Re: [secdir] secdir review of draft-ietf-mpls-tp-itu-t-identifiers

["Adrian Farrel" <[e-mail]>]

Hi Dan,

The authors may care to comment further, but I think there are two points...

The regulatory codes are centrally allocated so the "screw-ups" that you refer
to are within those domains. That reduces the scope of the problem and also
reduces its impact. But maybe the document would benefit from a note on the
"risks of misconfiguration". I am not sure that that would belong in the
Security Considerations section, but space could certainly be found in the
document.

Yes, Corrigenda are stable ITU-T references. They are found on the same web page
that gives download pointers for the Recommendations they correct.

Cheers,
Adrian

> -----Original Message-----
> From: [e-mail] [mailto:[e-mail]] On Behalf Of Dan
> Harkins
> Sent: 03 January 2013 19:05
> To: [e-mail]; [e-mail]; draft-ietf-mpls-tp-itu-t-
> [e-mail]
> Subject: secdir review of draft-ietf-mpls-tp-itu-t-identifiers
> 
> 
>   Hello, and happy new year,
> 
>   I have reviewed this document as part of the security directorate's
> ongoing effort to review all IETF documents being processed by the
> IESG.  These comments were written primarily for the benefit of the
> security area directors.  Document editors and WG chairs should treat
> these comments just like any other last call comments.
> 
>   This draft creates a new globally unique identifier for the Transport
> Profile of MPLS. RFC 6370, which created identifiers for MPLS-TP,
> uses the operator's AS as a globally unique identifier but this draft
> proposes an alternative: use the ITU-T Carrier Codes. It then goes
> about changing the identifiers created by RFC 6370 by substituting
> the ITU-T Carrier Code for the AS.
> 
>   The security considerations state that the draft merely extends an
> information model and does not propose any protocol changes and
> therefore it does not introduce any new security concerns. This seems
> acceptable except that this extension relies on the global uniqueness
> of the ITU-T Carrier Codes (as RFC 6370 relies on the AS to be
> globally unique). Apparently "national regulatory authorities"
> ensure that they are unique in their regulatory domain (which is an
> ISO 3166-1 identified code) so as long as they don't screw up
> anything all is well. I think it might be worth mentioning the
> assumption that the "national regulatory authorities" will not make a
> mistake and what happens if they do. RFC 6370 relied on IANA to
> not make a mistake; this draft relies on all 249 entities that have an
> official code in ISO 3166-1 to not make a mistake.
> 
>   Also, there is a normative reference to a "Corrigendum" of an ITU-T
> recommendation on "OAM functions and mechanisms for Ethernet
> based networks". I have never encountered such a document. Is it
> a stable reference?
> 
>   regards,
> 
>   Dan.