Incubator Dashboard (GN4-3)

Self-sovereign identity and the AARC BPA

The AARC Blueprint Architecture (BPA) describes a "Community AAI" solution, a set of software building blocks that can be used to implement federated access management solutions for (inter)national research collaborations. The benefit of the BPA is that its proxy-based architecture provides both a technical integration point for authentication and authorisation, as well as a centralised point for implementing the research communities' policies. The BPA also identifies a "membership management service" which implements community-specific onboarding to help establish the researcher's status and may be used to issue community-specific attributes to establish roles and rights. Implementations of the BPA, like eduTEAMS and SRAM, have greatly improved the capability to use FIM for research communities.
At first glance, a SSI based model may offer similar benefits as the AARC BPA model, while reducing the number of impediments as a wallet model may take away the need to have a proxy as the central authentication gateway. This activity will further explored the potential use of SSI technology in the context of the AARC BPA. It will describing where SSI technology may be leveraged, explore benefits and challenges and describe how that may be implement. A number of technical pilots will test the assumptions.

Account Linking

Account linking is, in principle, a simple process of joining users digital identities. Users have to prove that they own two or more digital identities, which are consecutively linked together in the identity management system. CESNET has very positive feedback on ELIXIR workflow for account linking. User signs in with one identity, then with a second one, and the identities are linked. The positive feedback is coming not only from the ELIXIR community but also from users from other communities, even outside the life-science area. The significant issue of the workflow is its user interface, which is still too complicated for users to navigate through smoothly. Because the process consists of several steps, skipping one step or leaving the linking process unfinished makes it even more prone to errors. Users need a straightforward interface that will guide them step-by-step through the whole process.
The goal of the topic is to develop a new web application for account linking, test it with end-users and improve it based on user feedback. The application has to support a recommendation system that tries to guess linkable identities. It also has to deal with all possible situations in a user-friendly way. The situations include linking identities that already belong to different registered users or linking identities neither of which belongs to a registered user.

Alternative MetaData eXchange

With the growth of the metadata feeds all over the world, together with the increased need for catering for multiple federations at once, Service Providers are increasingly reliant on the MetaData Query (MDQ) protocol, for Relying Party metadata lookup. (MDQ is also commonly referred to as MDX (MetaData eXchange) for historical reasons: the name of a mailing list where the spec was discussed.) This protocol is however not widely deploy at this time, due to technical complexity In addition, for MDQ there is only one implementation in common use: pyFF. This is a risk as it creates a high dependence on that piece of technology.
This activity seeks to simplify the scalable use of MDQ, both for federations that currently operate a MDQ as well as for federations tat do not provide an MDQ service. In addition it wants to create an alternative implementation, next to pyFF, for creating and managing XML metadata. The aim is to deliver EntityDescriptors as flat files in such a way this mimics MDQ protocol and combine that with Global DNS to create a super fast and highly redundant distribution mechanism for MDQ.

Acceptable Use Policy Service

The primary goal of the Authentication and Authorization Infrastructure (AAI) is to provide centralized authentication and authorization mechanisms. In an environment based on the AAI, one of such authorization rules might be fulfilling the policy requirements, set i.e. in an AUP document defined by the community, service provider, or organizational unit. In general, when the resource owner or provider decides that a specific policy has to be enforced when the resource is used, AAI needs to provide a mechanism for both sides to manage and fulfill this requirement. These rules need to cover several cases, i.e. a situation of actively using the resource (i.e. web-based service login), as well as just-in-case scenarios, like the data provisioning.
This topic aims to explore the area of enforcing acceptance of Acceptable Use Policies as an activity of managing access to resources. We like to understand which parties (e.g. user communities, e-infrastructures, resource owners, …) need to be involved in the process and how to combine their requirements together. Based on the analysis we will develop a web-based application which will provide tools to manage AUPs on a central level (within the AAI) and let users approve (whilst recording this act) such a policy document.

Community Tagging

Research communities have a need to express and potentially share certain trust marks on IdPs and SPs. These trust marks may differ from existing trust marks issued by identity federations. This is why this activity implemented a proof of concept based on a given set of community requirements and investigated in potential scenarios and impacts.

Community-Based Trust

This activity examined how identity vetting and token registration can be scaled for second-factor authentication scenarios where participants are distributed over EU and beyond. As part of this activity a specific flow - based on a community-based approach - was investigated. It takes into account the concept of the Web of Trust. While this mechanism typically does not work well in broad user groups, it is very well suited to distribute trust between small groups where a pre-existing trust fabric is already in place, as we typically have in research communities. What was missing is a means to make this trust network auditable and transparent. In order to make the trust network transparent, this activity identified and developed tools to support this flow.

Cryptech HSM

In many of the T&I services in the R&E sector, there is a need to securely store sensitive data like key material. Currently, this is rarely done using Hardware Security Modules (HSM) as they are associated with high costs. The goal of this activity was to evaluate the capability and applicability of affordable devices from the Cryptech project for use cases within the GÉANT project and to set up devices for testing purposes.

Distributed Identity Revisited

Discovery Pilot

Discovery is used in Federated Identity Management to locate the users home organization. This activity evaluated the pilot discovery service ran in the previous GÉANT project and investigated if/how the implementation technology Seamless Access developed within the RA1 project can be used for an implementation in eduGAIN.

Distributed Identity for researchers

A new paradigm, Distributed Identity, tries to let user be in direct control of the profile information they share with services. Rather then letting others provide claims towards a service, the users collects claim themselves from various sources and independently provides these when so requested by services. The services can check the validity of these claims against a central verifiable claims registry. This activity investigates the functional requirements for such a system by interviewing key stakeholders.

SFA - Distributed Vetting

Several research communities, especially in the life sciences domain, have a need to use second factor authentication to improve the quality of their authentication. One of the challenges identified was how to securely vet the second factor tokens of the participants of a collaboration in a case where the members of the collaboration are very distributed, as is the case in most pan-EU research collaborations. This activity investigated, together with research communities, how the token registration can be scaled for scenarios where participants are distributed over the EU and beyond. The aim of this task was to identify ways this vetting can be done.

eduLNK

eduLNK aims to provide software and potentially a servcie for a secure, persistent and privacy preserving link shortener for research and education. Link shorteners reduce a lengthy URL on the internet to something short and snappy, and sometimes easy to remember. As such it is a very well used and liked functionality within our community. Many ‘free’ link shorteners exist, however, their free nature comes with a number of challenges, including serving advertisements and data mining the users behaviour and interests. Next to privacy concerns, many free link shorteners have a limited lifetime for the links they present. eduLNK aspires to deliver a product from and for the community which can be trusted and used securely in a privacy preserving way.

Instant User Provisioning

Some systems, like non-web services, cannot be federated easily because they need user accounts to be provisioned before they can login. A prototype of an instant deployment tool called FEUDAL was developed by KIT. It facilitates provisioning of user accounts using virtual organisations (VO). Feudal is based on OIDC: It is an OIDC client, and it simply transports the information of the /userinfo endpoint along. Feudal is based on the concept of VOs (or authorisation Groups), i.e. the end services provide the information which VOs it supports. Feudal web fronted will only display services for provisioning to a given user based on his VO membership.

Identity Validation Broker

Both research communities as well as institutions have a need for proper proof of the identity of their users. This ranges from cases where e.g. new students living abroad need to be identified as part of the boarding into an institution, to scenarios where access to (medical) data puts regulatory demands on the research community. Identity proofing is however expensive and scales very poorly, especially in cases where the users are (very) distributed. This activity investigates, based on requirements collected from multiple stakeholders, if and in which way this situation may be improved.

IdP as a Service

The former GN4-2 project developed a solution to offer an IdP as a Service solution (IdPaaS) for hosted IdPs. This activity aimed at investigating the business case of this solution and to create a reference design and implementation for an easy to use software solution to enable NRENS to provide an IdPaaS offering.

Metadata push MDQ

Metadata is at the heart of the trust fabric of current R&E Identity Federations. For the trust to properly propagate, this metadata is first collected from and then distributed by the federation towards the federation members. This activity had investigated a new proposal called "push MDQ", which introduces a new, potentially highly scalable way of distributing metadata.

Porting oidc-agent to Windows

oidc-agent is a command line set of tools that enables the users to manage and obtain (mostly) OIDC tokens on command line. It follows the design of the ssh-agent and, as such, it can be easily integrated in the user's flow. However, current version only covers Linux and MacOS users. This activity would port the oidc-agent to Windows OS. The expected impact is potentially large, since users from many communities that may use this tool (such as HPC, or any community that need constant access to computing and storage resources as part of their work) typically rely on Windows as their main operating system. This activity investigates the best approach to porting the oidc-agent.

SimpleSAMLphp OIDC OP

SimpleSAMLphp (SSP) is a commonly used software product for both SP and IdP deployments in Research and Education. In addition it may also be deployed as a proxy. Next to SAML, various other authentication protocols are supported.While SSP already supports the OpenID Connect (OIDC) Relaying Party (RP) interface, an OIDC OpenID Provider (OP) implementation is missing.
This activity seeks to implement an OIDC OP in accordance with the OIDC specification into SSP. There is an existing module available at Github. It needs to be investigated whether this is suitable for this activity.

ORCID as IdP of last resort

Many research collaborations as well as campus services need a solution to deal with guest identity, as in many cases not all users are members of the academic Identity Federations. As a result, several federation operators as well as research collaborations operate IdPs or proxies to allow users to authenticate through external identity providers like social ones. This has led to serious reinventing of the wheel. This pilot aimed to bring the widely used ORCID service into the GÉANT IDhub as Identity Provider of last resort. Furthermore organisational and legal aspects as well as technical improvements were investigated.

pyFF Optimizations

pyFF is an open source and widely used product which is used to provide Discovery and Metadata Query services for identity federation. This topic investigates optimizations in pyFF operations, including but not limited to, performance and memory consumption. When processing the eduGAIN metadata, pyFF memory usage balloons to the gigabytes, hereby inflicting some extra cost when running on procured VM's like AWS. The startup/restart process speed, and service behavior while being started/restarted may also be improved. In particular, the service should never throw 5xx errors while in a normal startup/shutdown process.
The goal of this project is to optimize pyFF memory consumption and (re-)start behavior.

eduGAIN Service Catalogue

The eduGAIN Support Team, Secretariat and business development received several requests from SPs, IdPs and NRENs for an eduGAIN Service Catalogue. Currently the eduGAIN Entities Database tool and REFEDS Metadata Explorer Tool (MET) are the two services closest to a service catalogue, but there are inconsistencies between the two tools.
The Incubator attempts to improve the solution based on the outcome of this pilot phase. Depending on the requirements and use cases that need to be supported, new features can be added in following an iterative process. If there are not enough participants to provide a meaningful amount of data, no additional developments will be carried out by the Incubator. In this case, the software is decommissioned at the end of the pilot phase and the project is terminated.

Service Status Reporting

The T&I activities of the GEANT project run a multitude of services ranging for eduroam and eduGAIN to InAcademia, eduTEAMS and various smaller helper services for eduGAIN without a single overview of the state of all these services. The highly distributed nature limits our ability to present in a consistent way the status of these services and it hinders us in explaining issues when something is wrong. This is relevant both for members of the GEANT community as well as other other stakeholders like service owners and funding agencies. This activity wanted to create a comprehensive, high level and user friendly publicly facing service dashboard for T&I services. Since many freely or commercially available services for measuring availability already exist this activity investigated and compared them to choose a suitable solution for the GÉANT project.

Shibboleth OIDC Extension

Up to now, R&E federations are predominantly relying on the SAML2 protocol. With upcoming needs from industry and commercial service providers the OpenID Connect (OIDC) protocol is increasingly moving into focus. This activity supported the development of an Shibboleth OIDC Extension to the Shibboleth IdP and was paving the way to make the extension a fully sustained product.

IdP/SP software testbed

It is currently very hard to test new releases of IdP and SP software against other IdP/SP products. This activity attempts to design an automated test environment. A suitable architecture for an IdP/SP continuous integration platform is to be created. This involves creating use cases and identifying technologies that support them. An iterative approach, in which use cases are successively implemented and products are integrated one after another, is intended to quickly create a functional platform. If the CI approach is successful, ready-to-use software packages can be provided using container technologies such as Docker, to deploy the entire software locally by NRENs or other parties.

SSH in a Federated World

To allow easy access to SSH based services DeiC has made a SSH Certificate Authority proof-of-concept that issues short-lived SSH certificates based on a federated login. The system requires no specific client - or service side installed programs and makes it possible for the user to use all standard ssh services - as long at the certificate is valid. Depending on the configuration of the participating services the CA allows the user to use the same username/uid across all services. Optionally it can be combined with systemd-userdb services to allow for fully automated user management. The CA can also optionally issue host certificates so the users do not have to trust the servers on first use (TOFU). Initial goal of the activity is to hold a workshop to gather requirements and showcase and discuss existing solutions.

Status Reporting & Notification

To make the global eduGAIN interfederation scalable and interoperable, it is paramount participants adhere to agreed upon standards and community best practices. There are currently already several tools that examine the quality of federation metadata, the connection status of their IdPs and the attribute release status of their IdPs. This activity is about creating a tool to aggregate information from various sources and presenting this in an attractive, easy to understand way to put this information readily into the hands of the federation operators.

Test IdP

This activity creates a freely available, realistic test IdP for the GÉANT community. Based on the needs of the GÉANT community, the solution may offer SAML, OIDC and supports all the latest best practices. The requirements and use cases for such a service will be defined in collaboration with a group of community representatives. Potential features of such a tool are: test for a specific set of attributes, create an account to save a test profile, an open (REST) API to configure the IdP using a (web) client. There are different potential business cases for deployment: as part of the eduGAIN support tools, by GÉANT operations or NREN hosted.

IdP User Profile

Currently, there is no way for federated users to know at which service providers their account was used to login. This activity developes a mechanism to track federated signing in events. That would allow an user to check the recent activity on his account in regards of authentication. Users could see the list if authentications containing datetime, ip and relying party etc. That would help them to spot suspicious activity.
A user profile page deployed as part of Shibboleth IdP and SimpleSAMLphp would enable end users to gain insight into where their personal data is used and when it was last released to various services, as far as the IdP is aware. This feature should only release information to appropriate user (so after login). It needs to be considered how storing user data to facilitate this plugin would impact data retention policy of the IdP.

WebauthN

WebAuthn​ (Web Authentication), part of the FIDO2 Project, is a web standard published by the W3C that enables strong authentication with public-key cryptography, passwordless authentication, and secure two-factor authentication. The standard defines a JavaScript API which allows token registration and subsequent authentication. This activity implemented this API for the open source software SimpleSAMLphp and SATOSA, which enables them to provide 2FA functions to the benefit of the R&E community.

(De)provisioning connector for Windows

Identity provisioning and deprovisioning are a necessity for building modern authentication and authorization infrastructures. They are straightforward yet technically complicated part of identity and access management. The basic idea is to deliver identity and authorization information to the managed services, which is complicated by a lack of applicable standards in this area. Therefore, most of the Identity and access management solutions rely on a custom solution for provisioning. To overcome this obstacle, this project extends existing IAM capabilities by implementing a connector to easily provision data to services hosted on Windows OS based on SSH.