Last Call Review of draft-ietf-stir-rfc4916-update-06
review-ietf-stir-rfc4916-update-06-secdir-lc-smyslov-2024-11-12-00

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.

The draft defines a way for the called party to provide its authenticated
identity to the calling party in the SIP protocol. This ability helps to thwart
a number of attacks thus making this extension important for security of SIP.

The draft re-uses PASSporT construction from RFC 8225, that defines a format
for a cryptographically signed assertion of identity, but it uses this
contruction with a slightly different semantics (the 'dest' field is signed
instead of 'orig') and in a different SIP message (in the response, instead of
the request). Overall, it looks safe and the Security Considerations section
looks comprehensive, but I'm not familiar with SIP and STIR, thus I may have
missed some nuances.

Nits (may be ignored if I missed how SIP works).

1. It seems to me that adding PASSporT contruction into the SIP response may
lead to a higher risk of amplification attack (since the size of the response
is increased). If this can happen, then some discussion should be added to the
Security Considerations section.

2. The draft states that the 'rsp' PASSporT MUST NOT be sent in the request,
but provides no guidance what to do if this happens. I guess the called party
should send an error code (which?) and terminate connection, or what?

3. Perhaps adding few examples would be helpful (as in RFC 8946).