hosts file for release-check.edugain.org to point to this host )
- Production infrastructure: prod-edugain01.geant.net (service name http https://release-check.edugain.org/)
Operational Information: No regular operational maintenance needed as far as we know
Roadmap/ToDos:
...
- EARC - REFEDS Research and Scholarship Test -> REFEDS R&S Test with Requested Attributes
- EARC - REFEDS Research and Scholarship NG -> REFEDS R&S Test
- EARC - GEANT Data Protection Code of Conduct Test -> GEANT Data Protection Code of Conduct Test
- EARC - No Entity Category Test -> No Entity Category Test Changes SP MDUI DisplayName to
- EARC - REFEDS Research and Scholarship Test -> EARC - REFEDS Research and Scholarship with Requested Attributes Test
- EARC - REFEDS Research and Scholarship NG -> EARC - REFEDS Research and Scholarship Test
...
- API should allow to query results of a particular IdP and to ask which grades an IdP would get if releasing certain attributes to a test SP. Some API calls need yet to be defined and then documented somewhere (e.g. wiki.edugain.org). Should not have high priority as Tomasz/Maja asked for this based on false assumptions about EARC initially. However, others (e.g. Niels) also have some use for an API.
...
- See above, should not play a role anymore if both are treated the same way (= without penalty)
...
- Add links to existing R&S, CoCo documentation if grade is worse than an A.
...
- We don't have a good idea how to grade the results of this test because we don't know what should have more precedence: usability or data privacy or a mix of both.
- Remove grading as it is not easily possible to do a proper grading but keep information on what is released.
- Replace verdict with a hint (no attributes = good data privacy but bad usability, all attributes = the other way around)
...
Deployment Information: ansible-playbook -i ./hosts -l $ENV ansible.yml (where $ENV is test or prod)
Operational Information: No regular operational maintenance needed as far as we know
...