Re: DRAFT Montreal minutes - X#NodeArchitecture comments

Should this one go into the implemeters guide then?

(I know I still owe the list an update addressing Ken's
other point about an implementation that does not log and
does not transmit the key.)


Julian Satran wrote:
> 
> Paul is right. Julo
> 
> 
> *Paul Koning <[email protected]>*
> 
> 08/08/06 00:24
> 
> 	
> To
> 	[email protected]
> cc
> 	[email protected], [email protected]
> Subject
> 	RE: [Ips] DRAFT Montreal minutes - X#NodeArchitecture comments
> 
> 
> 	
> 
> 
> 
> 
> 
>  >>>>> "Ken" == Ken Sandars <Sandars> writes:
> 
> Ken> Hi Paul, In [RFC3720] it is ALWAYS an error when the value of a
> Ken> declared key is "NotUnderstood". Hence this document needs to
> Ken> state the (slightly) altered processing of this key.
> 
> Actually, what we have here is a 3720 bug.  If a new key is
> declarative, an old implementation does not know that and will (or
> may?) treat it as a negotiated key, and reply with NotUnderstood.
> 
> That case is not specific to X#NodeArchitecture.  So it shouldn't be
> treated as a special case there; instead it should be noted as a
> bugfix to 3720.
> 
> The fix is easy: a "NotUnderstood" value for a declarative key is
> ignored.  
> 
>                   paul
> 
> 
> _______________________________________________
> Ips mailing list
> [email protected]
> https://www1.ietf.org/mailman/listinfo/ips
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> Ips mailing list
> [email protected]
> https://www1.ietf.org/mailman/listinfo/ips

_______________________________________________
Ips mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/ips