Page History

...

• Kevin: CVII call of last week was positive, Q. can we embed our CCS poc in the bigger CVII picture ?

• quick look at CCS Proof-Of-Concept - Work Breakdown Structure

• Gunnar: there are 2 areas where it is still worth completing the implementation

  • the graphql implementaion

  • the access management

• Iyyaz: concenring access management, are you looking for an implementation on the cloud side ?

• @kevin: is it for the neutral server to authenticate or for 3rd parties ?

• Kevin: this is for both

• Gunnar: it does not replace the authentication on the pipe to the vehicle, VISS protocol does have a definition of access control

• Gunnar: shows the vss-graphql repository

• Christian H: why don't we collaborate with kuksa project on this, since they have implemented it

  • https://github.com/eclipse/kuksa.val

• Gunnar: is it aligned with OAuth ?

• Christian H: this is to be checked with Sebastian Schidlt

• discussion on the various approaches of authentication

• Gunnar / Ulf: we use the same authentication mechanism but not the same authentication token on the cloud side and the vehicle side

• Iyyaz: what are the actions resulting from this discussion ?

• Gunnar: IMHO there is a need to review the kuksa code to check what can be reused

• Gunnar: there is a need to cross-check whether kuksa project supports OAuth with Sebastian until next call

• Gunnar: IMHO a deep knowledge of the OAuth2 specfication is necessary

• Gunnar: we might be willing to explore which implementation exist in the open source

• Iyyaz: I have been using the okta implementation of OAuth2, it is actually a service, but provides some code examples of how the service can be used

  • https://developer.okta.com/docs/guides/

  • okta supports jwt

  • okta claims you do not have to bother about the updates of the various software components used

• Christian H: the authenticator should be the code we are looking for

• Iyyaz: points to https://medium.com/decentralized-identity/the-universal-resolver-infrastructure-395281d2b540

• Christian H: points to  https://dev.uniresolver.io/

• TODO look into the kuksa code

  • Christian: on Bosch side the developer is "wenwenchen", we should open a ticket in github to get in touch with the developer

  • Christian: I will check with wenwen which material is available for explaining the authentication approach in kuksa

  • Jira
    server JIRA serverId 121ddff2-c571-320f-9e4d-d5b9371533bd key CCS-193

    created and assigned to Christian H

• TODO Iyyaz to explain the rationale for using an external authentication service (like okta)

  • Jira
    server JIRA serverId 121ddff2-c571-320f-9e4d-d5b9371533bd key CCS-194

    created and assigned to Iyyaz

• Philippe: I recommend that we use the mailing list for this because it gives more visibilty than comments in a Jira ticket

• Jira
  server JIRA serverId 121ddff2-c571-320f-9e4d-d5b9371533bd key CCS-127

  thrown into the current sprint

Vehicle data model

• Jira
  server JIRA serverId 121ddff2-c571-320f-9e4d-d5b9371533bd key CCS-141

  • Gunnar: presents the results of the review of autopilot project docs he made
  • Gunnar: data files containing values measured in test trials, however we do not have a lot to learn for the development of VSS
  • Philippe: a short comment, it might be worth checking whether those data files might used as inputs for testing in CCS poc

• Jira
  server JIRA serverId 121ddff2-c571-320f-9e4d-d5b9371533bd key CCS-53

  • Gunnar: IMHO we should Benjamin K to look further into it

...