Revocation of Persistently Non-functional Delegated RPKI CAs

Summary of proposal:

This proposal suggests providing a mandate to the RIPE NCC to revoke resource certificates associated with longtime non-functional CAs to reduce Relying Party workloads.

Abstract:

The RIPE NCC offers users of its RPKI certification service two deployment models: "Hosted CA setup" and "Delegated CA setup".

In the Hosted setup the RIPE NCC is responsible for timely issuance and publication of new RPKI Manifests and CRLs, but in the Delegated setup resource holders themselves manage their CA on their own infrastructure.

It is possible for Delegated CA infrastructure to be offline for extended periods of time, or for the contents of publication repositories to become stale or otherwise invalid. This proposal suggests providing a mandate to the RIPE NCC to revoke resource certificates associated with longtime non-functional CAs to reduce Relying Party workload.

This policy proposal targets only pathologically non-functional CAs. An example of a situation considered out-of-scope for this policy would be a publication repository service advertised to also be available via IPv6 and RRDP but in practice only reachable via IPv4 and Rsync: the associated CA would still be considered functional (provided a valid and current Manifest could somehow be retrieved and validated sometime in the previous three months). In other words: this policy proposal isn't about CAs that didn't achieve 100% uptime, but about CAs that are down all the time.

Policy Text:

If the RIPE NCC is unable to discover and validate a Delegated RPKI Certification Authority's (CA's) current Manifest and Certification Revocation List (CRL) for more than three months, that Delegated CA's resource certificate shall be revoked by the RIPE NCC.

The RIPE NCC shall make reasonable efforts to discover new Manifests, to notify the Delegated CA operator if a current Manifest and CRL cannot be validated, and to notify the operator if the delegation is revoked.

Rationale:

a. Arguments supporting the proposal

Persistently Non-functional Delegated CAs have subtle adverse effects within the RPKI ecosystem which may become more pronounced over time.

• Non-functional CAs offer nothing of value to validators (without a current and valid Manifest, signed payloads are unavailable).
• Validator synchronisation becomes more economic with fewer purposeless publication points to traverse.
• Non-functional CAs besmirch RPKI validator syslog message archives and waste CPU cycles and network traffic.
• Revocation is only a minor inconvenience for non-functional CAs (the configuration already was broken for an extensive period of time), but a big deal for any Relying Party (RP) - especially when taking into account the many synchronisation attempts made over long periods of time.
• This policy doesn't affect many resource holders; there only is a small group of persistently non-functional CAs, some of which have been non-functional for more than 2 years already. An overview is available here: https://console.rpki-client.org/nonfunc.html

b. Arguments opposing the proposal

• Resource holders might require more than 3 months to complete the initial Delegated CA setup.
  
  [Counterpoint to the above: initial setup procedures usually only takes a few minutes. Resource holders of course are free to simply retry the delegated CA setup procedure following revocation. The goal of this policy is NOT to permanently bar resource holders from setting up Delegated CAs, but to curb persistently non-functional delegations. Aka, "come back when you are ready".]