List Info

Thread: Question: in RFC3046 why did Agent Subnet MaskSub-optiondie




Question: in RFC3046 why did Agent Subnet MaskSub-optiondie
country flaguser name
United States
2007-03-19 04:23:57
On Mar 19, 2007, at 9:43 AM, Bernie Volz ((volz)) wrote:
> Our assumption has always been that you need to
configure the DHCP
> server with this information (and DHCPv6 has no similar
options).

True enough.   To me this feels like an experimental RFC,
not  
necessarily something that's presently on its way to full
standard.    
If the author is expecting this to appear in relay agents
next week,  
he's deluded, but he's a reasonable guy and I don't get the 

impression that's what he's expecting.   Since you can't get
an  
option code allocated without writing an RFC, it is
necessary for him  
to write an RFC.   I'm assuming that's the motivation.

Since we placed that requirement on options allocation, we
are  
obliged to consider drafts like this.   I think the idea has
merit,  
and is worth experimentation.



_______________________________________________
dhcwg mailing list
dhcwgietf.org
https://
www1.ietf.org/mailman/listinfo/dhcwg

Re: Question: in RFC3046 why did Agent Subnet MaskSub-optiondie
country flaguser name
United Kingdom
2007-03-28 14:51:39
Ted Lemon wrote:
> On Mar 19, 2007, at 9:43 AM, Bernie Volz ((volz))
wrote:
> 
>>Our assumption has always been that you need to
configure the DHCP
>>server with this information (and DHCPv6 has no
similar options).
> 
> 
> True enough.   To me this feels like an experimental
RFC, not  
> necessarily something that's presently on its way to
full standard.    
> If the author is expecting this to appear in relay
agents next week,  
> he's deluded, but he's a reasonable guy and I don't get
the  
> impression that's what he's expecting. 

You're right, Ted. It's not something I need now, but it's
good to be 
ahead of the game, eventually, someone is going to want to
use a relay 
on a shared network with many subnets, and the ability to do
that (even 
if it requires an RFCxxxx-implementing relay) is a good
thing to have.

> Since you can't get an  
> option code allocated without writing an RFC, it is
necessary for him  
> to write an RFC.   I'm assuming that's the motivation.
>
Right again.

> Since we placed that requirement on options allocation,
we are  
> obliged to consider drafts like this.   I think the
idea has merit,  
> and is worth experimentation.
> 

Should  I post the guts-of-draft I wrote to cover this, do
you think?


Cheers,

Simon.



_______________________________________________
dhcwg mailing list
dhcwgietf.org
https://
www1.ietf.org/mailman/listinfo/dhcwg

[1-2]

about | contact  Other archives ( Real Estate discussion Medical topics )