Re: [radext] Fwd: New Version Notification for draft-ietf-radext-radius-fragmentation-05.txt

[Alejandro Perez Mendez <[e-mail]>]

El 18/03/14 14:27, Alan DeKok escribió:
> Alejandro Perez Mendez wrote:
>> we could extend that line to something like:
>>
>> The
>>    authors acknowledge this is formally violating [RFC2865], but no
>>    operational issues are expected as no known implementation 
>>    perform that verification when proxying Access-Requests, since doing so would
>>    preclude them to support future extensions. 
>   Perhaps this (word-smithing)
>
> The authors acknowledge that this specification violates the "MUST"
> requirement of [RFC2865] Section 4.1.  We note that a proxy which
> enforces that requirement would be unable to support future RADIUS
> authentication extensions.  Extensions to the protocol would therefore
> be impossible to deploy.
>
> All known implementations have chosen the philosophy of "be liberal in
> what you accept".  That is, they accept traffic which violates the
> requirement of [RFC2865] Section 4.1.  We therefore expect to see no
> operational issues with this specification.  After we gain more
> operational experience with this specification, it can be re-issued as a
> standards track document, and update [RFC2865].

Thanks for the improved text. I will use it.

>
>   Alan DeKok.