FIWARE Data Space Connector

The FIWARE Data Space Connector is an integrated suite of components implementing DSBA Technical Convergence recommendations, every organization participating in a data space should deploy to “connect” to a data space. The implementation of these recommendations is developed as soon as they become enough mature.

This repository provides a description of the FIWARE Data Space Connector, its technical implementation and deployment recipes.

Overview

The FIWARE Data Space Connector is an integrated suite of components every organization participating in a data space should deploy to “connect” to a data space. Following the DSBA recommendations, it allows to:

• Interface with Trust Services aligned with EBSI specifications
• Implement authentication based on W3C DID with VC/VP standards and SIOPv2 / OIDC4VP protocols
• Implement authorization based on attribute-based access control (ABAC) following an XACML P*P architecture using Open Digital Rights Language (ODRL) and the Open Policy Agent (OPA)
• Provide compatibility with ETSI NGSI-LD as data exchange API
• Supports the TMForum APIs for contract negotiation

Note: Although the FIWARE Data Space Connector provides compatibility with NGSI-LD as data exchange API, it could be also used for any other RESTful API by replacing or extending the PDP component of the connector.

Above listed functionalities can be used by an organization to connect to the data space in its role as data (processing) service provider, consumer of data (processing) services, or both.

Technically, the FIWARE Data Space Connector is a Helm Umbrella-Chart, containing all the sub-charts and their dependencies for deployment via Helm.
Thus, being provided as Helm chart, the FIWARE Data Space Connector can be deployed on Kubernetes environments.

Components

The following diagram shows a logical overview of the different components of the FIWARE Data Space Connector.

Precisely, the connector bundles the following components:

Note, that some of the components shown in the diagram above are not implemented yet.

Description of flows in a data space

This section provides a description of various flows and interactions in a data space involving the FIWARE Data Space Connector.

Onboarding of an organization in the data space

Before participating in a data space, an organization needs to be onboarded at the data space's Participant List Service by registering it as trusted participant. The user invoking the onboarding process needs to present a VC issued by the organization to the user itself, a VC containing the self description of the organization and a VC issued by a trusted Compliancy Service for the organization self description.

The following displays the different steps during the onboarding.

Steps

• The organization validates that the VC containing its description as organization is compliant with Gaia-X specifications using the services of a Gaia-X Digital Clearing House (GXDCH) - as a result, a VC is issued by the GXDCH (steps 1-2). That VC will end stored in the wallet of the LEAR either as part of the same process (once the GXDCH implements the OIDC4VCI) or via an issuer of VCs that exists inside the organization (step 3)
• The API for registering the organization is inspired in the DID-Registry API defined by EBSI but extending it to allow:
  • creation, update and deletion of entries beyond read(ing) of entries
  • authentication with VCs (including the VC issued by a GXDCH)
• Using an onboarding application (or a web portal) the organization’s LEAR requests the authentication into the Participant Lists service which ultimately translates into a request to the Verifier (step 4-6)
  • a page is accessed where a QR for authentication is displayed (step 4)
  • the QR code is scanned through the wallet (step 5) which translates
  • into a request to the verifier (step 6)
• The verifier checks in the PRP/PAP what VCs it has to request to the wallet (step 7). In principle it will find the following VCs to be requested:
  1. the LEAR VC accrediting the user as LEAR of the organization,
  2. the VC containing the description of the organization, and
  3. the VC issued by some GXDCH acredditing that the previous VC is Gaia-X compliant
• The verifier responds to the previous request sending a VP request to the wallet which responds with the requested VCs (steps 8-9)
• The verifier checks that the LEAR VC has been signed using proper eIDAS certificates and that the GXDCH VC has been issued by a trusted GXDCH (step 10). It finally produces an access token (steps 11-12) which the onboarding application can then use to invoke the EBSI DID-Registry+ API in order to register the organization as data space participant (step 13)

Consumer registration

Before being able to procure access to the provider's data service, a consumer organization needs to be registered at the provider's Trusted Issuers List as trusted issuer of VCs including claims representing a buyer of products in the provider's connector.

The following displays the different steps for the consumer registration.

Steps

• Using a contracting app (or a web portal), a Legal Entity Appointed Representative (LEAR) of the consumer organization will request authentication into the connector of the service provider (steps 1-3 involving scanning of QR code using the wallet)
• The Verifier will request from the user’s wallet a VC that acredits him/her as LEAR of the organization, eventually other VCs (steps 4-5). The wallet will check whether the verifier belongs to a participant in the data space (step 6) and return the requested VCs (step 7)
• The Verifier checks whether the LEAR’s VC was issued by a trusted participant of the data space (steps 8-9), and also checks whether other VCs required were issued by trusted issuers (step 10)
• If verifications were ok, it issues a token (step 11) that is transmitted to the user (step 12)
• Using the returned token, the user invokes TM Forum API to register the consumer organization at the Connector (steps 13-17) establishing the necessary access control (steps 13-14)
• Once the organization is registered and fulfills all the necessary information (which may take some time), the organization is registered in the local trusted issuers list as trusted issuer of VCs which include claims as buyer of products in the connector (step 18)

Contract management

After the registration, the consumer organization can perform contract negotiation, e.g., in order to procure access to a specific service linked to a product of the provider.

The following displays the different steps for the contract negotiation.

Steps

• A LEAR of the consumer organization will start authentication into the contract negotiation module of the connector of a service provider (steps 1-3 involving scanning of QR code using the wallet)
• The Verifier will request to the user (via his/her wallet) for VCs that acredit
  1. the user is a LEAR of the organization,
  2. (s)he owns credentials connected to roles meaningful for contract negotiation that the organization issued to the user and
  3. some other VCs (steps 4-5). The wallet will check that the verifier belongs to a participant in the data space (step 6) and return the requested VCs (step 7)
• The Verifier checks whether the LEAR’s VC was issued by a trusted participant of the data space (step 8), and rest of VCs required were issued by trusted issuers (step 9). Note that the VC for accessing contract negotiation functions requires that the organization were previously registered in the contract negotiation module, otherwise it will not be found in local trusted issuers registry
• If verifications were ok, it issues a token (step 10) that is transmitted to the user (step 11)
• Using the returned token, the user invokes TM Forum API to perform operations on the contract negotiation module (steps 12-17) establishing the necessary access control (steps 12-14)
• Once the organization is registered and fulfills all the necessary information (which may take some time), the organization is registered as trusted issuer of VCs which include claims as valid user of products accessible via the connector (step 18)

Service interaction

Once the procurement has been completed, a user or an application of the consumer organization can interact with the actual service offered by the provider, e.g., an NGSI-LD based data (processing) service.

In the case of a user interacting with the service, this is a Human-To-Machine (H2M) interaction.

In the other case of an application interacting with the service, this is a Machine-To-Machine (M2M) interaction.

The following displays the different steps for the two different types of interactions

Service interaction (H2M)

Steps

• A user of the product (employee or customer of the consumer oganization that was issued a VC in step 0 that acredits him/her as user playing a role relevant to the business logic of the product) request authentication in the connector (steps 1-3 involving scanning of QR code using the wallet)

• The Verifier will request to the user (via his/her wallet) for VCs that acredit

  1. the user owns credentials connected to roles meaningful for the given product/application and
  2. some other VCs (steps 4-5).

  The wallet will check that the verifier belongs to a participant in the data space (step 6) and return the requested VCs (step 7)

• Verifier verifies whether the VC was issued by an organization that
  1. is a trusted participant of the data space (step 8) and
  2. is a trusted issuer of the VCs meaningful for the application (that is, VCs only organizations that ordered the product can issue), also checks whether other VCs required were issued by trusted issuers (steps 9)
• If verifications were ok, it issues a token (step 10) that is transmitted to the user (step 11)
• Using the returned token, the user invokes services of the product (step 12)
• The PEP proxy and PDP will verify whether a user with the claims (attributes) included in the VCs extracted from the token is authorized to perform the given request (steps 13-15)
• If authorization is ok, the request is forwarded (step 16) and a response returned to the user (step 17)

Service interaction (M2M)

Steps

• An application from the consumer organization that acquired rights to use a product requests its authentication in the connector (steps 1)

• The Verifier will request to the application for VCs that acredit

  1. the application owns credentials connected to roles meaningful for the given product/application and
  2. some other VCs (steps 2-3).

  The wallet will check that the verifier belongs to a participant in the data space (step 4) and returns the requested VCs (step 5)

• Verifier verifies whether the VC was issued by an organization that is a trusted participant of the data space (step 6) and is a trusted issuer of the VCs meaningful for the application (that is, VCs that only organizations that ordered the product can issue), also checks whether other VCs required were issued by trusted issuers (steps 7)
• If verifications were ok, it issues a token that is transmitted to the application (steps 8)
• Using the returned token, the application invokes services of the product (step 9)
• The PEP proxy and PDP will verify whether the application with the claims (attributes) included in the VCs extracted from the token is authorized to perform the given request (steps 10-12)
• If authorization is ok, the request is forwarded (step 13) and a response returned to the app (step 14)

A detailed description of the steps to be performed by client applications and service providers can be found in the Service Interaction (M2M) documentation.

Deployment

Local Deployment

The FIWARE Data Space Connector provides a local deployment of a Minimal Viable Dataspace.

• Find a detailed documentation here: Local Deployment

This deployment allows to easily spin up such minimal data space on a local machine, by just using Maven and Docker (with k3s), and can be used to try-out the connector, to get familiar with the different components and flows within the data space or to perform tests with the different APIs provided.

Deployment with Helm

The Data-Space-Connector is a Helm Umbrella-Chart, containing all the sub-charts of the different components and their dependencies. Its sources can be found here.

The chart is available at the repository https://fiware.github.io/data-space-connector/. You can install it via:

    # add the repo
    helm repo add dsc https://fiware.github.io/data-space-connector/
    # install the chart
    helm install <DeploymentName> dsc/data-space-connector -n <Namespace> -f values.yaml

Note, that due to the app-of-apps structure of the connector and the different dependencies between the components, a deployment without providing any configuration values will not work. Make sure to provide a values.yaml file for the deployment, specifying all necessary parameters. This includes setting parameters of the connected data space (e.g., trust anchor endpoints), DNS information (providing Ingress or OpenShift Route parameters), structure and type of the required VCs, internal hostnames of the different connector components and providing the configuration of the DID and keys/certs.
Also have a look at the examples.

Configurations for all sub-charts (and sub-dependencies) can be managed through the top-level values.yaml of the chart. It contains the default values of each component and additional parameter shared between the components. The configuration of the applications can be changed under the key <APPLICATION_NAME>, please see the individual applications and there sub-charts for the available options.
Example: In order to change the image-tag of Keycloak, the values.yaml looks as following:

keycloak:
    # configuration for the keycloak-sub-chart. Its used as a dependency to the application, thus all config is accessible under the dependency name
    keycloak:
        image:
            tag: LATEST_GREATEST

The chart is published and released on each merge to master.

Testing

In order to test the helm-charts provided for the FIWARE Data Space Connector, an integration-test framework based on Cucumber and Junit5 is provided: it.

The tests can be executed via:

    mvn clean integration-test -Ptest

They will spin up the Local Data Space and run the test-scenarios against it.

Additional documentation and resources

Additional documentation

Additional and more detailed documentation about the FIWARE Data Space Connector, specific flows and its deployment and integration with other frameworks, can be found here:

• Additional documentation

Additional Resources

Following is a list with additional resources about the FIWARE Data Space Connector and Data Spaces in general:

• FIWARE Webinar about Data Spaces, its roles and components (by Stefan Wiedemann)