[Sidrops] Re: Call for Adoption of draft-xie-sidrops-moa-profile

[Linda Dunbar <[e-mail]>]

I think the problem described in the document is valid. It highlights a clear threat where a malicious actor could announce a false IPv6 mapping prefix for an IPv4 block, diverting traffic to unintended locations. This can lead to a loss of service, traffic interception, or other security vulnerabilities.

By introducing the MOA as a cryptographically signed object that links IPv4 prefixes with authorized IPv6 mapping prefixes, the document provides a robust method for ensuring the integrity and authenticity of these mappings. The use of MOA for Mapping Origin Validation helps prevent unauthorized mapping announcements, reducing the likelihood of hijacking.

I support the adoption of the document, with the following suggestions:


  *   The Abstract of the document does not really reflect the core of the document. It would be nice to move the following sentence from the Intro section to the beginning of the Abstract:
"This document proposes a new approach by leveraging Resource Public Key Infrastructure (RPKI) architecture to verify the authenticity of the mapping origin of an IPv4 address block."


  *   Expand on Security Considerations:
The Security Considerations section already outlines several important aspects of MOA security, but it could be improved by addressing potential failure modes:

     *   Misconfigurations: Address the potential impact of misconfigured MOAs or PKI failures. For instance, what happens if an operator incorrectly assigns an IPv6 mapping prefix? How should networks mitigate such issues?
     *   Compromise Scenarios: Discuss possible scenarios where the MOA PKI infrastructure itself could be compromised or where MOA signers act maliciously, and how to detect or mitigate such scenarios.

Best regards,
Linda Dunbar