Re: [dhcwg] DUID on a Virtual Host

On Feb 20, 2007, at 6:47 AM, Thomas Narten wrote:
>> This could be a one-time ID, and you could form a DUID-LLT-like
>> identifier from it, except that instead of being (link-layer, time)
>> it would be (link-layer, time, uid).
>
> To be clear, Do you think you can reuse the existing DUID-LLT type, or
> do you think you need to define a new one?

We would have to define a new one.   I don't think it makes sense to  
use the same format, because the documentation doesn't really match  
the proposed new use, and I would expect that to lead to needless  
confusion.

> In looking at the IANA considerations for DUIDs (see RFC3315), it says
> new DUIDS are created (only) through Standards Action. I wonder if
> that bar is too high. I'd think it would be sufficient to grant them
> under other types of RFCs. I.e., is it appropriate/necessary to
> _standardize_ all DUID types? I would imagine there will be special
> cases that fairly unique to specific scenarios which don't justify
> standardization.

I think relaxing it to "documented in an RFC" might be okay - someone  
might define a private use and publish an informational RFC.   I'm  
not sure how important this is, though - it seems like a fairly  
unlikely scenario.

> That said, in this case (virtualized hosts), I suspect there is
> utility/value in having a standardized approach, as it seems the
> scenario is somewhat general.

Agreed.

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