Re: draft-ietf-dhc-dhcpv6-opt-timeconfig-02.txt

[Harald Tveit Alvestrand <harald@alvestrand.no>]

My agreement with Rob.

re CALSCH: the big advantage DHCP has is that if I understand it right, 
it's able to specify what the timezone rule is NOW, and leave updates of 
the rules as "someone else's problem".

But going on in that vein....
Query in general about DHCP: What does the spec say about the validity of 
the stuff in a DHCP info packet that is not the address? Can we 
(reasonably) expect the client to get the TZ info on some DHCP-server-set 
schedule, and notice that the rules have changed, or do we have to assume 
that the client will reboot at least once between the time of a rule 
definition being changed and the time the definition is in effect?

I'm all in favour of pointing to POSIX and Olson and saying "here's what 
these things say timezones are, and here is how you pass it in DHCP" - but 
no matter what, this whole thing is FAR better than the single-value, 
no-update-time value for timezone that is in a current DHCP option....

                Harald


--On onsdag, april 23, 2003 14:51:27 -0400 Rob Austein <sra@hactrn.net> 
wrote:

> It really depends on what information they're trying to convey.  Some
> of what this draft proposes to send around is just a label into an
> externally supplied rule database (upon which sender and receiver
> would have to agree), and that database isn't static, it changes over
> time as governments change their minds.
>
> I suspect that a bit of explanation about the purpose and intended use
> of the timezone option might help.