Re: Issue 24 MUST/SHOULD in clock skew (section 7.1)

At Fri, 04 Feb 2005 13:35:17 +0900,
Kazunori Miyazawa <[email protected]> wrote:
> 
> #24 [*] MUST/SHOULD in clock skew (section 7.1)
> 
> 	The time difference MUST be computed and SHOULD be stored
> 	and used?  Why the different requirement levels?  (And is this sort of
> 	thing in the domain of KINK or Kerberos?)  (Ken Raeburn)
> 
> I think the clock skew processing is an implementation and/or operational issue
> in KINK. However the server must provide the ctime because the client may want
> to continue the process though there is skew.
> 
> I propose that KINK filling out ctime is MUST but KINK should not decide the
> processing.

I agree that the server MUST fill out ctime (and the current draft
also says so).

Regarding clients, do you mean KINK should not describe the
processing, or should not use capital letters?
I like SHOULD, or have no problem with describing small letters if the
WG prefers it.  But I think "not describing the processing" is not
appropriate because it is not specified by Kerberos (as far as I know)
and clients don't know how to process ctime without it.

-- 
KAMADA Ken'ichi <[email protected]>