Re: [lamps] Paul Wouters' No Objection on draft-ietf-lamps-5g-nftypes-08: (with COMMENT)

[Russ Housley <[e-mail]>]

>>> ----------------------------------------------------------------------
>>> COMMENT:
>>> ----------------------------------------------------------------------
>>> 
>>> Thanks for the document.
>>> 
>>> I just have one question.
>>> 
>>> This extension MUST NOT be marked critical.
>>> 
>>> Why not? One can argue this is greenField deployment but it would be a rather big one :P
>>> 
>>>> From what I am reading, this extension is required for 5g, so why not mark it critical if
>>> the extension is not understood?
>> 
>> TLS is being used for authentication, integrity, and confidentiality within the 5G core, so I do not think we can ensure  that this is a greenField deployment.
> 
> I think I was unclear.  The document says these NFtypes are required for 5G. So they MUST be present. If that is the case, a system that doesn’t understand NFtypes fails. This type of extension understanding requirement is usually signalled with marking it critical, so I didn’t understand why this critical marking was a MUST NOT.

Paul, the document does not say that NFTypes are required for all certificates in 5G.  In fact the Release 17 documents from 3GPP do not require this extension.  It was not specified at the time those documents were written.  It is expected that future 5G documents will make use of this extension, but marking it critical would hinder transition from Release 17 to those future releases/stages.

Russ